Telecommunication Call Management and Monitoring System With Voiceprint Verification
Disclosed is a secure telephone call management system for authenticating users of a telephone system in an institutional facility. Authentication of the users is accomplished by using a personal identification number, preferably in conjunction with speaker independent voice recognition and speaker dependent voice identification. When a user first enters the system, the user speaks his or her name which is used as a sample voice print. During each subsequent use of the system, the user is required to speak his or her name. Voice identification software is used to verify that the provided speech matches the sample voice print. The secure system includes accounting software to limit access based on funds in a user's account or other related limitations. Management software implements widespread or local changes to the system and can modify or set any number of user account parameters.
Latest Globel Tel*Link Corp. Patents:
This application is a Continuation application of U.S. application Ser. No. 14/613,153, filed Feb. 3, 2015, which is a Continuation application of U.S. application Ser. No. 12/218,145, filed Jul. 11, 2008, now U.S. Pat. No. 8,948,350, issued Feb. 3, 2015, which is a Continuation application of U.S. application Ser. No. 10/893,575, filed Jul. 16, 2004, now U.S. Pat. No. 7,403,766, issued Jul. 22, 2008, which is a Continuation-in-Part application of U.S. application Ser. No. 10/215,367, filed Aug. 8, 2002, now U.S. Pat. No. 7,333,798, issued Feb. 19, 2008, all of which are incorporated herein in their entirety.
FIELD OF THE INVENTIONThe present invention relates generally to the field of telephone communication systems in penal institutions or similar facilities. In particular, the present invention relates to a computer-based telecommunication system with the capacity to allow an institution to authorize, control, record, monitor, and report usage and access to a telephone network using speaker independent voice recognition and speaker dependent voice identification to ensure that access is restricted to authorized users only.
BACKGROUND OF THE PRESENT INVENTIONGenerally, the need to monitor, control, record and provide detailed records of the usage of a telephone system in a controlled institutional environment is well recognized. It is common to utilize a controlled telephone system capable of monitoring outgoing telephone connections in many types of institutional environments, such as, but not limited to, penal institutions, military institutions, hospitals, schools, businesses, or specific types of government institutions. The reasons for monitoring and controlling institutional telephone systems are evident. To prevent such institutions from incurring unaccountable telephone costs, the institutions must either restrict access to outbound telephone lines or employ a telephone monitoring system to charge the responsible party for making the outbound communication. Otherwise, unaccountable telephone costs would severally hinder the availability of the telephone systems in institutions.
Therefore, it is imperative for many institutions to utilize a communication system that provides an accurate identification means for administrators to determine the individual responsible for each outbound telephone call. A communication system must also provide a monitoring means for maintaining a useful record of the communication. Additionally, the system may include a means for restricting access or a means for providing options to particular users. Considering the number of users in a large institution, different payment methods available, and the excessive call volume at many institutions, it is evident that an effective telephone management system is essential.
Providing telephone systems in specific types of highly restricted institutions, such as in penal institutions, results in the consideration of numerous additional complicating factors. Generally, outbound communication means in penal institutions are heavily regulated by the government. Therefore, communication systems implemented in penal institutions or similar facilities must meet greater security requirements often mandated by regulatory bodies affiliated with the county, state, or federal institution. Thus, the communication system used in a regulated institution must employ unique functions.
In its most general form, a penal institution's telephone system utilizes a call processor to approve and place a call, surveillance equipment or monitoring equipment, and a recording device for evidencing the conversation. Generally, these simple systems are not equipped to restrict an inmate from calling any individual or to verify the identification of a user utilizing advanced authentication techniques. However, it is preferable for the call system devices now employed in such institutions to have the capability to thwart an inmate from calling certain specific individuals or types of individuals. Without the necessary constraints on an inmate's use of the telephone system, inmates have often harassed outside parties or individuals. For example, it is generally preferred that an inmate should not be able to place a telephone call to the prosecutor who prosecuted the inmate's case or another attorney responsible for the sentencing of the inmate. In another example, it may be preferred that an inmate be prevented from contacting the victim of the inmate's crime or witnesses from the inmate's case. It has also been documented that inmates have used previous penal institution call systems to perpetrate additional criminal activities such as fraudulent schemes or specific criminal conspiracies. Specifically, inmates have been known to arrange credit card fraud attempts, the smuggling of contraband into the facility, and have even been known to arrange escape attempts over the penal institution's telephone system. Therefore, it is critical in an efficient penal institution to carefully monitor all outgoing telephone calls making a regulated penal institution telephone system a necessity.
Another concern in implementing an efficient institution telephone system is cost control. In order for a system to be cost effective the system must critically monitor and record the activities of each individual user to properly charge each individual caller for his or her outgoing calls. Typically, telephone communication systems in penal institutions provide an inmate with a telephone account upon arrival. Each individual receives an account number. There are several options for an inmate to select with respect to payment on the account. For example, an inmate may place prior personal earnings into the account. The cost of each call is then deducted from the total amount in the inmate's account until no balance remains. The inmate may choose to utilize collect call means. In addition, or alternatively, an inmate may be assigned a commissary account, where funds are added to the account based on work performed by the inmate. As the funds increase, the inmate may apply these funds to the cost of placing telephone calls.
The inmate debit account may be located onsite, at a central office facility, or at a third-party site. The inmate debit account may alternatively be controlled by the inmate's family. For example, the inmate's family may control the inmate's access to the debit account either remotely (e.g., by using the Internet, accessing a toll-free/pay to dial telephone number, using a mail form, etc.) or by visiting the prison facility. The inmate's family may add funds to the debit account and thereby control the call volume allowed to the inmate.
Another requirement of a secure telephone management system in a penal institution is the accurate identification of the telephone call participants. Generally, it is common in a penal institution to assign each inmate a personal identification number (PIN). When an inmate attempts to place a telephone call, the inmate must supply a valid PIN to gain access to the telephone system. However, a primary problem with this identification method is the ease of obtaining another inmate's PIN. For example, individuals who commonly forget their PIN may write it down, increasing the possibility that an unauthorized individual will view the PIN and use it. In addition, if a PIN number is compromised and utilized by an unauthorized inmate, the unauthorized inmate may then be able to call certain individuals who are restricted to that inmate, since the unauthorized inmate is no longer using the proper PIN associated with that inmate. In known systems, the PIN identification method is incapable of verifying that the individual who supplies the PIN is the actual specified inmate. Some systems have attempted to improve security by requiring the use of a debit card in conjunction with a PIN. The use of the debit card will only allow access to an inmate's account if the correct associated PIN is supplied. This method, however, provides only minimal additional protection because a debit card and its associated PIN can often, with or without force, easily be taken from another inmate, or given to another inmate, especially in the violent atmosphere of a penal institution. For example, one inmate may threaten another inmate in order to obtain such information. Alternatively, one inmate may provide certain services in exchange for the use of another inmate's telephone privileges. The possibility that two inmates will exchange accounts also exists, thereby allowing them to contact people that would normally be restricted to them.
Further attempts to obviate security concerns include requiring personal information, in addition to a PIN, to be supplied by the inmate/user. For example, a user might be prompted to supply a PIN as well as certain information that may only be known to the user. A common example is a request by the call system to provide their mother's maiden name. This provides an additional security measure, but again is minimally secure because such information can easily be obtained in a correctional facility. It would therefore be desirable to develop a telephone management system that incorporates an improved method of identification and/or verification.
Another required feature of a telephone management system for a penal institution or similar facility is a means for restricting calls placed by a user (e.g., an inmate). It is well documented that inmates often try to harass individuals related to their arrest or confinement, such as judges, prosecutors or witnesses, etc., through telephonic communications. Penal institutions have attempted to prevent this by restricting the telephone numbers each inmate is able to access. For example, a system may utilize a PIN or other identification means to access a list of telephone numbers that the inmate may not call, or alternatively, the system may access a list of numbers that the inmate is authorized to connect to (i.e., the inmate can only call the numbers appearing on the list). Telephone numbers placed on the restricted list can include any individual related to the conviction (e.g., the arresting police officer, the prosecuting attorney, etc.), while telephone numbers placed on the permitted list may be, for example, close family relatives. The system may also limit the amount of time each inmate/user is permitted to conduct each outbound telephone call through the system. Furthermore, restrictions may be regularly updated. For example, if an inmate misbehaves, the inmate's telephone privileges may be further limited or revoked completely.
Penal institutions are also concerned with monitoring the activities and communications of inmates. Monitoring telephone activities is necessary to restrict connections to illegal activities outside of the institution. Three existing types of call monitoring techniques are known in the art. The first technique is live monitoring. Live monitoring requires an operator or other individual to listen to each telephone call and alert the proper authorities if necessary.
The second type of monitoring involves recording the telephone conversation via a common recording device. A common example of this is a recording device such as a magnetic tape drive. This type of monitoring may be continuous or intermittent depending on the degree of security required for each inmate.
The third type of monitoring is known as passive monitoring. Passive monitoring may be activated when certain keywords are spoken. In addition, passive monitoring may be activated if the telephone call at the termination end is transferred to a third party via certain known detection means such as “click and pop” detection, etc.
Penal institutions currently record most inmate telephone calls, with the exception of lawyer-inmate communications which are generally prohibited by law. Typically in the art, monitoring may occur using any combination of the three methods (e.g., live monitoring, electronic recording monitoring, or passive monitoring). However, it would be desirable for a telephone management system to embody a means for determining which level of telephone monitoring should be employed for each telephone call. For example, it would be advantageous to flag certain individuals in an inmate's profile as highly suspicious. If the inmate initiates communication with the flagged individual, the system will alert a live operator to monitor the system. In such a system it is essential that the system correctly identify the called individual to avoid unnecessary expenditure of live operators.
Alternatively, the inmate telephone call system may utilize a remote alert notification system wherein the system contacts an operator when a violation has occurred. The system may contact the operator utilizing telephone means, paging means, etc. This notification system may be set to call the operator a limited number of times or until the alert has been noted in the inmate telephone call system. The operator may then access information about the alert remotely using the telephone, Internet, or any other such remote access means.
In order to alleviate some of the problems and concerns discussed herein, many penal institutions have implemented certain task-specific advanced systems. Generally, these “advanced” systems known in the art comprise several features.
For example, it is known in current systems to employ permanent call blocking. Specifically, it is known in the art to block an inmate or group of inmates from dialing certain telephone numbers. Most systems also prevent inmates from talking directly to live operators. This prevents inmates from requesting that the operator forward a call or provide additional telephone numbers allowing the inmates to harass or locate additional parties. Furthermore, current systems block “1-800,” “1-900” and other like telephone numbers including toll-free and pay-to-dial telephone numbers. In addition certain institutions may elect to block country codes, specific area codes, or other third-party numbers.
Current systems known in the art may also utilize a feature commonly referred to as “selective” call blocking. As discussed, “selective” call blocking may be employed to thwart inmates from establishing a connection with a selected group of individuals (i.e., with the home telephone of prison guards, wardens, indictment witnesses, trial witnesses, police officers, judges, etc.). It is also foreseeable that the telephone numbers of the family members of these specific individuals may also be blocked.
Some current systems also limit the use of specific long-distance carriers. This feature proves useful in limiting unnecessary costs incurred by employing alternating carriers.
Several current systems utilize features commonly referred to as “flash hook” prevention or “click” and “pop” prevention modes. These systems prevent inmates from extending the current outgoing telephone call and entering a new telephone call with a new number without fully terminating the original telephone call. For example, this prevents an inmate from utilizing common call forwarding features and the like.
In addition, some current institutional telephone systems electronically or manually disable the keypad after a telephone number is dialed and the telephone call is connected. This feature prevents inmates from interacting with telephone games and lotteries, and in certain older systems, prevents the inmate from achieving an unrestricted dial tone.
Another common feature employed by institutional systems is three-way call prevention. This feature prevents an inmate from instructing the called party to bridge the telephone call to another telephone number.
Other known systems in the art may exhibit other regulatory features. For example, generally, telephone communication systems allow an institution to limit the duration of a telephone call and/or to limit the cost of the telephone call. These types of features further allow a facility to customize the telephone call systems thereby preventing unrecoverable expenditures.
Another control used by current institution telephone systems is the use of certain aspects of biometric recognition for the identification of users or inmates (i.e., the calling party). However, systems known in the art have only used biometrics to a limited extent It is highly beneficial for communication systems in penal institutions to incorporate biometrics as an additional security device. Biometric recognition is commonly available in a number of fields. For example, biometrics recognition has found a number of security uses, including common usage, in credit card systems and building security systems. Biometric information includes fingerprints, rand geometry, voiceprints, retinal patterns, iris scans, signatures, infrared facial patterns, and all other sources which constitute unique physiological characteristics and which can assist in establishing a person's identity. Various devices exist which can scan one or more biometric characteristics and digitize the information.
The features discussed herein are present in several known systems. For example, one such system includes automatic account number validation and billing management. This system prompts a user for an account number and compares the number inputted to a number stored in a database to determine validity. If the account number is valid and found in the database, the system completes the predetermined telephonic connection. If the number is not in the database, and therefore invalid, the system will utilize voice prompts to request re-entry of the number or provide further instructions. The system attempts to locally automate and simplify the process of payment for routing calls without live operator assistance, but does not address additional security concerns that may exist in specific facilities, such as in a penal institution. Furthermore, this system does not provide for protection measures to confirm that the individual supplying the account number is the individual entitled to the use of the account (e.g., the system does not use any biometric identification to verify the identity of a user). In such a hostile environment as in a penal institution, an account number may easily be obtainable through coercion or by force. Additionally, the system does not provide any means of monitoring the conversations taking place or restricting which individuals are accessed by the user.
Another known call management system enables prepayment of telephone calls utilizing a debit system. Specifically, a user of the system obtains a special code by depositing a prepayment. The prepayment is stored in a database on the call management system for use in verifying calling party calls. To access the system, a user dials a special number and inputs a user-specific code for verification followed by the number of the party to be called. Next, the code is verified by the system. If verification is successful and sufficient funds are available, the call is connected. The prepayment amount, minus deductions for the running cost of the call, is tabulated as the call progresses. The call terminates either when the prepaid funds are exhausted in the user's account or when either party disconnects. This system also includes steps to prevent the same access code from being used at different terminals. However, it does not have a means for selecting the call type or a call monitoring means. The system also fails to perform an advanced verification means to identify and authorize a user.
Also known in the art is a multilingual prepaid telephone system capable of interfacing with a public switched telephone network (PSTN). In this system, each user is assigned a PIN and a credit account. To access the system, a user first dials a number to access the telephone system and chooses a language for all subsequent voice prompts. The user then supplies a PIN, which is compared against a list of numbers in a database. If sufficient credit is available for the duration of a telephone call to the destination number, the connection is completed and a timer is set for the available duration of the call. The call terminates either when the allowed amount of time for the call expires or if one party member hangs up the telephone line. If the latter situation occurs, the system computes a new available credit balance for the user's account. However, this system fails to provide a selection means for the user, such as the ability to choose the type of call to be placed (e.g., collect, debit, international, etc.). Further, it does not have any call monitoring means and would therefore be unacceptable as a communication system for a penal institution.
Yet another known system in the art is an integrated commissary system for receiving and processing orders in an institutional setting. The commissary system is designed for use without access to a PSTN. User status and inventory status are stored in an onsite database. To access the database, a user provides identifier information and item selections through selected telephones. The selections are compared against the onsite database using a processor. If the user is authenticated and the requested items are available, the processor generates transaction records, updates user commissary information, and correctly adjusts inventory. The updated information is stored in a file that may be used for record keeping or archival purposes. However, as described this system does not teach a commissary system for use with a PSTN. It also fails to teach multiple authentication means and would therefore be unacceptable for use in a penal institution.
Another known system includes a software process for real-time call rating and debiting so that a subscriber's account balance is not exceeded. The method used in this system estimates the time when the user's balance will expire by using the total charge per second average. The process then determines the time remaining by dividing the account balance by the average charge per second of all telephone calls, and the time limit for the call is then set accordingly. This method is useful if the rate for long distance calls is not known locally. However, the system does not allow for other types of calls, such as collect calls, to take place. Furthermore, the system does not provide an advanced call monitoring apparatus with an advanced authentication apparatus (e.g., no voice identification for verification is used in this system).
Still yet another known system provides automated public telephone control for charge or collect call billing. The apparatus embodies a microprocessor system controlling voice prompting, recognition of responses, network signaling, recording of calling details, and verification of account numbers. The system provides for an automated telephone billing for public telephone systems. The system offers a plurality of billing methods, such as billing to a credit account number, to the called party (collect calling), or to a third party. An additional aspect of this known system is the recognition of voice utterances from other signals and called party spoken words (i.e., the system can recognize the word “yes” when spoken by any individual). However, the system does not identify or verify the individual speaking. Furthermore, this system does not provide a means to identify the user or verify that the user is not partaking in fraudulent activities. Finally, this system fails to teach a monitoring and call control means.
Still another known system in the art includes a means for detecting three-way call attempts by monitoring for pulses of energy indicative of a hook-flash. The system includes a low pass filter for passing energy signals having frequencies below 500 Hertz (“Hz”), preferably in the range of 100 to 300 Hz, and an energy detector for detecting specific electrical energy pulses passing through the filter and having a predetermined minimum magnitude. The system also includes a software window analyzer, which cooperates with the energy detector to detect specific events, such as sound, occurring on the telephone line during a predetermined time window after the detection of the aforementioned energy pulse. The software window analyzer includes a timer means that is activated by the detection of the energy pulse, and a sound means for detecting the occurrence of sound on the telephone line during at least one of multiple windows of time defined by the timer means. The non-occurrence of sound on the telephone line during a specified time window is used by the system to confirm that the detected energy pulse is in fact a three-way call attempt. A counter means is further implemented for counting specific energy pulses detected by the energy detector during the time window when the remote party is using a pulse-dial telephone.
Also known is a collect call system which automatically routes long distance calls without intervention of an outside service or operator. This feature enables private public telephone owners, as opposed to primary telephone companies, to receive revenue for completion of the call. The invention comprises the steps of providing the calling party with voice prompts, receiving voice or dialed signal information about the calling party in response to the voice prompts, either voice or dialed signals, locally recording the information about the calling party, providing the called party information about the calling party, and reacting to a variety of provided signals by either the called or calling party. The known system does not provide for other possible payment methods. The system is further limited by its lack of telephone call monitoring ability and calling party identification means, and is therefore unsuitable for use in penal institutions.
There is also a known system in the art that utilizes biometric verification means for determining if a user is authorized to use a credit or identification card. This system introduces a method embodying the steps of receiving biometric data from a cardholder, scanning the card to obtain user information, and retrieving authorized card owner information stored in a database. The information obtained from the three sources (i.e., cardholder, card, and database) is analyzed and compared to verify that the cardholder is the authorized card owner. A number of possible biometric features may be used such as voiceprints, fingerprints, digital photography, and retinal scans. The system compares the user information to an already existing database, which lessens the possibility of incorrect identification. Further, the system provides a reliable means for verifying a user in a credit or debit card system. However, the system fails to implement additional biometric means useful in identifying a called party and fails to apply these features to specific institutional functions.
Systems are known in the art for controlling, monitoring, recording and reporting telephone communications. The system deals primarily with the identification of a user through use of a PIN and restricting telephone communications through a profile accessed by the PIN. The system further contemplates means for monitoring and recording communications. The system may also incorporate an improved method of monitoring calls. The method includes a means for detecting tones commonly associated with call bridging and call forwarding attempts. For example, it can detect tones such as ring signals, busy signals, special information tones (SIT tones), dual tone multi-frequency tones (DTMF), call progress tones or other similar tones characteristic of the placement of a telephone call. However, this system is limited by detection of certain sounds which may not be readily machine-recognizable. For example, it is foreseeable that interference, background noise, or compressed voice data may inhibit the detection of the tones.
Another known system is used for the verification of a calling party, called party and a secure connection by using secure telephone devices, such as the Micro MMT Sectel® 1500BDI. Specifically, in this system, the calling and called parties supply voice data which is encoded and transmitted over a telephone network. Both users hear the alternate party's recorded voice data and verify that the supplied voice data is correct. The call is established only if both parties verify that the called party has provided the correct voice data. However, as is known in the art, the secure telephone devices are often costly and would not provide adequate security in a penal institution or similar facility. Additionally, the system does not allow for different possible payment methods for calls or call management. Another system known in the art uses voice recognition as an identification means, including speaker recognition by using two processing units. The first unit receives voice characteristic information by recording specific words spoken by the user. The information is analyzed and stored in a database. The system prompts the user for additional information, which is then received by a second processing unit. The results of the analysis of the second processing unit are sent to a first processing unit with the previously stored information. A comparison of the analyses determines if the user is authorized. This system is limited by its inability to manage call restrictions and lacks monitoring capabilities.
Another system known in the art permits users repetitive access to a multitude of systems. The system requires an initial enrollment phase for access. The enrollment phase consists of extracting biometric data to be stored for future use. The format of the data is compatible with a plurality of verification identification systems including voice recognition, fingerprint identification, and retinal scan identification. However, there are no restrictions to the system or further monitoring means during use of the system, which are essential for systems within a penal institution.
Finally, a system is known that uses acoustic and non-acoustic attributes to identify. An initial profile is created by both utterances spoken by a user and non-acoustic information, such as keying in a user's customer number or social security number. The acoustic recognition contemplates the usage of a plurality of voice recognition methods. The system is limited in its ability due to the lack of monitoring or call management abilities.
In view of the foregoing, clearly there exists a need for an improved method and apparatus for managing an institution's telephone call system. There also exists a need for an improved method for controlling access to an institution's telephone call system by use of advanced authentication and verification techniques including integration of voice or other biometric identification means. Furthermore, a need exists for an improved telephone call monitoring system for a penal institutions or similar facilities that addresses the increased concerns of monitoring the call activity of the calling party. In particular, a need exists for a computer-based telecommunication system with the capacity to allow an institution to control, record, monitor, and report usage and access to a telephone network.
SUMMARY OF INVENTIONThe present invention embodies an improved telephone call management system using improved identification means including biometric identification. For example, the identification means may include the utilization of speaker independent voice recognition in conjunction with speaker dependent voice identification. This, in combination with a user-specific personal identification number (“PIN”) provides an improved method of identifying and restricting a user's access and utilization of an institution based telecommunications network. In addition, the present invention incorporates control means, monitoring means, recording means, and a reporting means for an institution based telecommunication network. The present invention may implement a debit card platform or other such payment methods. The system of the present invention may be implemented in a variety of facilities including, but not limited to, penal institutions, mental institutions, nursing homes, rehabilitation centers, correctional facilities, government agencies; private and public business, and the like.
One objective of the present invention is to provide a telephone call system including a means for identifying and authenticating an institutional calling party. For example, it is foreseeable that these means may include, but not be limited to, the use of an institution-assigned inmate specific debit card requiring specific authentication data, the use of biometric recognition devices, the use of radio frequency identification devices, etc.
As one specific example, the system may require that a user provide a valid PIN and corresponding voice identification to access the system. The system can verify the PIN and match the provided voice identification with a stored record. Further, the invention ensures that the voice identification includes certain words (e.g., the caller's name). This prevents an inmate from using an easily mimicked voice sample (e.g., the system would reject a voice sample that consists of a user simply whistling, coughing, or remaining silent).
The system of the current invention is designed to operate in either a local area network (LAN) or a wide area network (WAN). In a LAN configuration, the telephone system is controlled by a configurable switchboard device that routes calls, provides voice prompts, and responds to menu selections. Calls placed by users of the system are routed through the switchboard device and connected to the proper outgoing trunk based on the type of call placed (e.g., collect, debit, etc.). The switchboard tests outgoing trunks as calls are placed. If no dial tone is detected, the trunk is taken out of service for a pre-programmed period of time. An integrated cross point switch enables any telephone to access any available outgoing trunk. The cross point switch also allows calls from other switchboard modules to be routed to alternate modules for completion. During periods when all trunks are in use on the module at the telephone connection, calls can still be completed provided that some trunks are available on alternate modules. The switchboard device also has an integrated channel bank, allowing for fully integrated T-1 capability. This allows calls to be processed either over analog or digital trunks as required. The architecture of the switchboard allows it to accommodate multiple processors, eliminating system overload during extremely busy periods of telephonic communications.
The switchboard is connected to a site server, which is commonly referred to as an ITAC (Inmate Telephone Access Control) in penal institutions or a UTAC (User Telephone Access Control) in certain other types of institutions. This device serves as the main database for the telephone management system. It has the ability to log and record details of all telephone calls placed through the system and store them for a period of time defined by the institution. The ITAC/UTAC also digitizes all information for the digital T-1 trunk. A recorder, which may be integral to the system or remote to the system as a separate storage means attached to the ITAC/UTAC, is responsible for recording the telephone calls and storing them in one or more databases depending on the size of the institution or the amount of data which must be archived by the institution and the capability of the storage means.
In the system of the present invention, a new user receives a PIN, account number, cell location and other identification information. This identification information is stored in a database preferably located at the ITAC/UTAC. Further, a new user is required to speak his or her name when subscribing. This sample speech, also stored in the database, is utilized as voice identification data every time a user places a call. Specifically, the system prompts the user to speak his or her name and the user's response is compared with the original speech sample using speaker dependent voice identification. Importantly, the system includes methods for ensuring that the original speech sample is the user's name and is not an easily mimicked speech sample, and that the speech captured when placing a phone call matches this original sample.
Connected to the ITAC/UTAC are a number of administrative and investigative workstations used to create, edit, and monitor user accounts and telephone calls. The investigative workstations may be used to listen to the outgoing telephone calls in real time or to access calls stored on the server or other type of database or storage means.
In a WAN configuration, the site server is connected to multiple switchboard devices that are located in separate institutions. In this embodiment, the ITAC/UTAC serves as the database location for the entire system. Administrative and investigative workstations may be located at every facility. Alternatively, it is foreseeable that one or more sets of workstations at a central facility may be used to administrate all user accounts.
User-friendly software utilizing a graphical user interface (OUI) or other types of on screen display (OSD) capable devices may be employed to administer all user accounts of the telephone management system. The software allows a system administrator to provide calling restrictions at all levels of operation. Such restrictions may include, but are not limited to, the total number of minutes allowed, the total number of calls placed, dates and times calls are allowed, telephone exchanges allowed to be accessed, the number of times the debit inquiry system may be used, and other like restrictions. If a WAN configuration is being used, such restrictions can be set for each institution on the network or for the entire telephone network. In addition, it is contemplated by the present invention that different divisions of each institution, such as cellblocks in a correctional facility, may also be given global restrictions. Since each division of the institution is linked by a LAN/WAN, changes can be made at any of the different institutions and then be applied globally or locally.
Additional restrictions and options are available for individual user accounts. For example, each user may be assigned a language for the telephone system's voice prompts. Another option is the ability to limit the telephone terminals a user may call from or the payment method utilized for each call made.
The called party that a user may contact may also be restricted. For example, certain exchanges or telephone numbers may be blocked. Alternatively, users may have to submit a pre-approved list of numbers to the system administrator. Once the user list has been checked to ensure that the inmate is allowed to call those people, the inmate may only call the people on this list.
Certain options for contacts on the pre-approved list may be set for each person on the list. For example, a language for each person may be set. The voice prompts for that contact will then be in that language. Contacts may also be assigned specific recording restrictions. For example, a conversation between an inmate and an attorney may require specific recording restrictions. Other options such as if a caller can only place calls collect or by debit, may also be specified.
An additional feature of the software is that it can create a debit account for each user and monitor the account balance. The cost of each call is subtracted from the account after the call is completed. Also, an account administrator can manually add or subtract funds to individual accounts. The inmate's access to the account may alternatively be controlled by the inmate's family. In this configuration, the inmate's family controls the inmate's funds and thereby controls the inmate's access to the account.
Since the site server logs data about each call, the present invention is designed to provide reports, either in electric or hard copy form, utilizing specific data from the database. Therefore, a system administrator can track important statistics such as the net profit of the telephone management system during a certain time period. It is foreseeable that the software may incorporate extensive data processing services for providing a user a multitude of correlated dates.
In one embodiment of the present invention, when a user attempts to access his or her account to place a call, the user may be requested, via voice prompts, to select a voice prompt language and enter a user-specific personal identification number. The information entered by the user is compared with information stored in the database for that specific user. If a corresponding PIN match is not achieved, a verification error may be provided and the system may request a re-entry of the PIN. It is foreseeable that if the second attempt fails to provide a match, the individual may be denied access to the telephone system and an official may be notified.
The system can also use speaker dependent voice recognition as an added level of security. For example, when a user is registered (i.e., given a PIN), the system can also store a sample of the user's speech and associate that sample with a PIN. Then each time an inmate attempts a call by entering the PIN, the system prompts the inmate to repeat the previously provided sample. The system utilizes speaker dependent voice identification to ensure the inmate placing the call has the same voice as the stored sample.
The system can also use speaker independent voice recognition to ensure the sample of speech is a certain key word (e.g., the inmate's name). This prevents an inmate from storing a sample (e.g., a whistle, cough, etc.) that is easy to mimic.
Once the inmate's identity is verified, the system may announce the inmate's call restrictions. For example, the system may access a pre-recorded menu of restrictions informing the inmate of specific restrictions, such as total telephone usage time, individual call length, and different payment options, such as collect call means or debit account means. The system may then connect to the desired called telephone number, provided that the number is on the accessible number list for that user. The system may first prompt the person called to select a language for future voice prompts. The language selected may then be saved in the database. Then, the called party may be informed, via voice prompts or other like means, the identity of the calling party and location from where the calling party is located (e.g., John Doe from Peters State Penitentiary). The called party can accept or reject the caller through voice response, DTMF tones, or other like input means. If the called party provides a negative response (i.e., rejecting the call), the called party may select an option blocking calls from this caller, institution or similar calls in the future. The data may then be stored in a database. After accepting the call, the called party may be prompted to supply identification information (e.g., social security number, user name, date of birth, etc.). Additionally, the called party may be asked to provide a PIN to be used for future authentication. The information may be used to later identify the called party and verify permission for communication with the calling party. Alternatively, if the called party is on a pre-approved list, the call may proceed without any voice prompts in a similar manner as a normal PSTN telephone call.
The system may also utilize other authentication means to provide access to the telephone management system. For example, biometric data may be required to access the system. Biometric data includes, but is not limited to, voiceprints, face architecture, signature architecture, fingerprints, retinal prints, hand geometry, and the infrared pattern of the face. Such data may be acquired from users either from prior supplication to biometric systems or from the acquisition of the data from the user upon the creation of a telephone account for use with the system. This data may be stored along with the user's PIN in the user's account profile or some other storage means to be used later as an authentication device.
When a user attempts to access the telephone system at a later time, the user may hear a series of voice prompts directing the user to first supply a PIN and then supply the same form of biometric information that is stored in the database. For example, if the user's thumbprint was stored digitally in the database, the user would have to supply a thumbprint to a device capable of scanning it and converting the resulting data to the same format as the information in the database. The scanned data would then be compared to the information maintained in the storage database. If a positive match occurs based on the PIN and biometric data entered, then the user would be granted access to the system subject to user specific restrictions.
Biometric authentication means may also be implemented to authenticate the called party. In this type of system, the authentication may include one or more biometric authentication means in addition to non-biometric authentication means. In this embodiment, before the called party is allowed to converse with the caller, the called party may be asked to supply voice authentication and/or provide a PIN. This information may be stored in a database either prior to a user's first call or when the first call is made. Similar to the method of speaker independent voice recognition and speaker dependent voice identification (utilized to verify inmates), the system may require the called party to speak a specific word or phrase (e.g., the caller's name). Speaker independent voice recognition is used to verify what the called party spoke. Again, this additional step ensures that the called party provides a significant voice sample that is not easily mimicked. If the data has been stored prior to the call, the called party would have to state the pre-recorded phrase. The recorded data would then be compared with information in the database. If the data is in compliance with the information in the database to within some pre-assigned statistical threshold, the system would allow the call to proceed and a connection would be made. If the data had not been stored prior to the call, it would be retrieved from the called party and used for future voice authentication.
A further authentication means that may be utilized by the present invention is the use of radio frequency (“RF”) authentication. The user can be required to have in the user's possession some type of radio frequency identification technology to access the telephone system. This may be accomplished in a number of ways.
In a first example, each user of the system would be required to wear an RF band attached preferably to the ankle or wrist. The RF band may be active, passive, or neutral. For example, the frequency of the band is unique to each wearer. An active RF band contains a transponder that either intermittently or constantly emits an intermittent RF pulse that is detected by a series of sensors placed about the user's location. The sensors relay the detected data to a remote or central database containing a processor that calculates the location of the wearer in one of two usual methods known in the art, as well as those methods not yet contemplated.
The first method involves triangulating the user's source, which requires two or more sensors or a sensor on a rotating platform. The analyzing system detects the angle of the incident pulse with the sensor and uses this data from multiple detectors to triangulate the source of the user.
A second method of detection involves the analysis of the time of flight of the emitted RF pulse. This method requires the same sensor configuration as the triangulation method. In this method, the sensor measures the time it takes each pulse to arrive at the sensor location as compared to the pulses that occur at a regular rate in an active RF device. From this data, the analyzing system can determine an accurate distance from the sensor that the user must be located within some minor degree of error. By combining this information from a multitude of sensors, the location of the person can be triangulated.
Active RF bands may be used in the present invention to locate an individual within the facility housing the telephone system. When an individual attempts to access a telephone terminal, the system can determine the user accessing the terminal since the location of each individual is known. The options for that specific individual can then be used by the phone terminal allowing the user to place a call.
It is advantageous to use active RF bands because they are very accurate for determining the location of an individual. However, most active systems have a relatively short battery life and require constant recharging. A passive RF device may also be used with only slight modification to the system. A passive RF device works by emitting a pulse only when activated by some other emitted pulse. In this embodiment, the sensors may also be equipped with RF pulsing devices designed to activate the user's RF devices. When a user attempts to access a telephone terminal, the system sends out a pulse. As a result, the user's RF device is activated and the person using the telephone terminal is authenticated. The user's specific user options can then be forwarded to the telephone terminal the user is utilizing.
Passive RF bands may be used in conjunction with the present invention for a variety of applications. For example, each telephone may be located in its own booth and fitted with an RF emitter and sensor. When a person attempts to access a telephone account at a terminal, an emitter terminal can send out a pulse activating a passive RF user device. In turn, the RF device may emit a responsive pulse. Subsequently, the sensor on the terminal can detect the responsive pulse. Using time of arrival analysis, the user can be located and verified, thereby authenticating the telephone terminal.
Neutral RF bands may also be used in the present invention. Neutral RF bands function by reflecting an incident RF pulse with a slight modulation, which is unique to each user. In response, the sensor on the telephone terminal can record the reflected wave and forward the information to analyzing software. The analyzing software subtracts off the original pulse signal to determine the modulation frequency of the user's RF band, thereby authenticating the user at the terminal. The correct user options can then be supplied to that specific terminal.
In another example of the present invention, a debit card may also be used in conjunction with a PIN. At each terminal, the user may be instructed to scan a user-specific debit card into a debit card reader and enter a corresponding PIN in order to gain access to the system. Alternatively, a user may enter information related to a debit card in any of a number of known procedures followed by a PIN. This method of authentication may also be combined with biometric and/or RF identification means.
After a user is authenticated and a calling party is contacted, the present invention provides monitoring and safety means that are active when a call is in progress. Call recording is the primary security means. There are three types of call recording utilized by the present invention. However, numerous other call recording systems in accordance with the concepts and objectives of the present invention may be incorporated.
In a first call recording protocol, by default, all calls are logged and recorded unless otherwise specified in a user's contact file (e.g., calls to an attorney are not recorded). It is foreseeable that calls may be archived on a site server or other database storage means for future reference.
A second type of call recording implemented in the present invention is referred to as passive recording. This type of recording may utilize voice recognition software to listen for certain key words or phrases in a conversation. If a key word is uttered, the system records the rest of the call and alerts the system administrator, proper authorities, or implements other security procedures. For example, in a penitentiary, if the word “drugs” is used in a conversation, the system may be activated thereby recording the remainder of the conversation.
The third type of call recording that may be utilized in the present invention involves an operator scan of a user conversation. In response to a flagged conversation, the operator may have the discretion to selectively record the conversation. For example, using the same software used to administrate accounts, an operator can selectively listen to conversations and record suspicious conversations for future reference.
Third-party call detection is another security feature that the present invention may utilize. The system software may have the ability to detect whether a third party is connected upon the origination of the call. If a third line is detected, but not authorized, the software will end the communication and notify the authorities that such an event has occurred. The telephone management system may also include a feature that terminates the telephone call if more then a certain number of individuals are included in the conversation. It is foreseeable that this may be accomplished by voice recognition means.
Another type of third-party call detection involves the detection of hook-flash events. A hook-flash event occurs when a user hangs up the telephone for only a brief instant during a telephone call, just long enough to not disconnect the line. This is usually characteristic of a switch between lines on call waiting or to initiate a three-way conference call. If the telephone system's software detects a hook-flash event by either party to the call, it will terminate the call and alert the authorities. This feature of the system prevents unauthorized third parties from taking part in telephone conversations.
An additional security means incorporated in the present invention is the locking of the telephone keypad after a connection is made. This prevents the caller from dialing out to a third party that has not been approved after a pre-approved connection has been made. Alternatively, the system may allow the user to only press a predetermined number of keys after a connection has been made. This allows an authorized user to access an automated service and enter menu selection keys.
The human voice has the capability of creating DTMF tones also. In order to differentiate tones created by a user's voice from tones created by the telephone keypad, the present invention incorporates software which monitors the frequency of DTMF tones. Such software is capable of determining the source of the DTMF tones.
One objective of the present invention is to provide an efficient and reliable user identification system using biometric identification means wherein speaker independent voice recognition and speaker dependent voice identification are used to verify the identity of a user each time a user attempts to place a telephone call.
Another objective of the present invention is to capture a significant sample of a user's speech when the user subscribes to the system and to request the user to repeat this sample every time the user places a telephone call to provide an improved voiceprint identification system for use in a telecommunications call management and monitoring system.
An additional objective of the present invention is to provide an improved telephone communication monitoring system. Another object of the present invention is to provide an improved call management system using a user-friendly operating system.
An additional object of the present invention is to provide an improved call management system capable of operating in a LAN or WAN.
A further objective of the present invention is to permit alterations of call restrictions and options through a computer control unit.
Still another objective of the present invention is to provide a means of restricting communication options to certain users.
An additional objective of the present invention is to incorporate biometric verification, including voiceprints, face architecture, signature architecture, fingerprints, retinal prints, hand geometry, infrared pattern of the face, etc., with a debit card system. Yet another object of the present invention is to utilize a combination of a PIN and a voiceprint to accurately verify a caller.
Still another object of the present invention is to restrict voiceprints to certain keywords or phrases to ensure that the voiceprints can be used accurately for subsequent identification of a caller.
Yet another objective of the present invention is providing called parties with the option of rejecting the call and blocking similar calls in the future.
Another object of the present invention is to incorporate an active RF device for use in authentication in a call management system.
A further object of the present invention is to incorporate a passive RF device for use in authentication in a call management system.
Another object of the present invention is to incorporate a neutral RF device for use in authentication in a call management system.
Still another object of the present invention is to incorporate advanced software incorporating at least a monitoring module and an accounting module.
A further understanding of the present invention can be obtained by reference to the preferred embodiment and alternate embodiments set forth in the illustrations of the accompanying drawings. Although the illustrated embodiments are merely exemplary of systems for carrying out the present invention, both the organization and method of operation of the invention, in general, together with further objectives and advantages thereof, may be more easily understood by reference to the drawings and the following description. The drawings are not intended to limit the scope of this invention, which is set forth with particularity in the claims as appended or as subsequently amended, but merely to clarify and exemplify the invention. For a more complete understanding of the present invention, reference is now made to the following drawings in which:
As required, a detailed illustrative embodiment of the present invention is disclosed herein. However, techniques, systems and operating structures in accordance with the present invention may be embodied in a wide variety of forms and modes, some of which may be quite different from those in the disclosed embodiment. Consequently, the specific structural and functional details disclosed herein are merely representative, yet in that regard, they are deemed to afford the best embodiment for purposes of disclosure and to provide a basis for the claims herein, which define the scope of the present invention. The following presents a detailed description of a preferred embodiment as well as alternate embodiments such as a simpler embodiment or more complex embodiments for alternate devices of the present invention.
Referring first to
Multiple processors may also be incorporated into the architecture. This allows call processing even after parallel component failure. The architecture also provides for a sharing of the load between processors, which eliminates system overload during extremely busy periods. The multiple processors enable the system to handle large volumes of calls at any time, and to ensure system integration.
Additionally, electronic switchboard device 105 performs the voice prompts heard by the inmate and the recipient of the call allowing the parties to respond to the menu selections. For example, before each telephone call is placed, preferably, the user of the system is required to provide a personal identification number (PIN) and a speech sample used to verify the identity of the user. Electronic switchboard device 105 prompts the user to provide these identifications and receives the responses (DTMF, and spoken) from the user. Electronic switchboard device 105 tests outgoing trunk lines as calls are placed and digitizes telephone audio for recording and/or biometric voice identification purposes. If no dial tone is present, one of trunk lines 111 may be taken out of service for a pre-programmed amount of time for maintenance. These capabilities are pre-programmed into the device's firmware. However, it is foreseeable that software and software upgrades may provide these services in addition to other services useful in the present invention.
A central site server 113 interfaces within the telephone call system 101 via a first serial port 115. In the preferred embodiment of the present invention, an RS-232 serial port is employed for the interference connection. However, it is foreseeable that other types of serial ports 115 commonly known in the art may be utilized. Serial port 115 may also be comprised of a direct hardware connection or may consist of a series of ports and connecting means commonly known in the art for connecting electronic devices. Serial port 115 is designed to allow firmware driven systems, such as electronic switchboard device 105, to interface with software-based systems, such as a PC designed system operating as a site server. All inmate and call information is routed through central site server 113. At central site server 113, user call information is digitized for efficient data transfer and efficient record keeping. Central site server 113 stores at least each user's financial transaction data. It is preferred that central site server 113 also stores the digitized audio used for voice prompts as well as each user's call restrictions, PIN, biometric verification data, etc. However, depending on the memory requirements, numerous site servers may be employed. It is foreseeable that older archived data may also be stored on an integral or a remote computer system database (not shown) or kept on additional storage devices on the central site server 113. Preferably, the voiced data, PIN, etc., are stored in a database on the central server 113. The database may be any commercially available database such as Microsoft Access, Microsoft SQL server, an Oracle database, an IBM database, etc.
Connected to central site server 113 via one of serial ports 115 is audio recorder 117. In the preferred embodiment of the present invention, an RS-232 serial port is employed for the interference connection. However, it is foreseeable that other types of serial ports 115 commonly known in the art may be utilized. Serial port 115 may also be comprised of a direct hardware connection or may consist of a series of ports and connecting means commonly known in the art for connecting electronic devices. Audio recorder 117 may either be a stand-alone unit or incorporated into the hardware of central site server 113, or incorporated into other hardware devices within the system. Although it is preferred in the present embodiment that audio recorder 117 is digital, it is foreseeable that other known types of recording devices, as well as those not yet contemplated, may be employed in accordance with the teachings of the present invention. Audio recorder 117 records the conversations performed under the direction of telephone call management system 101. Audio recorder 117 may be activated for each call unless the number being called is specifically flagged for no recording or monitoring, such as calls to or from an attorney. Furthermore, audio recorder 117 can monitor multiple telephone lines simultaneously, using a different recorder channel number for each of trunk lines 111. The recorder channel number further enables the institution's staff to identify the call record they wish to review associated with a desired outgoing telephone call. Each user telephone 102 is further associated with a station identification number which allows the staff of the institution to identify the particular user telephone 102 a call was initiated and conducted from. It is foreseeable that the embodiment described herein supports up to 32 inmate telephone stations 103 and 24 trunk lines 111. However, multiple units 105 may be configured to expand the system to meet the capacity demand for the institution. Audio recorder 117 is also used to receive voice samples from the users for identification and authorization purposes. Specifically, when an inmate first “subscribes” he or she must state his or her name. Voice recognitions software stored on central site server 113 verifies that the user spoke his or her name. Alternatively, central site server 113 may be equipped with specialized hardware for receiving and identifying voice samples. Voice identification solutions are known in the art and available from companies such as Nuance, Buytel, Veritel, Integrated Wave Technologies, Intervoice Brite, etc.
Central site server 113 is controlled by software associated with administrative workstation 120. In the preferred embodiment, administrative workstation 120 is connected to central site server 113 via a local area network (LAN). However, it is foreseeable that other types of electronic connections may be employed. The administrative workstation's 120 software can modify call restrictions for individual users in addition to all telecommunication activity of the institution. Additionally, the software can also track a user's commissary information, such as the account balance if a debit system is being used. Furthermore, depending on the needs of an institution, the database may perform other functions.
Commissary workstation 121 is used in conjunction with administrative workstation 120 to manage and record a user's financial transactions. In the preferred embodiment; commissary workstation 121 and administrative workstation 120 are connected to central site server 113 via a LAN. However, other known connections, or connections not yet contemplated may be utilized. Commissary workstation 121 can also record other financial information, such as the total amount spent on collect calls by each inmate, amount spent on debit calls, the total net financial transactions for each user, etc.
Shadow workstation 123 and investigative workstation 125 are also employed in the present embodiment. Shadow workstation 123 and investigative workstation 125 are connected via the local area network linked to central site server 113 in the present embodiment. Shadow workstation 123 utilizes a live operator to monitor telephone calls without detection. It is foreseeable that this function may be performed by software integrated with shadow workstation 123. The shadow workstation 123 software provides a means for patching into a call using circuitry without alerting the user or called party to the operator's presence. If the operator finds that a call being monitored is suspicious, the operator may manually (or by using software) activate the audio recorder 117 to record a portion of an active telephone call. The called party's number may also be flagged in the inmate's profile (stored on administrative workstation 120 or central site server 113) to provide future monitoring of calls from the specific user to the specific called party.
Alternatively, software located on central site server 113 or investigative workstation 125 may be used to passively monitor calls. For example, when certain key words or phrases are spoken, voice recognition software may activate audio recorder 117 via electronic means and alert the proper authorities that a violation has occurred. The same voice recognition solution that is utilized in verifying a user's speech sample (i.e., that a user provides his or her name when subscribing to the system) may be used to identify such key words and phrases.
Furthermore, investigative workstation 125 controls other monitoring and security features interfaced in call system. For example, investigative workstation 125 can be to access past conversations stored on audio recorder 117. Software on investigative workstation 125 may also be configured to detect if a third party is present during a user's conversation. Investigative workstation 125 or central site server 113 may also contain voice recognition software to aid in calling or called party voice authentication. Administrative workstation 120, shadow workstation 123, investigative workstation 125, and commissary workstation 121 may alternatively be combined into one or several units. Furthermore, administrative workstation 120, shadow workstation 123, investigative workstation 125, and commissary workstation 121 may be integral within the central site server. It is also foreseeable that any component may be alternately located off site from the other apparati of the present invention.
Referring next to
In the present embodiment of the call system, a series of bilingual voice prompts are provided. These voice prompts guide the user through placing a call through the telephone call management system. Pre-recorded voice prompts instruct the user how to place the call and announce the call to the called party, providing identification of caller and location of call. These voice prompts may be digitally produced and presented in a concatenated form as is presently known in the art or other common form in the art. The voice prompts may be pre-recorded by the institution and retained in an integrated or remote database, or may be recorded in any form as is known in the art. Furthermore, these voice prompts can be played in the language specified by the user's profile, the language specified by the institution, or in a multitude of languages.
When a new inmate is entered into the system, in addition to assigning a PIN and storing identification information for the inmate, the system also stores a sample of the inmate's speech (step 241). In the preferred embodiment of the present invention, this sample speech should be the inmate stating his or her name. Although other speech segments can be recorded, it is important to require the inmate to record a sample of speech that is not easily repeated. Preferably, audio recorder 117 is utilized to record the utterance of speech avoiding the need for any additional hardware to incorporate voice recognition as part of the system of the present invention. The recording can then be stored with the database in any standard recording format (.wav, .mp3, .wma, etc.).
It is known in the art to utilize speaker dependent voice recognition algorithms, software, hardware, programs, etc. to identify a speaker. However, it is also known in the art that such systems can easily be defeated if the sample of speech can easily be mimicked. For example, a speaker identification program may fail to distinguish a speaker if the sample of speech is simply a whistle or a blow into a microphone. Therefore, the system of the present invention preferably ensures that when the inmate is prompted to record a speech segment (step 241), that the inmate speaks his name.
The system of the present invention utilizes speaker independent voice recognition software to automatically ensure that the inmate has provided a valid speech segment (i.e., his or her name) (step 243). As is known in the art, speaker independent voice recognition software is now commonly available as off-the-shelf software for computers that utilize standard operating systems (e.g., Windows, UNIX, Macintosh, etc.). This software includes voice recognition software modules that can easily be integrated into existing system software. Therefore, the voice recognition software can be readily installed on central site server 113 and can be incorporated into the system of the present invention. Of course, alternative speaker independent voice recognitions solutions (e.g., hardware solutions such as integrated circuits (les) or digital signal processors (DSPs)) can be utilized without departing from the spirit of the present invention.
Regardless of the specific implementation, the speaker independent voice recognition module receives the speech segment spoken by the inmate (and captured by audio recorder 117) (step 243) and outputs data representative of the words spoken by the inmate (step 245). The words spoken by the inmate are compared with the stored identification data to verify that the speaker indeed spoke his or her name (step 247). If the speaker did not state his or her name, the inmate may be given a second attempt (step 251). If it is already the inmate's second attempt, an account for that inmate can be denied (step 253), and the sample speech spoken by the inmate can be forwarded to site personnel for manual analysis (step 255).
If, however, the inmate speaks his or her name, as verified by the speaker independent voice recognition module, then the sample speech is stored in the database with the inmate's other information (PIN, identification number, cell location, etc.) (step 257). The initial subscription of a user may also include receipt of biometric information (e.g., thumbprint, retinal scan etc.) to be retained by the call system for future authorization (not shown in
After a user provides the necessary verification information, a user profile can be established including restricted call number, amount stored in a debit account, call time restrictions, and other such information. The user profile may contain lists of valid telephone services and types of screening used for that specific user. The information may be displayed to a user via an associated display means, may be electronically provided via a verbal call prompt or may be hidden from the user.
Turning next to
If a valid PIN is entered, the system next prompts the user to state his or her name (step 269). The user's spoken sample is received and compared with the stored utterance of the inmate's name (step 271). If the two samples match (i.e., the speaker dependent voice identification solution utilized by the present invention determines the samples were provided by the same person), the inmate is verified (step 273). If there is no match, the user may be provided with a second attempt (step 275). Any subsequent attempts may be blocked (step 267). However, if a user is verified (step 273) the user will be able to use the telephone management system of the present invention in accordance with the user's designated rights and restrictions.
Referring back to
Multiple trunk lines 111 are supported by electronic switchboard device 105. In addition, different routing to the PSTN may be assigned across multiple trunk groups. Use of multiple trunk groups may be required or preferred due to lower network access charges, routing facilities and usage costs, (i.e. local, international, long distance debit, long distance collect, etc.).
In addition, electronic switchboard device 105 can provide digit analysis based on dialed number identification system (“DNIS”), other dialed digits, etc., and can route a call via the appropriate trunk lines 111. Trunk lines 111 may interface the network on direct digital T1 circuits, analog circuits, or other like network interfaces.
The distributed architecture of the hardware and associated software of the electronic switchboard device 105 comprises Station Interface Subsystem (SIS) 203, Call Processing Subsystem (CPS) 205, Digital Interface Subsystem (DIS) 207, Data Communication Subsystem (DCS) 209, and Concentrator Communication Subsystem (CCS) 212.
Station interface subsystem (SIS) 203 provides switched connections to call processing subsystem (CPS) 205. CPS 205 controls digit collection, interactive voice prompts, call screening, network access, etc., during the inmate calling process. Specifically, during the call routing process, if trunk lines 111 in the primary trunk group are all occupied, a variation of call treatments may be implemented. For example, the call may be routed to a secondary trunk group, a voice message may be played, a congestion busy signal may be provided, etc.
In the preferred embodiment, SIS 203 contains four main components including SIC (Station Interface Card) 203 which provides power to all telephones, CSC (Cross-Point Switch) 201 which routes telephone calls from telephone bank 203 to the proper outgoing trunk line 111, LIC (Line Interface Card) 215 which converts analog telephone signals to a format compatible with the call management system, and Spybox 211 which is used for audio monitoring of user telephone calls. The basic function of SIS 203 is to detect and process off/on-hook call service requests from the telephones located at inmate telephone bank 103. SIS 203 also CO1U1 ects the line to an available call processor card (CPC) 210 port for processing the inmate call. In addition, SIS 203 provides switched audio connections to Spybox 211.
The originating user off-hook requires connecting station interface card (SIC) 213 voice paths via cross-point switch card (CSC) 201 to line interface card (LIC) 215. Voice path connections are switched and controlled by CSC 201. The selected LIC 215 outbound port connects the line to CPC port 210 for processing the call. Dial tone is provided to the inmate when CPC 210 is connected and ready for the inmate to enter digits.
Call processing system (CPS) 205 controls all routing and subsystem, interaction processes required by the call management system. CPS 205 contains one or more call processing cards (CPCs) 210 which provide voice prompts to users and receives and record DTMF and voice responses. Thus, CPS 205 is used extensively during the initial login wherein a potential user provides a PIN and a speech sample. Specifically, CPS 205 (specifically CPCs 210) provide prompts instructing the user to provide a PIN, and if the PIN is indicative of an authorized user, a speech sample for voice verification. CPS 205 then receives a user's response to the prompt for a PIN (preferably in the form of DTMF tones), and the user's spoken response to the prompt for a speech sample.
Station voice paths switched through the SIS subsystem 203 are connected to call processing card (CPC) 210. In the present embodiment, CPCs 210 have four ports per card. However, additional ports per card may be utilized in accordance with the objectives of the present invention. For example, in the preferred embodiment, call processing subsystem (CPS) 205 can accommodate up to six CPCs per call system unit allowing each electronic switchboard device to support up to twenty-four call processing ports.
CPS 205 can accommodate multiple CPCs 210 which allows system redundancy and system availability. Real-time call processing loads are distributed across the number of configured CPCs 210. In the preferred embodiment, the subsystem is configured with a minimum of two CPCs 210 per electronic switchboard device. For example, it is preferred that a minimum of two CPCs 210 are utilized as a fault protection. If one CPC fails, call processing would continue on the other active CPCs 210.
Call processing cards (CPCs) 210 support the specialized call processing features and controls required for an institution telephone service. When a user originated call is connected to CPC 210, a dial tone is returned to the user. The dial tone indicates that the call system is ready for the caller to enter digits. During the call process, CPC 210 interacts between other subsystems and the call, thereby supporting the necessary system call sequence control and prompts for completing the call.
CPCs 210 collect dial tone multi-frequency (DTMF) digit information, or like information, entered by the user and provides pre-recorded voice prompts stored in system memory card (SMC) 227 delivered to the user via an audio record/playback buss. CPCs 210 connect the audio/record playback to the user telephone. Interactive voice prompts instruct the user to enter a series of identification and/or authentication information. For example, a user may be required to provide voice information for authentication or recording, DTMF information responses such as a PIN, biometric information for authentication, or provide RF data. Prompt responses are detected and recorded via CPCs 210. Biometric responses are recorded via separate hardware in the call system using a voice buss that couples to the DIS. Voice responses are played on the audio record/playback buss to SMC 227 for processing. Processed voice signals are digitized and stored in memory (not shown). Once the originating call is processed and approved, CPC 210 will either connect the call to its associated network trunk lines 111 and outpulse the call or otherwise be released from the call so that the call can be connected by SIS 203 to an alternate CPC 210 for outpulsing the call. If the user's call is not approved (e.g., if the user's speech does not match the stored speech sample provided by the user during the user's initial subscription to the system), a special call treatment is returned to the user. Special call treatments can comprise voice prompts, busy signals, etc. For example, these special call treatments generally provide information concerning why the call could not be completed and processed.
When the call is outpulsed and answered, CPC 210 provides called party prompts to announce the call, which may include asking the calling and called party for voice verification. The called party may also have to enter a PIN to be authenticated. The system of the present invention supports subscribing called parties. When a potential called party subscribes to the system a similar process is used as when a user subscribes (see
The CPC hardware is laid out on a PC board design that supports two plug-in daughter boards. The main PC board is identified as the line card. The larger daughter board is identified as the line card extension board. The smaller board of the two is identified as the CLICK board.
Digital interface subsystem (DIS) 207 converts analog voice information to a digital format. Integrated analog/digital conversion card (IDC-MSI) 217 handles analog to digital (AID) conversion for the telephone call management system. Digital T1 interface card TLC 219 routes calls to CPU 223. CPU 223 contains software which controls user access to the telephone call management system. CPU 223 can also store all financial and authentication data. Furthermore, CPU can be capable of processing any other data as may be required within the system.
Digital interface subsystem (DIS) 207 provides an integrated digital T1 network interface capability for the call system. DIS 207 interfaces call processing system (CPS) 205 lines/trunk ports. DIS 207 formats the digital voice signals into a 24 channel digital T1 interface. In addition, DIS 207 processes user inquiries and performs account update transactions via the LAN. DIS 207 can include an integrated analog/digital conversion card (IDC-MSI) 217, digital T1 interface card (TLC) 219, PCM extension buss (PEB) 221, and a digital subsystem CPU controller card (CPU) 223.
The integrated analog/digital conversion card (IDC-MSI) 217 is a commercial design commonly employed in the art. The design is a proven technology and is utilized in a large number of switching applications.
This design is based on a Dialogic® modular station interface (MSI) board or other similarly designed boards. A PC-AT form factor board was developed by Dialogic® to support integrated digital switching functions. The board is compatible with PCM extension buss (PEB) 221 based designs. Furthermore, the board is compatible with the North American (1.544 Mb/s transmission rate, u-law PCM coding) and European (2.048 Mb/s transmission rate, A-law PCM coding) digital interface standards.
The Dialogic® MSI board consists of a motherboard that can accommodate up to six base modules or two add-on modules. The six-module version supports four analog port interfaces per module. The two add-on module version supports twelve analog port interfaces per module. Each version fully configured supports up to 24 inbound analog ports. These analog ports are connected to distribution blocks for grading to the CPC's line-side interface ports. The CPC and IDC blocks are used for cross-connection the CPC ports 240 to the IDC ports.
Data Communication Subsystem (DCS) 209 controls data communications between multiple call management systems. DCS 209 contains data communication card (DCC) 225, system memory card (SMC) 227, and PC Interconnect Card (PCI-1) 221.
Multiple unit systems require communications between units. This is supported by equipping one of the units with a communications concentrator subsystem (CCS) 212. CCS 212 contains system concentrator communication card (SCC) 229, concentrator memory card (CMC) 231, and a second PC interconnect card (PCI-2) 233.
Completed calls to trunk lines 111 require that caller identification and instructions be provided to the called party. A variety of programmed voice prompts can be used to announce the call and to instruct the called party. Typical voice prompts include information regarding where the call originated from, the type of call (i.e., collect/prepaid), how to accept or decline the call, how to deny future calls from the same caller, etc. For example, if the user (John Doe) places a collect call, the message “You have a collect call from John Doe. Dial -55- to accept the call or hang-up to decline the call,” may be played. If the called party enters positive acceptance, the caller is provided a transmission path.
Throughout the duration of the call, the system monitors the called party line for switch hook flashes. Detection of these flashes may indicate potential three-way calling/conference feature activation by the called party. If a hook flash is detected, the system may be programmed to limit the call to a certain time duration, and/or to play a warning tone or play an announcement to both parties thirty seconds prior to disconnect.
In addition, design features prevent the user from reaching live operators or the ability to chain dial. Each call process requires that a specific disconnect duration to the network is completed. New call attempts are forced through a rigid call state sequence and screening, which includes a number of authentication means such as a PIN, biometric information, and/or RF authentication.
A number of features are provided to aid in call screening. Each user profile may contain a list of telephone numbers to which calls may be placed. Certain exchanges or prefixes, such as 1-800, may also be blocked. Other options, such as the number of calls allowed, call minutes allowed, or restriction to specific forms of call payment may be specified.
Concentration communication subsystem (CCS) 212 is responsible for supporting communications between the call system units and to the servers. CCS 212 includes a system concentrator communication card (SCC) 229, a concentrator memory card (CMC) 231, and a second PC interconnect card (PCI-2) 233.
CCS 211 is configured using the same basic hardware cards as utilized in the data communication card system. However, each memory card is independent and operates under different software systems.
Call detail records (CDRs) collected in the CPCs 210 are typically communicated to DCS 209 over the CPC COM port disposed within call processing system (CPS) 205. The CDRs are then buffered in SMC 227. The CDRs may also be transmitted to CCS 212 and buffered in CMC 231. CMC 231 may act as an interim backup for the CDR records. When requested, the buffered CDR's are also transmitted via CCS 212 to the server(s).
Each CDR transmitted and acknowledged by the server is flagged by the memory cards. Servers use a polling method during low traffic periods to upload CDRs from the memory cards. During the next polling sequence, only the CDRs that have not been acknowledged are transmitted to update the server CDR database. Basically, CDRs may at one point be buffered in a DCS memory card (SMC 227), a CCS memory card (CMC 227), or a server database (located in central site server (113 from
Referring next to
As shown in
SIC 213 has subscriber line interface circuits 301 to connect and provide power to user telephones 103. Individual switch-hook detect circuits 305 monitor the telephone lines for on/off-hook states. Generally, the normal line state is either idle or on-hook. However, detection of an off-hook state represents a new call service request.
SIC 213 converts the 32 balanced 600 ohm two wire line circuits into 32 individual single wire two-way voice busses. The 32 voice busses connect to [A] point array 313 and [C] X-point array 315 on CSC 201. Busses then may be connected via CSC 201 to idle CPC ports 240 for call processing. When a valid connection is established, the CPC returns a dial tone to telephones located within user telephone bank 103.
Calls originating from SIC 213 that cannot be connected to idle paths generate a busy tone from generator 317. CSC Z8 control processor 309 selects the send busy circuit in the SIC 317 (via circuitry not depicted). SIC 213 connects the busy tone signal 319 to a telephone in phone bank 103. When the telephone disconnects, the circuit is released and the telephone is ready for a new call origination.
CSC Z8 control processor 309 on CSC 201 continuously queries switch hook memory detect circuit 305 to detect new off-hook service requests. When a permanent off-hook or an invalid connection is detected, CSC Z8 control processor 309 enables send busy signal 317. This places the station line in a permanent or lockout busy state. The condition may be cleared when the telephone or faulty off-hook status is restored. When a condition is restored, the station line can originate new calls.
The second main component of SIS 203 is cross-point switch card (CSC) 201, which provides a 32 by 24 cross point switching matrix for connecting stations to call processors and network trunks. In addition, it also performs switch connections for Spybox 211 monitor access.
CSC 201 supports an automatic reroute feature. Connections to CPC ports 240 during the call process may be switched to a second CPC port. This reroute may be required to access an idle trunk for a different call type route. This feature allows call connections to be rerouted in the system to pre-designated call treatment types or alternate trunks.
Upon initialization, CSC Z8 control processor 309 reads and loads programs from onboard E-PROM based program memory 307. During initialization, CSC Z8 control processor 309 performs subsystem reset 321. At this point, CSC Z8 control processor 309 performs hardware diagnostics and data validation. Configuration control information is sent to the Data Communication Subsystem (DCS) 209 concerning the in-service station location addresses which are mapped into program memory 307. Under control of CSC Z8 control processor 309, address decoders decode the on/off states. The CSC 201 decodes the addresses to select [A] X-Point array 313 or [C] X-Point array 315X-Address location for that individual station.
During operation, onboard watch dog time (WDT) 323 monitors the processor operation and sanity. WDT 323 will automatically reset the hardware if any abnormal condition would prevent the recovery of CSC Z8 control processor 309. Reset 321 then causes the subsystem to re-initialize and return to a normal in-service state. In the case of an invalid on/off hook state or connection, CSC Z8 control processor 309 reset logic can restore the call connection to an idle state, set busy connect for permanent off-hook stations, or reset LIC connect relay 325. LIC connect relay 325 may be enabled or reset under control of the CSC Z8 central processor 309 to connect or release the CPC port. Release of the CPC port will disconnect the forward connection to the network.
In the present embodiment, concentration may be introduced by CSC 201 between its 32 station ports and 24 LIC ports. The level of concentration per unit, provided all 32 stations and 24 trunks are assigned, is 32:24. This would limit the number of simultaneous calls per unit to 24. Concentration levels may be varied by the assignment of stations and trunk lines across the equipped unit. For example, these levels can be 1:1 for a non-concentrated configuration. It is foreseeable that alternate concentration levels may be provided depending on the number of stations and trunk lines utilized in the telephone call management system.
Still referring to
When CSC Z8 central processor 309 detects a new call origination, the outbound 24 voice busses on [A] cross-point array 313 are selected first. The first choice routes are to the CPC ports 240 within the same unit. Secondary routes to CPC ports 240 in other companion units are connected through the [C] X-Point array 315.
[B] X-point array 327 provides a 16 by 24 matrix. Eight of the sixteen voice busses represent the inbound unit/unit OUT and unit/unit IN busses. These busses are used to switch station connections to and from other companion units. Outbound access to the busses is via [C] X-Point array 315. Unit/unit inbound access to the 24 LIC 215 voice busses is through [B] X-Point array 327.
The other eight busses are unit/unit IN busses, commonly referred to as half busses. These busses support inbound voice connections for Spybox 211 monitor connections. Monitor access is provided through [B] X-Point array 327, establishing the monitor connection on the trunk side of the call path. A set of dipswitches (16 switches) on CSC 201 provides the option to connect or disconnect the unit/unit voice busses.
The processor's serial COM port (located on the processor) provides data communications between CSC 201 and data communication subsystem (DCS) 209. COM port 329 supports inter-processor communications between units for call connections and unit/unit call control. In addition, in an offline mode, the port may be used to support external maintenance and debug access.
A third main component of SIS 203 is line interface card (LIC) 215, which interfaces SIS 203 to call processor subsystem (CPS) 205. LIC 215 converts the 24 outbound voice busses from CSC 201 to 24 (balanced 600 ohm) two-wire interface circuits. These circuits are connected to individual CPC ports 240. The CPC ports 240, under control of CSC Z8 control processor 309, provide access for call processing and network trunk lines.
Each of the 24 LIC inbound ports directly interface the voice buss from CSC 201. The audio path conversion includes a balanced 600/600 ohm transformer coupled circuit (not shown) and connect relay 325.
Connect relay 325 controls the seizure and the release of the associated CPC port 240. When the relay circuit is enabled, the LIC port extends an off-hook to CPC port 240. In a normal or release state LIC 215 extends on-hook status to CPC port 240. Control of the relay is performed by CSC Z8 control processor 309 address/data buss via I/O address decoder 331. A current detect circuit 333 output is read by CSC Z8 control processor 309. Current detect circuit 333 monitors the loop current in the connection to the CPC port. This permits CSC Z8 control processor 309 to detect the call path connect and disconnect status.
A coupled component of SIS 203 is Spybox card (SBC) 211, which supports non-intrusive access to monitor and record user calls. In the present embodiment, the eight SBC inbound ports interface to the CSC X-point single wire voice busses, which then convert the single wire technology to a balanced 600 ohm two-wire tip and ring voice circuit. The eight outbound two-wire ports connect to remote audio monitor devices. Monitor access in each unit is connected by X-point array [B] 327. Each monitor path is cross connected to one of the 24 inbound voice paths to the LIC.
Spy channel access is connected through the [B] X-Point array 327. The actual monitor connection is made at each unit's LIC inbound voice buss. Therefore, each monitor point connection is made at the trunk side of the telephone call.
The eight spy channel busses have access to a set of eight unit/unit voice busses. These unit/unit buss connections provide access to other companion units in the system. The unit/unit busses connect to the [B] X-pt array 327 in each equipped unit.
The CSC Z8 control processor 309 controls activation of a spy channel connection to an individual telephone call. A monitor request by a Spybox workstation routes a unique broadcast message to the call system units. The broadcast message is sent to each equipped CSC 201 via the COM port. CSC 201 (with the actual telephone call) acknowledges the broadcast message and then completes the monitor connection.
The eight inbound spy busses connect to line current detect circuit 333 on SBC 211. This interface circuit converts the voice buss into a 600/600 ohm balanced two-wire tip and ring line circuit. These line circuits from SBC 211 connect to Remote Audio Interface (RAI) circuit module 335. Each SBC line circuit to RAI 335 has a line current regulator and a line current detector circuit.
The line current regulator is enabled or disabled under control of CSC Z8 control processor 309. The line current regulator controls loop current to RAI 335 when the monitor connection is established. The line current detect circuit senses the loop current to RAI 335. Output of the detect circuit is addressed and read by CSC Z8 control processor 309. This allows the processor to determine the RAI module line connection state.
The final component of SIS 203 is spy remote audio interface module (RAI) 335, which is a single port audio monitoring module. Each RAI 335 uses line power from SBC 211 to detect an active monitor connection. An isolation transformer connects the inmate audio to the monitor OUT and speaker jack. RAI 335 can be placed at various monitor locations at the inmate facility. Each RAI 335 provides a control circuit for activating the recording device.
An isolation transformer provides a balanced 600 ohm tip and ring circuit to the device for recording the inmate conversations. A ⅛″ speaker T, R, and S stereo jack located on the module may be used to support an external speaker connection.
Now referring to
The LIC outbound ports connect to CPC telephone side ports 407. The telephone side port signal leads 450 are identified as T-out and R-out.
The LIC outbound ports extend a loop closure signal to the CPC 210. This initiates CPC 210 seizure and connects the station voice path. CPC 210 enables DTMF receivers 409 and returns a dial tone that confirms originating access. This signals the system is ready for the system user to press digits on the telephone keypad. In addition, onboard tone generator 411, which is under control of the call processing card processor 401, sends a dial tone to the station via the line interface logic.
CPC line interface logic design 413 includes both forward connect circuit (FCR) 415 and second connect circuit (SCR) 417. Processing a call requires the station and line side paths be independently controlled by call processing card control processor 401. This allows call processing card control processor 401 to control transmit and receive path information during the user call process.
After the initial seizure of CPC telephone side port 407, the station and line side are split via FCR 415. Enabled DTMF receivers 409 detect the information digits entered by the user. A typical dial plan allowed may include 1+10 digits for debit account type calls terminated to national numbers, 0+10 digits for collect type calls terminated to national numbers, or any other foreseeable combination of digits. Biometric and/or voice verification may occur either before or after DTMF tones have been entered.
Call processing card control processor 401 collects the digits entered by the user and routes the digits to a database capable of providing digit comparison. Digit analysis is performed to determine the call type (collect, debit, speed dial, national, international, etc.).
The call system may require users to enter a Personal Identification Number (PIN). Any range of digits may be used. However, to handle most institution requirements, the range of the PIN is one to nine digits or characters typically. Each number is unique to a user. The PIN may be used to index a discrete user file by the User Telephone Account Control (UTAC) server or an inmate telephone account control (ITAC). COM port 419 on CPC 210 supports communications with the server via data communication control card (DCC) (i.e., 225 of
Bong tone 421 is played to the user after the digits are entered. Bong tone 421 signals the user that the system is in the auto attendant mode. In the auto attendant mode, interactive voice prompts will interface to the user and guide the user through the calling process, possibly asking for biometric authentication, and providing a means to interface the user entered data to the call system.
Each CPC has audio record/playback logic 423 interfaced to four I/O busses 425. I/O busses 425 connect to the call system voice memory card (SMC) (i.e., 227 of
The telephone call management system of the preferred embodiment supports playing voice prompts that guide a user through the call process. Audio record/playback logic 425 supports recording of the user's name and the called party's name for later use in voice prompts and/or voice authentication functions. In addition, audio record/playback logic 425 supports playback of pre-recorded voice announcements to the called party when answered. Based on call type and the user data profile, different voice prompt menus may be selected according to the user's preferred language and other like options.
The interface logic's line side is open until CPC control processor 401 is ready to extend the call. Line side ports interface and connect toward the network. Port signal leads 408 are identified as T-in and R-in. Port signal leads 408 connect to either PSTN network analog trunk line 111 facilities or integrated analog/digital conversion card (IDC-MSI) (i.e., 217 of
A trunk group is determined during the digit analysis based on the dial plan. If the current CPC trunk interface is a member of the selected trunk group, the trunk can be seized. CPC processor 401 enables SCR circuit 417 to seize the appropriate trunk circuit.
Each line side port may be either a loop or ground start operation. Ground start relay circuit 427, under control of CPC processor 401, provides the ground start feature.
When the CPC trunk interface is not in the trunk group selected, the call must be switched to an alternate CPC path. The alternate CPC path selection is initiated by message via COM port 419 to the DCC (i.e., 225 of
DTMF senders 409 under control of CPC processor 401 are attached to outpulse the network information digits. Based on the network trunking plan, various interface protocols may be supported (debit, collect, long distance, etc.).
During network seizure, information outpulsing and call setup, various call progress tones or states may be encountered. The CPC line interface logic supports detection of these network progress tones and supervisory states. Progress tone decoder circuit 429, under control of CPC processor 401, detects the various network progress tones. Complex software and hardware algorithms are used to detect network progress tones and states. The states include dial tone 431, congestion busy signal, subscriber intercept tone, call intercept announcement, line busy signal, ringback tone 433, ring no-answer, answer supervision, and quiet. Loop polarity detection circuit 435 in CPC 210 supports hardware answer supervision. The detect circuit looks for a reverse tip and ring loop polarity.
Optionally, CPC 210 has “CLICK” detect circuit 437 which monitors the network line when answer supervision has been declared. The circuit supports detection of a rotary dialed called party acceptance. Call acceptance by the called party using a touch-tone telephone is detected via an enabled CPC DTMF receiver. “CLICK” detect circuit 437 also supports flash-hook detection used for third-party/conference call setup by the called party telephone.
The PSTN analog trunks utilized with the present invention are typically two-wire line interfaces used for local TELCO access. For this type of line, CPC 210 connections appear on type 66 distribution blocks, which are used to cross-connect the CPC ports 240 to the TELCO lines, or to the IDC ports.
Referring next to
The CODEC's digital audio signals are gated onto the time slot interchange (TSI) switch. The TSI receives digitized audio signals from the CODEC. Channel switching/connection signals are provided from the onboard control processor for each of the 24 channels.
In the preferred embodiment, the TSI switch acts as a traffic coordinator to buffer and gate the digital data from each channel. After the digital data is managed by the TSI switch, the digital signals are routed to PCI-I 221 (shown in
IDC-MSI card 217 is slave to the digital subsystem CPU card (CPU) 223. In the present embodiment, IDC-MSI card 217 edge connector supports PCI/ISA control buss format, however, other formats may be supported.
In the preferred embodiment of the present invention, DTI hardware 505 is a commercial design presently known in the art. Deployed in a large number of switching applications, the design is a proven technology. Specifically, the design is a derivative of a hardwire design similar to Dialogic's® D/24-SC-T1 digital T1 interface card.
The basic function of TLC 219 is to provide an integrated digital T1 network interface. When using the institution call system of the present invention, it replaces the need to use special channel bank type equipment.
Digital subsystem CPU controller card 223 hardware is a commercial design. The design is a proven technology and supports various TELCO system applications. Key functions of CPU 223 include the processing of the digital interfaces, subsystem statistics, user inquiries, etc.
Digital subsystem controller card 223 is a full size card. The Pentium processor based CPU 223 accommodates up to 256 MB DRAM memory. A secondary level 512 KB cache is also provided. Communications between digital subsystem CPU controller card 223, TLC 219, and IDC-MSI 217 is provided by the PCI/ISA buss. Floppy disk controller 509 supports loading updated programs/data files. Hard disk 511 provides storage media for digitized voice and data files.
Inquiry requests may be initiated by a user of the system from one of the available telephones. Unique access codes along with the PIN number, biometric authentication, voice samples, or RF authentication, may be entered once the user receives dial tone from digital subsystem CPU controller card 223. When digital subsystem CPU controller card 223 receives the digits, it connects to the line side and resends the digit information to IDC-MSI 217. The analog information is digitized by IDC-MSI 217 and routed via PEB buss 550 to TLC 217. Information digits may then be processed by digital signal processors (DSPs) in TLC 217.
TLC 217 has drop-and-insert capability to support digital data outputs that may be processed by digital subsystem CPU controller card 223. The drop-and-insert feature in addition supports playback of digital voice prompts during the inquiry process.
Digital voice files are preferably stored on hard disk 511 of digital subsystem controller card 223, although the digital voice files may be stored anywhere accessible to the system, whether local or remote. The processing of digit information and the use of special menus support the interactive inquiry process. In this embodiment, the recorder is incorporated into CPU 223. CPU 223 also contains software capable of analyzing biometric data from biometric sensor 109 (all depicted in
Digital subsystem CPU controller card 223 LAN interface supports communication with the servers and associated inmate account information. This supports account transaction processing between ITAC and other inmate data files.
Digital subsystem CPU controller card 223 board connectors support connection to external I/O devices. For example, CPU 223 may include display 513, keyboard 515, and/or COM ports 517 and 519.
Now referring to
PEB buss 550 supports up to 24 time slots in the preferred embodiment of the present invention. However, the number of time slots may be altered depending on the desired capacity of the call system of the present invention. This enables the telephone call management system to route channels to/from the IDC-MSI card. Each time slot is a digitized bit stream and represents one voice channel. This enables T1 interface card 603 to switch voice channels on PEB buss 550 to and from the IDC-MSI card analog interface ports.
PEB buss 550 time slot data may be routed to a series of digital signal processor (DSP) 607 interface. DSP 607 processes the digitized audio signals data on each channel. This design supports channel drop-and-insert capability. Under control of TLC onboard control processor 605, digital data may be extracted from the bit stream and/or inserted into the bit stream.
Digital signal processor 607 supports user inquiry features of the call system and any voice authentication that may be employed. Digital signal processors 607 may be programmed to perform signal analysis, to automatically adjust gain control, to compensate for variations in the level of incoming audio signals, to compress digitized voice data compression, to send and/or receive DTMF or inband signaling, to monitor channel conditions and status, to detect presence of tones (DTMF, MF, etc.), detect silence/quiet, to determine if a caller is not responding, to decompress stored audio data, to compress audio data for playback, to adjust the volume and rate of speed of playback, to signal bit control (off-hook, on-hook, etc.) based on trunk types (FXS, E&M, etc.), etc.
TLC onboard control processor 605 controls TLC 603 operation via local control buss 609. Local control buss 609 interprets and executes commands from TLC onboard control processor 605. Communications between TLC onboard control processor 605 and the host CPU is via dual port shared RAM memory 613. Dual port shared RAM memory 613 acts as an input buffer and/or output buffer. Upon initialization, the operating firmware that controls TLC 603 is downloaded from the CPU. It is downloaded into the onboard code/data RAM via dual port shared RAM interface 614 via PCI/ISA control buss 503.
Specifically, control of the digital interface subsystem is provided by configuration manager 611. Configuration manager 611 determines and sets various board level operational parameters. This feature eliminates the need to set confusing jumpers or dipswitches.
Referring next to
In the preferred embodiment of the present invention, one of UART ports 705 supports an internal modem circuit for remote dial access. In addition, six of UART ports 707 are dedicated for connecting up to six CPC serial COM ports 701 from the CPS. Other serial COM ports 701 include one port dedicated for CSC communications and a second COM port dedicated for communications with the system concentrator communication card. Communications between the data communication card and the system memory card is via internal processor COM ports 708.
A battery backed real time system clock 709 on DCC 225 supports an accurate date and time stamp function. This capability is used for time stamping call detail records (SMDR) and reports.
Data communication card central processor 717 sanity is monitored by internal watch dog timer (WDT) 711. This feature provides a hardware type reset for data communication card central processor 717. For example, if a non-recoverable error condition affects the processor sanity, the feature would force reset data communication card central processor 717. Furthermore, WDT 711 feature may force data communication card central processor 717 and/or SMC processor to reset.
DCC 225 has two types of memory located on the board, which include ROM (read only memory) 713, equipped at either 256 or 512 kB in the present embodiment, and SRAM (static random access memory) 715, equipped at 512 kB in the present embodiment. ROM 713 contains the operating programs for DCC 225. On power-up, data communication card central processor 717 boots and initializes the operating programs.
Battery-backed SRAM 715 supports real time and configuration data requirements. Configuration data may be site specific. Alternatively, site programs can be downloaded from remote operations centers.
PCI-1 703 is based on a passive card design. It consists of connectors that distribute and connect signals between subsystem cards. Each call system unit is equipped with one PC card referred to as PCI-I. The PCI-1 card supports the distribution of signals between DCS, SIS, and CPS in each unit. The PCI cards plug into a standard card slot in the Call system unit.
DCC 225 controls and performs communications functions between electronic switchboard unit subsystems. In addition, DCC 225 supports communications with units and subsystems via the CCS. Some operational and maintenance features of the DCC 225 hardware include visual indicators, special line modem access jacks, external serial COM port access jacks, reset & write protect switches, memory equipped options, etc.
In the preferred embodiment, modem access for remote maintenance and administration is supported by line modem 720 and line modem jack 718, such as a RJ14 telephone jack. Generally, the access line usage is low and may be controlled. Therefore, an option makes it possible to share the line for user calling to allow for efficient operations of the call system.
Next referring to
The preferred embodiment of the present invention provides for four (8 by 4) cross-point arrays 809. However, other cross-point arrays and configurations may be used. Each cross-point array 809 provides access to the four voice processing circuits on the SMC provided in the preferred embodiment of the present invention. Each equipped CPC provides four audio line record/playback buss connections to SMC 227. Based on a fully equipped unit (6 CPSs per unit at 4 audio line busses each) a total of 24 audio line buss cross-connections are used in the present embodiment. For example, each one of the first three arrays supports four audio line busses from 2 call processing cards. Specifically, each array may access any of the four voice processor circuits 811 of the present embodiment.
Voice processor circuits 811 code user voice signals into digitized voice files for recording and use for authentication. Voice processor circuits 811 also decode user digitized voice files and convert the digital signals to audio signals for playback.
Furthermore, first-in, first-out (FIFO) buffer 813, along with the SMC processor's 805 DMA buss allow larger digitized voice data files to be moved fast and efficiently to and from memory. A multitude of FIFO buffers 813 may be utilized for providing a more efficient call system.
Alternatively, or in addition to the FIFO buffers, memory extension modules 815 may be utilized and designed as plug-in modules for the SMC 227 (as depicted in
Referring next to
Specifically, in this embodiment, one DART port 907 supports an internal modem circuit 909 for remote dial access. Typical COM ports used in SCC 229 include four serial COM ports which are used to connect to the Dec in each electronic switchboard device unit equipped and one serial COM port used to connect to a server COM port. In the present configuration, the SCC processor's COM ports 905 connect directly to the CMC 231 COM port. Furthermore, a battery backed real time clock circuit 911 is also part of the SCC design.
SCC internal processor 915 sanity is monitored by an internal watch dog timer feature 913. This feature provides a hardware type reset for SCC internal processor 915. If a non-recoverable error condition affects the processor sanity, the feature force resets the SCC internal processor 915.
SCC 229 comprises two types of memory located within the system. Specifically, SCC 229 includes programmed memory ROM 916 (read only memory), equipped at either 256 or 512 kB in the present embodiment, and data memory SRAM 917, (static random access memory) equipped at 512 kB in the present embodiment. Although the specific memory capacity for the system is provided for disclosure purposes, it is foreseeable that additional memory, either internal or external to the system, may be provided depending on he desired capacity of the call system. Generally, ROM is used for program memory and contains the SCC's operating software.
In the preferred embodiment, CMC 231 performs buffering for call detail records as a backup to the system memory card. The CMC 231 communicates with the system concentrator communication card via the processors internal COM ports 905. Together they support data communications in the telephone call management system. Watch dog timers 923 on the CMC and SCC processors monitor sanity. A reset from either processor will reset the companion processor. The CCS's CMC uses the same basic hardware as the SMC equipped in the DCS.
Call system software controls all monitoring, recording, financial transactions, and other call processing features. In the preferred embodiment of the present invention, call system software contains six main components. A general description of the call system software is provided herein to offer a general understanding of the possible software for use with an investigative call system. However, the description provided is not intended to provide the full scope of software functions compatible with the present invention. For example, a system administrator section controls which institution authorizes hardware and telephone access to the system. A user administrator section controls which prison authorities have access to the software. Specifically, in a prison environment, the warden may have access to modify all features within the system whereas a guard may only be able to change user profiles in his own cellblock. The account section allows inmate profiles to be created and modified. It monitors the inmate's calling data and financial transaction data. A shadow section allows the software to control the various Spybox monitoring stations to listen to inmate calls. A fifth section allows users of the software to compile various system reports, such as net monthly financial transactions and an extra digits dialed report. A final section processes the biometric information and the RF data for use in authentication. This section uses various algorithms to check a user's recorded information against data supplied to the biometric scanner and RF receiver.
System administration software allows an institution staff member to define defaults and to customize the system. Generally, only authorized staff members may have access to customize system settings, based on individual staff member security levels. However, a user security level may be determined when a user first logs into the system containing the call system software based upon username and the access level that has been set for each user name by a user manager. The sub-menus of software may include class of service (COS) maintenance, living unit maintenance, telephone location maintenance, facility telephone number control, nationwide telephone number control, call pricing, facility default maintenance, transaction type maintenance, etc. In addition, the institution employees may have access to update multiple telephone lists, view calls in progress, monitor calls in progress, manually modify transactions, enable/disable telephones, modify the class of service, etc.
Referring next to
Now referring to
Referring next to
In this example, all parameters except the line number can be changed by staff members' intervention. Further options may be accessible depending on the requirements of an institution.
Facility telephone number control screen 1300 additionally contains call type block 1317 that contains a number of fields indicating the type of call. Block field 1301 is selected if a telephone number is not allowed on the user's personal list. Allow field 1303 is selected if the telephone number in telephone number field 1302 is on the user's allowed list. If no list required field 1304 is selected, a user does not have to place the telephone number on their personal list in order to call it. Type block 1317 additionally contains collect field 1319 which is selected if a user is authorized to call the number collect. Similarly, type block 1317 also contains direct field 1321 which is selected if a user is authorized to call the number utilizing direct call means.
Facility telephone number control screen 1300 additionally contains a number of days field indicating the number of days a telephone number is actively on the list. In this example of the present invention, setting number of days to “*” will cause the number to be active on the list indefinitely. Any number greater than “*” will cause the number to only be active for that number of days.
Brief comment field 1311 may be used to describe the reason a telephone number is on the list or any other like comments. The user security level that added the number is indicated (and cannot be modified) in “User” field 1313.
Depicted in
Telephone number control screen 1400 contains a list of telephone numbers 1402 which can be assigned a “Blocked” designation 1401 or an “Approved” designation 1403. Blocked numbers will not be connected even if they are on a user's personal list. Approved numbers are connected even if they are not on the user's personal list. Numbers can also be assigned an “Exclude” designation 1404 set to exclude. Excluded numbers are not included in any lists or filters generated using wildcards. The number of extra digits dialed field 1405 can be used to limit the number of digits a user is allowed to enter after a telephone call has been connected to prevent a user from achieving an unauthorized connection or for certain-gaming telephone options.
A number of days field 1407 is used to set the number of days that a number is actively on a user's list. In this example of the present invention, setting number of days to ‘0’ will cause the number to be active on the list indefinitely. Any number greater than ‘0’ will cause the number to be active for only that amount of days, starting from when the number was added to the list as indicated by Date field 1409.
Comment field 1411 is used to describe the number or the reason the number is on the list. User field 1413 indicates the user security level needed to add/edit a number on this screen. In this example, user field 1413 cannot be modified.
Referring next to
A class of service parameter 1509 allows an authorized user to choose a specific class of service from a drop down list (e.g., “Default COS”, “COS 1”, “COS 2”, etc.). Number of living units field 1511 indicates the number of living units in each facility. Number of telephone stations field 1513 displays the number of telephone terminals in use in each living unit. Number of trunk lines field 1515 indicates the number of trunks available at the facility.
Facility default maintenance screen 1500 additionally includes user default block 1516. In the example of
Telephone number block 1520 contains the default settings used for all telephone numbers in class of service field 1509. Direct call field 1523 indicates if direct call access is allowed. Similarly, collect call field 1521 indicates if collect call access is allowed. Record field 1525 indicates if calls are to be recorded. Allow field 1526 indicates if added telephone numbers are to be allowed by default. Alert field 1527 indicates if all telephone numbers added under this particular class of service are to be flagged in a user's profile. Additionally, telephone number default block 1520 also contains an extra dialed digits field indicating the number of digits that a user may press after a call has been connected and a maximum active telephone numbers field 1531 indicating the number of allowed telephone numbers on any user's list.
Multiple telephone list update screen 1600 additionally contains fields limiting access to different call methods. Allow field 1613 is used to indicate if a user is authorized to contact the specified number. A not allow reason field 1615 allows a comment to be added indicating why the number is blocked. Collect field 1617 indicates if a user can access the number utilizing collect call means. Similarly, direct field 1619 indicates if a user can access the number utilizing direct call means.
Referring next to
Now referring to
By clicking monitor button 1817, an authorized software user may monitor live telephone conversations selectively. End monitoring button 1819 is used to end live monitoring.
Now referring to
Transaction type field 1905 indicates the type of transaction selectable from a list. The selection may be deposit, withdrawal, exception, amount of transaction, reference number, etc. Upon completing all manual transactions for a user, a manual transactions report will be generated automatically.
Manual financial transactions screen 1900 is also used to close a user's account. This is done by setting transaction type 1905 to “release”. When a user is released using the manual transaction screen, the user's account status automatically changes to inactive, and assuming the user is not using the account at the time, the remaining balance is transferred to a commissary account. No further calls or transactions will be possible with this account while their status remains inactive. In addition, a record of this release will automatically be made on the user release report.
Enable/disable telephones screen 2000 also allows authorized users to enable/disable all calls at the living unit level or the facility level. Living units block 2015 is used to control the line state status of all phones in each living unit. Telephone terminals are enabled by choosing enable button 2017. Disable after current call button 2019 is used to disable telephones after all telephone stations are free whereas cut off button 2021 is used to disable all telephones even if calls are in progress.
To enable/disable telephones at the facility level, buttons in facility block 2023 are utilized. Telephone terminals are enabled by choosing enable button 2025. Disable after current call button 2027 is used to disable telephones after all telephone stations are free whereas cut off button 2029 is used to disable all telephones even if calls are in progress.
Now referring to
Different lists of users are accessed from groups block 2107, typically located at 23 the bottom of the screen. User groups may include, but are not limited to, account operators 2109, administrators 2111, backup operators 2113, and guests 2115. Each user group may additionally be given a selection. User security profiles and groups can be modified by double clicking on the desired user or group.
Referring next to
Now referring to
A user must have a system account established in order to make telephone calls from a specific facility. This information will be stored on the site server which may be integral or remote from the call system architecture. When an inmate is transferred from one facility to another, only the inmate's account information, COS, and telephone lists are transferred to that facility. However, previous information may remain in an archived database or other storage system.
Register number field 2301 contains an eight-digit number unique to each user. Although eight digits are utilized for the present invention, numerous character strings may be utilized depending on the number of intended users. A name field lists the user's first, last, and middle name. Facility field 2307 indicates the facility at which the user is located. Living unit field 2309 indicates the living unit in which a user is located. A language field 2311 is used to select the language of voice prompts used by the call system. Personal identification number field 2303 displays the user's PIN used to access the call system. The PIN is used in conjunction with biometric authentication and/or RF authentication. For example, as discussed with respect to
In the example of a penitentiary, a find inmate button 2313 allows access to an inmate's account from any page on the inmate account information screen. User accounts can also be added from this screen via the Add Inmate (or Add User in other embodiments) button 2315. Using the “>” button 2317 and “<” button 2319 buttons allows for easy navigation through inmate profiles.
The account activation date field 2403 indicates the date a user's account was originally created. It is supplied automatically by the program and cannot be modified. Financial and call records for an inmate should be available back to this date. In the present example, the most recent data is kept locally on the site server, unless the inmate had been transferred within that time, in which case only the data since the transfer is kept.
The date of arrival field 2405 displays the date that an inmate arrived at their current facility. The date of arrival is automatically generated when an inmate is transferred to a new facility.
Inmates may be assigned a status code, displayed in status code field 2407, to help separate them into various classifications. The status code is a single letter code from A-Y. The definitions for these codes are established by the central site server. In the present embodiment, a status code ‘Z’ indicates that the inmate has been released, and their account is inactive. Status ‘Z’ cannot be set manually. It is done automatically once an inmate has been released using the manual transaction screen. An account with a status code ‘Z’ cannot make calls or process transactions from that facility.
Suspension field 2409 indicates whether or not a user's calling privileges are suspended. Similarly, alert field 2411 indicates a user's current alert status. Comments field 2413 can be used to display any information regarding a user's financial, alert, or suspension status.
Buttons located in the lower right corner of account screen 2400 are used to modify a number of user settings. Print PAC button 2417 is used to print a user's current PIN or other access number. Change button 2419 is used to modify data such as user name, living unit, user language preference, status code, and comments. Change register button 2421 allows authorized personnel to change a user's identification number. Change PAC button 2423 is used to modify a user's current PIN. Alert button 2425 toggles alert field 2411. Finally, suspend inmate button 2427 is used to toggle suspension field 2409.
The prison staff has the capacity to temporarily suspend an inmate's calling privileges using suspension block 2429. Number of days field 2431 is used to set the length of a user's suspension. Begin date field 2433 and begin time field 2435 are used to input the respective date and time that a user's suspension is to begin. Similarly, end date field 2437 and end time field 2439 are used to input the respective time and date a suspension is to end. Information regarding a user's last suspension remains displayed in suspension block 2429 even after the suspension has expired.
Financial transaction screen 2500 also includes a number of buttons located in the upper right hand corner. Display button 2517 refreshes the financial transaction list for the month and year selected. Display button 2517 must be pressed each time the year and date are changed. Current month button 2519 brings up financial transaction data for the current month. Sort order button 2521 determines the order in which the information will be displayed. In this embodiment, the choices are date/time (ascending or descending), or transaction type/date/time (ascending or descending). Refund button 2523 is used to perform a direct dial call refund.
Date column 2605 displays the date of each financial transaction. Time column 2607 displays the hour at which each financial transaction occurred. Dialed digits column 2609 indicates the sequence of digits used during each phone call. Duration column 2611 displays the time in minutes of each phone call and charge column 2613 indicates the resulting cost of the call. Charge type column 2615 indicates the calling method used for each telephone call (e.g., direct call, collect call, international, etc.). Call result column 2617 displays the appropriate completion code for each call. A completion code is chosen from a pre-assigned code list indicating possible outcomes of telephone calls. Recorder column 2619 displays the recorder used for each call. Alter type column 2621 indicates the type of alert triggered, if any.
Displayed in
Shown in
Call access control block 2903 contains collect column 2911, direct call column 2913, and total column 2915. Corresponding to these columns are a number of rows. At the intersection of the rows and columns a number of fields are formed. Calls/Day row 2905 is used to enter the number of collect, debit, and total calls allowed per day. Calls/Week row 2907 is used to enter the number of calls allowed per week. Calls/Month row 2909 is used to enter the number of calls allowed per month. A minutes between calls row 2919 indicates the amount of time that must lapse between calls before another can be made. Minutes per call row 2917 is used to enter the maximum duration allowed for each telephone call.
Call schedule block 2921 determines when the inmate may use the telephone system. For each day of the week, selected from the day drop down list, there is a checkbox for each half hour period of time. If checked, calls are allowed to begin during that half-hour. For instance, if 9:30 PM is checked, calls can begin anytime from 9:30-9:59:59 PM.
ITS access control settings block 2923 determine how many times inmates may perform an inquiry for each day, week, or month.
Maximum active telephone numbers field 2925 limits the number of telephone numbers on the inmate's approved list. Commissary controls button 2927 switches screens to allow control of when and how often inmates may access the commissary system and transfer funds.
A number of check box fields are also located on this screen, including require approved number list (collect) field 2929, require approved number list (direct) box 2931, allow special number list 2933, use call branding message 2935, and use intermittent message 2937.
Branding allows a pre-recorded message to be played for the called party at the beginning of every call to announce, “This call is from a Colorado Correctional Facility.” This message can be set for the individual inmate, a living unit, or all inmates at a facility. Call branding may be turned on or off at the discretion of the prison staff.
Much like call branding, intermittent messages play the pre-recorded message “This call is from a Colorado Correctional Facility” throughout the call. How often the message plays is randomly determined, within a set minimum and maximum duration between plays.
The Shadow software provides the capability to digitally record, store, playback and execute a possible keyword search. Shadow resides on a separate site server providing flexibility in implementation and sizing of the system. It can simultaneously record conversations from all telephones installed at a site regardless of the size of the site. As an integrated part of the system, the Shadow software is completely transparent to the user. An intuitive user interface is provided for playback of the conversation.
The call records screen, (
Play button 3107, pause button 3109, and stop button 3111 operate in the typical fashion. Step button 3113 permits users to “fast forward” in the recorded conversation to any point. Playback is automatically resumed when the step button is released. Back button 3115 is used in a similar manner. Mark loop begin button 3103 marks the beginning of a selected segment of the recorded conversation for repeated playback. This feature is extremely useful when a segment of the recorded conversation is not readily understood. Mark loop end button 3105 marks the end of a selected segment of the recorded conversation for repeated playback. This button is grayed out until mark loop begin button 3103 is activated.
Play loop button 3117 is used to replay the portion of the conversation from the mark loop begin to the mark loop end positions. Repeated playback continues until stop button 3111 is depressed. The button is grayed out until mark loop begin 3103 button is activated. Volume bar 3119 permits users to adjust the volume to a desired level. Close button 3121 closes the shadow button and returns the user to the call detail screen (as shown in
The Shadow software also incorporates a text scan, which can be used to find a particular call for audio review. A site to search is selected from a drop down list accessed from the main screen of the software. This brings up a text scanner selection screen shown in
After the desired criteria have been chosen from this list, the software executes the text scan and query result screen will appear, as shown in
The call system allows for quick and easy tabulation of data. Every report contains at a minimum general report capabilities. Reports may either be generated automatically at certain times or under certain conditions, or printed upon demand by prison staff members with appropriate user level access. Reports may be generated for one facility or a combination of facilities, depending upon the requesting user's security level.
All reports contain, at a minimum, time, terminal making request, parameters of the report, number of pages, report heading, end of report footer, report heading on each page, report title on each page, identified of the staff member creating the report. In the present example, the header on each page includes the prison name, report name, data and time of the report, page number, and field headings. The footer, contained at the end of the report, contains total for all columns containing dollar values, total count of inmates if the report contains inmate information, and total call counts or total call duration if report includes this information.
Various reports have different selection criteria and parameters to be defined before running the report. Many will prompt the user for a start date/time and an end date/time and an end date/time. Others require the user to select individual or multiple records at a time. Some reports utilize ‘wizards’, such as the telephone wizard, that allow the use of pre-saved selections. All reports can also be saved for later use.
An example report is shown in
Separate software controls the biometric and RF authentication. The biometric portion of the software will be contained first.
As discussed with respect to
The biometric scanner may be a retinal scanner, fingerprint scanner, body heat sensor, or any other like device. Such scanners typically include means of digitizing the information so it is readily available to the DIS.
As discussed with respect to
The called party may also be voice authenticated each time a user places a call (e.g., a record of voice prints for the called family members may be established). This provides additional security because it ensures that the user is not communicating with an unauthorized party. The biometric information provided by the called party can also be used for third party call detection. By continuously sampling voice data from the telephone conversation, the biometric software can be used to detect if a third party or an unauthorized person has spoken. Upon third party detection, the call can be disconnected and/or authorities may be notified. This will cause an alert in the inmate's profile.
The biometric authentication software can easily be extended for use with telephone cards. Upon purchasing a telephone card with limited or unlimited debit, the user provides voice initial voice data for future voice authentication. This may occur in a number of ways. For example, the user may be required to provide voice information at the institution that the telephone debit card was purchased, such as a supermarket or convenience store. The voice information may also be provided the first time that a user calls the access number for the debit card. A voice prompt asks the user to state and repeat their name. Additionally, the telephone number from which the user is calling from may be asked for and entered via DTMF tones or recording based on the ANI data. The telephone system may then hang up the line and call back the number provided by the user and ask for voice authentication utilizing the information previously provided. Upon authentication, the user does not have to repeat the initial voice supplication. If a user attempts to use the purchased telephone debit account again, the user only has to supply voice authentication and/or an account or PIN. In this way, the user does not have to carry the telephone card to access the telephone system and need only remember the telephone access number. This authentication process can additionally be used with prepaid cellular telephones.
A similar authentication process may also be used to access an internet telephone account, or any other secure internet information. The person simply uses the microphone attached to their computer to provide the voice authentication instead of the telephone handset. This information can be transmitted via the internet to the server containing the stored biometric data and be used to voice authenticate the user.
The voice database concept may be extended for use on a PSTN. Each user of the public telephone network would only be granted access if the user's name is in the database. The voice database may also be used to limit telephone access. For example, a convicted criminal would be blocked from ever calling his previous victims. Teenagers may also be blocked from accessing “1-900” numbers and the like.
A national voice database may also be used to track wanted criminals. For example, if a wanted criminal or suspect ordered a pizza from a local pizzeria, biometric software, located on the same server as the voice database, could recognize the calling party as the wanted criminal. Authorities would then be alerted that the fugitive has been identified. Since the called number is known, authorities already have a means of narrowing the search field for the criminal. Authorities may then call the pizzeria and inquire about previously placed orders. This would provide them with a list of possible locations that the criminal may be located. Furthermore, if voice recognition is also in the capability of the biometric software, the food order of the calling party may be ascertained. Authorities could then inquire who placed that specific order. Speech recognition capability may additionally allow the address of the calling party to be located without ever having to call the pizzeria because a delivery address or telephone callback number is typically provided when placing a food order. Once the suspect is located, the suspect can easily be apprehended. This technology may also be used to help apprehend criminals who steal any device interfaced with a telephone network, public or private. Such devices include, but are not limited to, cellular telephones, wireless internet appliances, and laptops. Whenever the criminal attempts to use the stolen device, a voice print is automatically recognized by the voice database as not being an authorized user of that device. The proper authorities could then be alerted.
The RF authentication portion of the software operates in a similar fashion. When a user picks up the telephone, the RF emitter is activated. The signal is reflected by a band the user wears and the reflected signal is received by the RF sensor. Using this information, the RF software determines the distance of the inmate from the telephone and the user's uniquely encoded RF frequency. This is done using complex algorithms known to one skilled in this particular art. If the user is not a valid user of the call system, the prison authorities are notified at which facility and station ID the violation occurred. If the user is a valid user, the supplied PIN and/or biometric information are also authenticated. The user is allowed access to the system only if the information supplied is valid and agrees with the other authentication method utilized.
Referring next to
Multiple processors may also be incorporated into the architecture. This allows call processing even after component failure. The architecture also provides for a sharing of the load between processors, which eliminates system overload during extremely busy periods.
Additionally, electronic switchboard device 105 performs the voice prompts heard by the user and the recipient of the call allowing the parties to respond to the menu selections. Electronic switchboard device 105 also tests outgoing trunk lines as calls are placed and digitizes telephone audio for recording and/or biometric voice identification. If no dial tone is present, trunk line 111 is taken out of service for a pre-programmed amount of time for maintenance. These capabilities are pre-programmed into the device's firmware. However, it is foreseeable that software and software upgrades may provide these services in addition to other services useful in the present invention.
A central site server 113 interfaces within call management system 101 via first serial port 115. In the preferred embodiment of the present invention, an RS-232 serial port is employed for the interference connection. However, it is foreseeable that other types of serial ports 115 commonly known in the art may be utilized. Serial port 115 may also be comprised of a direct hardware connection or may consist of a series of ports and connecting means commonly known in the art for connecting electronic devices. Serial port 115 is designed to allow firmware driven systems, such as electronic switchboard device 105, to interface with software-based systems, such as a PC designed system operating as a site server. All user and telephone call information is routed through central site server 113. At central site server 113, user call information is digitized for efficient data transfer and efficient record keeping. Central site server 113 stores at a minimum each user's financial transaction data. It is preferred that central site server 113 also stores the digitized audio used for voice prompts as well as each user's call restrictions, PIN, biometric verification data, etc. However, depending on the memory requirements, numerous site servers may be employed. It is foreseeable that older archived data may also be stored on an integral or a remote computer system database (not shown) or kept on additional storage devices on central site server 113.
Connected to central site server 113 via a second serial port 115 is audio recorder 117. In the preferred embodiment of the present invention, an RS-232 serial port is employed for the interference connection. However, it is foreseeable that other types of serial ports 115 commonly known in the art may be utilized. Serial port 115 may also be comprised of a direct hardware connection or may consist of a series of ports and connecting means commonly known in the art for connecting electronic devices. Audio recorder 117 may either be a stand-alone unit or incorporated into the hardware of central site server 113. Although it is preferred in the present embodiment that audio recorder 117 is digital, it is foreseeable that other known types of recording devices, as well as those not yet contemplated, may be employed in accordance with the teachings of the present invention. Audio recorder 117 records the conversations performed under the direction of telephone call management system 101. Audio recorder 117 is activated for each call unless the number being called is specifically flagged for no recording or monitoring, such as calls to or from an attorney. Furthermore, audio recorder 117 can monitor multiple telephone lines simultaneously, using a different recorder channel number for trunk lines 111. The recorder channel number further enables the site's staff to identify the call record they wish to review associated with a desired outgoing telephone call. Each user telephone 102 is further associated with a station identification number. The station identification number allows the staff of the institution to identify the particular user telephone 102 a particular call was initiated and conducted from. It is foreseeable that the embodiment described herein supports up to 32 inmate telephone stations 103 and 24 trunk lines 111. However, multiple units 105 may be configured to support up to xxx telephone stations and xxx trunk lines.
Central site server 113 is controlled by software associated with administrative workstation 120. In the preferred embodiment, the administrative workstation 120 is connected to central site server 113 via a local area network (LAN). However, it is foreseeable that other types of electronic connections may be employed. The administrative workstation's 120 software can modify call restrictions for individual users in addition to all telecommunication activity of the institution. Additionally, according to the present example, the software also tracks the inmate's commissary information, such as the account balance if a debit system is being used. Furthermore, depending on the needs of an institution, the database may perform other functions.
A commissary workstation 121 is used in conjunction with the administrative workstation 120 to manage and record inmates' financial transactions. In the preferred embodiment, commissary workstation 121 and administrative workstation 120 are connected to a central site server 113 via a local area network. However, other known connections, or connections not yet contemplated may be utilized. Commissary workstation 121 can also record other financial information, such as the amount spent on collect calls by each inmate, amount spent on debit calls, the total net financial transactions for each user, etc.
Shadow workstation 123 and investigative workstation 125 are also employed in the present embodiment. Shadow workstation 123 and investigative workstation 125 are created via a local area network linked central site server 113 in the present embodiment. Shadow workstation 123 is used by a live operator to monitor telephone calls without detection. It is foreseeable that this function is performed by software integrated with shadow workstation 123. Shadow workstation 123 software provides a means for patching into a call using circuitry without alerting the inmate or called party as to the operator's presence. If the operator finds that a call being monitored is suspicious, the operator may activate the audio recorder 117 to record a portion of an active telephone call. The called party's number may also be flagged in the inmate's profile stored on administrative workstation 120 or central site server 113.
Alternatively, software located on central site server 113 or investigative workstation 125 may be used to passively monitor calls. For example, when certain key phrases are spoken, voice recognition software activates audio recorder 117 via electronic means and alerts the proper authorities that a violation has occurred.
Furthermore, investigative workstation 125 controls other monitoring and security features interfaced in the call system. For example, investigative workstation 125 can be used to access past conversations stored on audio recorder 117. Software on investigative workstation 125 may also be configured to detect if a third party is present during an inmate's conversation. Investigative workstation 125 or central site server 113 may also contain voice recognition software to aid in calling or called party voice authentication. The administrative workstation 120, shadow workstation 123, investigative workstation 125, and commissary workstation 121 may alternatively be combined into one unit. Furthermore, 120, 123, and 120 may be integral within the central site server. It is also foreseeable that 120 may be alternately located off site from the other apparati of the present invention.
While the present invention has been described with reference to the preferred embodiments and several alternative embodiments, which embodiments have been set forth in considerable detail for the purposes of making a complete disclosure of the invention, such embodiments are merely exemplary and are not intended to be limiting or represent an exhaustive enumeration of all aspects of the invention. The scope of the invention, therefore, shall be defined solely by the following claims. Further, it will be apparent to those of skill in the art that numerous changes may be made in such details without departing from the spirit and the principles of the invention. It should be appreciated that the present invention is capable of being embodied in other forms without departing from its essential characteristics.
Claims
1. A system comprising:
- a database configured to: store biometric data of a plurality of users; store first identifying information associated with the plurality of users in correspondence with the biometric data; and
- a control platform coupled to the database, the control platform configured to: prior to a call: receive second biometric data associated with a calling party; receive second identifying information related to the calling party; find in the database third identifying information from among the first identifying information that matches the second identifying information received from the calling party; retrieve from the database third biometric data associated with the third identifying information; compare the second biometric data received from the calling party with the third biometric data retrieved from the database; and authenticate the calling party based on the comparison; and during the call: acquire and store biometric information of the called party; acquire, at regular intervals, biometric information of a person speaking; compare the acquired biometric information of the person speaking to at least one of the stored third biometric information and the stored biometric information of the called party; and detect the presence of an unauthorized call participant based on the comparison.
2. The system of claim 1, wherein the control platform is further configured to:
- receive further information related to the calling party;
- store the further information in the database in correspondence with the biometric data of the calling party.
3. The system of claim 2, wherein the further information includes at least one of an identification number, a name of the user, a phone number, a phone number of a called party, and a language preference.
4. The system of claim 1, wherein the biometric data calling party includes at least one of a voiceprint, a face architecture, a signature architecture, a fingerprint, a retinal print, a hand geometry, and an infrared pattern of the face.
5. The system of claim 1, wherein the identifying information includes a personal identification number.
6. The system of claim 1, wherein the identifying information includes a name of the user.
7. The system of claim 1, wherein the further information includes at least one of an identification number, a name of the user, a phone number, a phone number of a called party, and a language preference.
8. The system of claim 1, wherein the acquired biometric data of the called party includes at least one of a voiceprint, a face architecture, a signature architecture, a fingerprint, a retinal print, a hand geometry, and an infrared pattern of the face.
9. The system of claim 1, wherein the first biometric data of the calling party is received during a call, and the second biometric data is received during the same call.
10. The system of claim 1, wherein the biometric data calling party is received during a registration process that is not a part of a call, and the second biometric data is received during an authentication step of a call.
11. A method comprising:
- receiving biometric data of a plurality of users;
- storing the biometric data in a database;
- receiving first identifying information of the plurality of users;
- storing the first identifying information in the database in correspondence with the biometric data of the plurality of users;
- prior to a call: receiving second biometric data related to a calling party; receiving second identifying information related to the calling party; finding in the database third identifying information from among the first identifying information that substantially matches the second identifying information received from the calling party; retrieving from the database third biometric data associated with the third identifying information; comparing the second biometric data received from the calling party with the third biometric data retrieved from the database; and authenticating the calling party based on the comparing; and during the call: acquiring and storing biometric information of the called party; acquiring, at regular intervals, biometric information of a person speaking; comparing the acquired biometric information of the person speaking to at least one of the stored third biometric information and the stored biometric information of the called party; and detecting the presence of an unauthorized call participant based on the comparison.
12. The method of claim 11, further comprising:
- receiving further information related to the calling party;
- storing the further information in the database in correspondence with the biometric data of the calling party.
13. The method of claim 12, wherein the further information includes at least one of an identification number, a name of the user, a phone number, a phone number of a called party, and a language preference.
14. The method of claim 11, wherein the stored biometric data of the calling party includes at least one of a voiceprint, a face architecture, a signature architecture, a fingerprint, a retinal print, a hand geometry, and an infrared pattern of the face.
15. The method of claim 11, wherein the identifying information includes a personal identification number.
16. The method of claim 11, wherein the identifying information includes a name of the calling party.
17. The method of claim 11, wherein the further information includes at least one of an identification number, a name of the user, a phone number, a phone number of a called party, and a language preference.
18. The method of claim 11, wherein the acquired biometric data of the called party includes at least one of a voiceprint, a face architecture, a signature architecture, a fingerprint, a retinal print, a hand geometry, and an infrared pattern of the face.
19. The method of claim 11, wherein the first biometric data of the calling party is received during a call, and the second biometric data is received during the same call.
20. The method of claim 11, wherein the biometric data of the calling party is received during a registration process that is not a part of a call, and the second biometric data is received during an authentication step of a call.
Type: Application
Filed: Sep 16, 2016
Publication Date: Jan 5, 2017
Applicant: Globel Tel*Link Corp. (Reston, VA)
Inventor: Stephen Lee HODGE (Aubry, TX)
Application Number: 15/268,142
