Abstract: A method and apparatus for facilitating communications with a roaming mobile subscriber unit which roams beyond a microcellular communication network, such as a Digital Electronic Cordless Telephone (DECT) system. A mobility server of the microcellular communication networks is coupled to a macrocellular communication network, such as a Global System for Mobile communications (GSM) network. Wide-area mobility management functions of the macrocellular communication network are provided to the microcellular communication network and are used to facilitate call routing to and from the roaming, mobile subscriber unit.

Abstract: In order to provide a practical E-mail reader for reading out E-mails phonetically enabling easy grasping of their contents by a user with its vocal output even when quotation codes or header information are included in the E-mails, a phonetic E-mail reader of the invention comprises a speech synthesizer (102) for converting text data into vocal data, quotation code storing means (105) for storing quotation codes used for indicating a quotation line inserted at a top of the quotation line, and quotation code elimination means (106) for detecting and eliminating a quotation code inserted at tops of quotation lines referring to the quotation code storing means (105) before supplying the quotation lines to the speech synthesizer (102).

Abstract: A method and apparatus for discriminating non-sounds and voiceless sounds of speech signals, recorded on a recording medium, from each other when playing back the speech signals at a varied play-back speed. The method includes the steps of setting, as a reference voltage level, an optional value between a voltage level corresponding to non-sounds and a voltage level corresponding to voiceless sounds, detecting a pitch component of each waveform of the speech signals, comparing the absolute value of a voltage level of the detected pitch component with the reference voltage level, and distinguishing and outputting a portion of the speech signal associated with the detected pitch component based on the result of the comparison.

Abstract: A mandarin speech input method for directly translating arbitrary sentences of mandarin speech into corresponding Chinese Characters. The present invention is capable of processing a sequence of "mono-syllables," "(but each of the characters in the poly-character word is continuous)," "prosodic segments," or even a "whole sentence of continuous mandarin speech." A prosodic segment comprising one or more words is a segment that is automatically isolated by a speaker by pausing where characters in the prosodic segment are continuous.

Abstract: A pattern recognition scheme using probabilistic models that are capable of reducing a calculation cost for the output probability while improving a recognition performance even when a number of mixture component distributions of respective states is small, by arranging distributions with low calculation cost and high expressive power as the mixture component distribution. In this pattern recognition scheme, a probability of each probabilistic model expressing features of each recognition category with respect to each input feature vector derived from each input signal is calculated, where the probabilistic model represents a feature parameter subspace in which feature vectors of each recognition category exist and the feature parameter subspace is expressed by using mixture distributions of one-dimensional discrete distributions with arbitrary distribution shapes which are arranged in respective dimensions.

Abstract: With the signal encoding method and apparatus according to the present invention, noise components of plural channels are encoded individually by a first encoding unit 124, while noise components of plural channels are encoded in common by a second encoding unit 125. A discriminating unit 123 discriminates characteristics of noise components of plural channels. Based upon the results of discrimination, selective switching is made between an output of the first encoding unit 124 and an output of the second encoding unit 125. If the noise components of plural channels are encoded in common, the compression ratio for the noise components of plural channels may be improved. On the other hand, if the noise components of plural channels are not encoded in common, ill effects due to common handling can be prohibited.

Abstract: A method and apparatus for recognizing an identifier entered by a user. A caller enters a predetermined identifier through a telephone handset. A signal representing the entered identifier is transmitted to a remote recognizer, which responds to the signal by producing a recognized output intended to match the entered identifier. The present invention compares this recognized identifier with a list of valid reference identifiers to determine which one of these reference identifiers most likely matches the entered identifier. In performing this determination, the present invention employs a confusion matrix, which is an arrangement of probabilities that indicate the likelihood that a given character in a particular character position of the reference identifier would be recognized by the recognizer as a character in the corresponding character position of the recognized identifier.

Abstract: Transmission level control is provided to eliminate a sense of disorder during a television conference connecting a multiplicity of locations.In a speech input terminal included in connection between multiple locations, there is provided a speech input means for inputting speech, a communication control means for controlling communication between the speech input means and a communication line, a transmission volume adjusting means capable of stepless adjustment of transmission volume, and a reception volume adjusting means. Thus, the volume adjusting means which has conventionally been provided only at a receiving end is now provided also at a transmitting end to allow stepless volume control, which enables speech output in conformity with the speech level of another entity's terminal.

Abstract: A speech recognition system is disclosed useful in, for example, hands-free voice telephone dialing applications. The system will match a spoken word (token) to one previously enrolled in the system. The system will thereafter synthesize or replay the recognized word so that the speaker can confirm that the recognized word is indeed the correct word before further action is taken. In the case of voice activated dialing, this avoids wrong numbers. The token itself is not explicitly recorded; rather, only the lefemes may be recorded from which the token can be reconstructed for playback. This greatly reduces the amount of disk space that is needed for the database as well as provides the ability to reconstruction data in real time for synthesis use by a local name recognition machine.

Abstract: A Telecom Access Cost Management System provides the capability for a communication carrier service provider to substantially automate the payment to other communication carrier service providers for the use of their services and equipment. Billed charges are received in a variety of forms from the communication carrier service providers. The cost processor receives this information, checks its integrity, and converts it to a format in which it can be further processed. Once this information has been uploaded and converted, a validation process is performed in which the individual items of the bill are checked as to whether the rate information charged by the communication carrier service providers matches the rates which have been either negotiated or established by a third party. Any discrepancies noted in this comparison process are included in a report that is included with the item in the billed charges.

Abstract: The present invention relates to methods and devices for processing data frames representative of audio information in digitized and compressed form. The method comprises the steps of classifying successive data frames into frames containing speech sounds and non-speech sounds, altering parameters of the data frames identified as containing non-speech sounds for eliminating or at least substantially reducing artifacts that distort the acoustic background noise. In addition, the data frame identified as containing non-speech sounds are low-pass filtered. Finally, a signal level compensation is effected to avoid undesired fluctuations in the signal level.

Abstract: A code printing apparatus includes a speech input unit for inputting speech information, a code output unit for printing the speech information input by the speech input unit on a medium as an optically readable code pattern image, and a speech code printing control unit for controlling the code output unit to print the speech information input by the speech input unit on the medium as a corresponding code pattern image on the basis of the input operation at an operation unit controlled by the user. The user can thus easily achieve a function of printing speech data or the like as an optically readable code.

Abstract: Apparatus for processing acoustic features extracted from a sample of speech data forming a feature vector signal every frame period includes a first linear prediction analyzer, a vector quantizer, at least one partitioned vector quantizer and a scalar quantizer. The first linear prediction analyzer performs a linear prediction analysis on the feature vector signal to generate a first error vector signal. Next, the vector quantizer performs a vector quantization on the first error signal thereby generating a first index corresponding to a first prestored vector signal which is an approximation of the first error vector signal. The vector quantizer also generates a residual vector signal which is the difference between the first error vector signal and the first prestored approximation vector signal.

Abstract: A microphone array input type speech recognition scheme capable of realizing a high precision sound source position or direction estimation by a small amount of calculations, and thereby realizing a high precision speech recognition. A band-pass waveform, which is a waveform for each frequency bandwidth, is obtained from input signals of the microphone array, and a band-pass power of the sound source is directly obtained from the band-pass waveform. Then, the obtained band-pass power is used as the speech parameter. It is also possible to realize the sound source estimation and the band-pass power estimation at high precision while further reducing an amount of calculations, by utilizing a sound source position search processing in which a low resolution position estimation and a high resolution position estimation are combined.

Abstract: A first mobile station is correlated with a second mobile station as a permitted home zone party. The second mobile station is in speech connection with another telecommunications party. In the event that the first mobile station wishes to join in on the conversation, the first mobile station simply transmits a selected service code A to the serving mobile switching center (MSC). The serving MSC retrieves the data correlating the first mobile station with the second mobile station, determines that the first mobile station is currently located within the same home zone, and conferences the first mobile station into the existing call connection involving the second mobile station by utilizing one of the call conference circuits. As an illustration, members of the same household can designate and correlate each other as permitted home zone parties.

Abstract: An automatic gain control (AGC) algorithm is provided to the digital signal processor instructions of an audio processing unit of a multimedia multipoint server. The AGC algorithm operates to bring the power level of the audio signal of every active channel to within a fixed range. The AGC algorithm allows the use of non-calibrated microphones, corrects for long distance signal attenuation, and provides improved audio level reference for audio switching and audio level video switching. The AGC algorithm is structured into two steps: calibration and gain control. The calibration includes defining a noise threshold. Gain control includes first calculating a dynamic speech detection level. Second, with reference to the speech detection level, determining whether an incoming audio signal is a speech-based audio signal.

Abstract: A method for detecting fraud in a cellular radio telephone system. Fraud is suspected when the system detects a multiple access from a mobile station, when an activity collision occurs, when the system receives a premature registration from the mobile station, when auditing or operator-initiated locating of the mobile station reveals the existence of the mobile station in two locations simultaneously, or when tracing of mobile subscriber activity reveals unusual activity.

Abstract: An information processing system (10) receives input data from a keyboard (26), a mouse (28) and spoken data input device (30) in an input data sequence. A microphone (32) converts received speech into electrical signals which are digitized by an analogue to digital converter (100). A digital signal processor (102) converts the digital signals into multi-dimensional vectors which are stored in a temporary input buffer (104). A recognition processor (106) performs a recognition program in order to match the multi-dimensional vectors to speech models. In order that the system (10) may output data in an output data sequence which corresponds to the input data sequence, each data input receives a time stamp. A timing controller (24) ensures that instructions received from either the keyboard or the mouse are output only when those instructions have a time stamp which is earlier than the time stamp of the data most recently processed by the recognition processor.

Abstract: Hitherto, when speech data is quantized, since a quantum is determined irrespective of a deviation of a distribution of the speech data, an error of an output probability calculation by quantization is large and a recognition rate deteriorates. According to the invention, to solve the subject of the prior art, a quantization range of speech data is obtained, an integral value of an output probability of the quantization range is obtained by using an output probability distribution of statistic models of speech data which has previously been obtained, a quantum is determined so as to equally divide the integral value, and a quantization code book is formed, thereby reducing an error of an output probability due to the quantization and enabling a recognition rate to be improved.

Abstract: A method and system for providing an inverse transform for an audio compression decoding algorithm in software precalculates a plurality of identified values; each of which is computationally intensive. The method and system then performs a pre-inverse transform complex multiply utilizing a first portion of the identified values and an array of input coefficients to provide a plurality of intermediate values. Thereafter, an inverse transform complex multiply and a post inverse transform multiply are combined to provide a combined complex multiply operation. The combined complex multiply operation uses a second portion of the identified values and the intermediate values provides the inverse transform. Accordingly, through the use of the present invention, the number of instructions for implementing the inverse transform can be substantially minimized.