Abstract: A system is provided for fraud prevention upscaling with a fraudster voice print watchlist. The system includes a processor and a computer readable medium operably coupled thereto, to perform fraud prevention operations which include receiving a first voice print of a user during a voice authentication request, accessing the fraudster voice print watchlist comprising voice print representatives for a plurality of voice print clusters each having one or more of a plurality of voice prints identified as fraudulent for a voice biometric system, determining that one or more of the voice print representatives in the fraudster voice print watchlist meets or exceeds a first biometric threshold for risk detection of the first voice print during the fraud prevention operations, and determining whether the first voice print matches a first one of the plurality of voice print clusters.

Abstract: Techniques for determining whether audio is machine-outputted or non-machine-outputted are described. A device may receive audio, may process the audio to determine audio data including audio features corresponding to the audio, and may process the audio data to determine audio embedding data. The device may process the audio embedding data to determine whether the audio is machine-outputted or non-machine-outputted. In response to determining that the audio is machine-outputted, then the audio may be discarded or not processed further. Alternatively, in response to determining that the audio is non-machine-outputted (e.g., live speech from a user), then the audio may be processed further (e.g., using ASR processing).

Abstract: A feature vector having high class identification capability is generated. A signal processing system provided with: a first generation unit for generating a first feature vector on the basis of one of time-series voice data, meteorological data, sensor data, and text data, or on the basis of a feature quantity of one of these; a weight calculation unit for calculating a weight for the first feature vector; a statistical amount calculation unit for calculating a weighted average vector and a weighted high-order statistical vector of second or higher order using the first feature vector and the weight; and a second generation unit for generating a second feature vector using the weighted high-order statistical vector.

Abstract: An electronic device is provided. The electronic device includes a microphone to receive audio, a communicator, a memory configured to store computer-executable instructions, and a processor configured to execute the computer-executable instructions. The processor is configured to determine whether the received audio includes a predetermined trigger word; based on determining that the predetermined trigger word is included in the received audio; activate a speech recognition function of the electronic device; detect a movement of a user while the speech recognition function is activated; and based on detecting the movement of the user, transmit a control signal, to a second electronic device to activate a speech recognition function of the second electronic device.

Abstract: A spatial normalization unit generates a normalized spectrum by normalizing an orientation component of a microphone array for a target direction included in a spectrum of an acoustic signal acquired from each of a plurality of microphones forming the microphone array into an orientation component for a predetermined standard direction. A mask function estimating unit determines a mask function used for extracting a component of a target sound source arriving in the target direction on the basis of the normalized spectrum using a machine learning model. A mask processing unit estimates the component of the target sound source installed in the target direction by applying the mask function to the acoustic signal.

Abstract: A speech recognition method, a speech recognition apparatus, and a method and an apparatus for training a speech recognition model are provided. The speech recognition method includes: recognizing a target word speech from a hybrid speech, and obtaining, as an anchor extraction feature of a target speech, an anchor extraction feature of the target word speech based on the target word speech; obtaining a mask of the target speech according to the anchor extraction feature of the target speech; and recognizing the target speech according to the mask of the target speech.

Abstract: An electronic device is provided. The electronic device includes a microphone to receive audio, a communicator, a memory configured to store computer-executable instructions, and a processor configured to execute the computer-executable instructions. The processor is configured to determine whether the received audio includes a predetermined trigger word; based on determining that the predetermined trigger word is included in the received audio; activate a speech recognition function of the electronic device; detect a movement of a user while the speech recognition function is activated; and based on detecting the movement of the user, transmit a control signal, to a second electronic device to activate a speech recognition function of the second electronic device.

Abstract: A sound analysis system includes: first and second sound pressure acquisition means for respectively acquiring a sound pressure of a voice of a user and disposed in an equipment worn by the user at positions that differ in a distance from a mouth of the user under the state in which the user is wearing the equipment; distance estimation means for estimating a distance between either the first or the second sound pressure acquisition means and the mouth of the user based on the acquired sound pressures; and sound pressure correction means for calculating a difference between a reference value of the distance between the first or the second sound pressure acquisition means and the mouth of the user and the estimated distance, and correcting at least one of the sound pressures acquired by the first and the second sound pressure acquisition means based on the calculated difference.

Abstract: A voice recognition device includes a plurality of mics disposed toward different directions and a processor connected with the plurality of mics, wherein the processor is configured to determine, in a setup mode, a direction of a first sound received through the plurality of mics; set a non-detecting zone, which includes the direction of the first sound; determine, in a normal mode, a direction of a second sound received through the plurality of mics; and skip voice recognition for the second sound or an operation based on the voice recognition depending on whether the direction of the second sound belongs to the non-detecting zone.

Abstract: A method implemented by a computing system comprises generating, by the computing system, a fingerprint comprising a plurality of bin samples associated with audio content. Each bin sample is specified within a frame of the fingerprint and is associated with one of a plurality of non-overlapping frequency ranges and a value indicative of a magnitude of energy associated with a corresponding frequency range. The computing system removes, from the fingerprint, a plurality of bin samples associated with a frequency sweep in the audio content.

Abstract: In order to detect a replay attack in a speaker recognition system, at least one feature is identified in a detected magnetic field. It is then determined whether the at least one identified feature of the detected magnetic field is indicative of playback of speech through a loudspeaker. If so, it is determined that a replay attack may have taken place.

Abstract: According to some aspects, a method of monitoring an acoustic environment of a mobile device, at least one computer readable medium encoded with instructions that, when executed, perform such a method and/or a mobile device configured to perform such a method is provided. The method comprises receiving acoustic input from the environment of the mobile device while the mobile device is operating in the low power mode, detecting whether the acoustic input includes a voice command based on performing a plurality of processing stages on the acoustic input, wherein at least one of the plurality of processing stages is performed while the mobile device is operating in the low power mode, and using at least one contextual cue to assist in detecting whether the acoustic input includes a voice command.

Abstract: A system and method for storing data samples in discrete poses and recalling the stored data samples for updating a sound filter. The system determines that a microphone array at a first time period is in a first discrete pose of a plurality of discrete poses, wherein the plurality of discrete poses discretizes a pose space. The pose space includes at least an orientation component and may further include a translation component. The system retrieves one or more historical data samples associated with the first discrete pose, generated from sound captured by the microphone array before the first time period, and stored in a memory cache (e.g., for memorization). The system updates a sound filter for the first discrete pose using the retrieved one or more historical data samples. The system generates and presents audio content using the updated sound filter.

Abstract: A transcription of a query for content discovery is generated, and a context of the query is identified, as well as a first plurality of candidate entities to which the query refers. A search is performed based on the context of the query and the first plurality of candidate entities, and results are generated for output. A transcription of a second voice query is then generated, and it is determined whether the second transcription includes a trigger term indicating a corrective query. If so, the context of the first query is retrieved. A second term of the second query similar to a term of the first query is identified, and a second plurality of candidate entities to which the second term refers is determined. A second search is performed based on the second plurality of candidates and the context, and new search results are generated for output.

Abstract: An electronic device is provided. The electronic device includes a microphone to receive audio, a communicator, a memory configured to store computer-executable instructions, and a processor configured to execute the computer-executable instructions. The processor is configured to determine whether the received audio includes a predetermined trigger word; based on determining that the predetermined trigger word is included in the received audio; activate a speech recognition function of the electronic device; detect a movement of a user while the speech recognition function is activated; and based on detecting the movement of the user, transmit a control signal, to a second electronic device to activate a speech recognition function of the second electronic device.

Abstract: System and methods for processing audio signals are disclosed. In one implementation, a system may include a wearable apparatus including an image sensor to capture images from an environment of a user; an audio sensor to capture an audio signal from the environment of the user; and at least one processor. The processor may be programmed to receive the audio signal captured by the audio sensor; identify at least one segment including speech in the audio signal; receive an image including a representation of a code; analyze the code to determine whether the code is associated with the user and/or the wearable apparatus; and after determining that the code is associated with the user and/or the wearable apparatus, transmit at least one segment of the audio signal, at least one image of the plurality of images, and/or other information to a computing platform.

Abstract: The present disclosure provides an electronic device, a control method therefor, and a control program therefor capable of preventing an operation for activating a function protected by user authentication from becoming complicated. An electronic device includes: a keyword management DB for storing identification information of a registrant and a registered keyword in association with each other; a command management DB for storing a command and required authentication scores in association with each other; a data generator for creating grammar data including a registered keyword and a command; an utterance recognizer for matching the grammar data and extracted data extracted from an utterance of a user and acquiring a recognized authentication score and a recognized command; and an authenticator for determining that the command is recognized by comparing the required authentication score associated with the command determined to be the same as the recognized command and the recognized authentication score.

Abstract: [Object] To make it possible to estimate the position of a sound source resulting from a wearing displacement, even in the case where the wearing displacement occurs in a wearable device.

Abstract: A computer-implemented method includes listening for audio name information indicative of a name of a computer, with the computer configured to listen for the audio name information in a first power mode that promotes a conservation of power; detecting the audio name information indicative of the name of the computer; after detection of the audio name information, switching to a second power mode that promotes a performance of speech recognition; receiving audio command information; and performing speech recognition on the audio command information.

Abstract: Methods and systems populate a speech signature database with unique speech signatures that are associated with one or more speaker identities and are further associated with one or more mobile stations and/or telephone numbers. Real-time voice signals are compared to the speech signatures in the speech signature database. When a match is found, the mobile station from which the voice signal originated is located in real-time. Further, the associations in the speech signature database are leveraged to find other relevant mobile stations or users and to generate additional associations and to also locate associated users and mobile stations.

Abstract: A method is provided and includes estimating an approximate list of potential speakers in a file from one or more applications. The file (e.g., an audio file, video file, or any suitable combination thereof) includes a recording of a plurality of speakers. The method also includes segmenting the file according to the approximate list of potential speakers such that each segment corresponds to at least one speaker; and recognizing particular speakers in the file based on the approximate list of potential speakers.

Abstract: Disclosed embodiments provide for personalizing a voice user interface of a remote multi-user service. A voice user interface for the remote multi-user service can be provided and voice information from an identified user can be received at the multi-user service through the voice user interface. A language model specific to the identified user can be retrieved that models one or more language elements. The retrieved language model can be applied to interpret the received voice information and a response can be generated by the multi-user service in response the interpreted voice information.

Abstract: A system for generating one or more enhanced audio signals such that one or more sound levels corresponding with sounds received from one or more sources of sound within an environment may be dynamically adjusted based on contextual information is described. The one or more enhanced audio signals may be generated by a head-mounted display device (HMD) worn by an end user within the environment and outputted to earphones associated with the HMD such that the end user may listen to the one or more enhanced audio signals in real-time. In some cases, each of the one or more sources of sound may correspond with a priority level. The priority level may be dynamically assigned depending on whether the end user of the HMD is focusing on a particular source of sound or has specified a predetermined level of importance corresponding with the particular source of sound.

Abstract: An embodiment of the invention provides a method of preparing for speaker authentication. The method includes: receiving speech data that represents an utterance made by a user; extracting side information; examining the side information to determine whether to allow speaker model training using the speech data; and generating a feedback message for the user based on the side information if speaker model training using the speech data is not allowed.

Abstract: A method and system for assignee identification of speech includes defining several time intervals and utilizing one or more function evaluations to classify each of the several participants as addressing speech to an automated character or not addressing speech to the automated character during each of the several time intervals. A first function evaluation includes computing values for a predetermined set of features for each of the participants during a particular time interval and assigning a first addressing status to each of the several participants in the particular time interval, based on the values of each of the predetermined sets of features determined during the particular time interval. A second function evaluation may assign a second addressing status to each of the several participants in the particular time interval utilizing results of the first function evaluation for the particular time interval and for one or more additional contiguous time intervals.

Abstract: An electronic device for browsing a document is disclosed. The document being browsed includes a plurality of command-associated text strings. First, a text string selector of the electronic device selects a plurality of candidate text strings from the command-associated text strings. Afterward, an acoustic string provider of the electronic device prepares a candidate acoustic string for each of the candidate text strings. Thereafter, a microphone of the electronic device receives a voice command. Next, a speech recognizer of the electronic device searches the candidate acoustic strings for a target acoustic string that matches the voice command, wherein the target acoustic string corresponds to a target text string of the candidate text strings. Finally, a document browser of the electronic device executes a command associated with the target text string.

Abstract: A voice analyzer includes a plate-shaped body, a plurality of first voice acquisition units that are placed on both surfaces of the plate-shaped body and that acquire a voice of a speaker, a sound pressure comparison unit that compares sound pressure of a voice acquired by the first voice acquisition unit placed on one surface of the plate-shaped body with sound pressure of a voice acquired by the first voice acquisition unit placed on the other surface and determines a larger sound pressure, and a voice signal selection unit that selects information regarding a voice signal which is associated with the larger sound pressure and is determined by the sound pressure comparison unit.

Abstract: The present invention relates to a method and apparatus for generating speaker-specific spoken passwords. One embodiment of a method for generating a spoken password for use by a speaker of interest includes identifying one or more speech features that best distinguish the speaker of interest from a plurality of impostor speakers and incorporating the speech features in the spoken password.

Abstract: A voice analyzer includes an apparatus body, a strap that is connected to the apparatus body and is used to hang the apparatus body from a neck of a user, a first voice acquisition unit provided in the strap or the apparatus body, a second voice acquisition unit provided at a position where a distance of a sound wave propagation path from a mouth of the user is smaller than a distance of a sound wave propagation path from the mouth of the user to the first voice acquisition unit, and an identification unit that identifies a sound, in which first sound pressure acquired by the first voice acquisition unit is larger by a predetermined value or more than second sound pressure acquired by the second voice acquisition unit, on the basis of a result of comparison between the first sound pressure and the second sound pressure.

Abstract: A voice analyzer includes an apparatus body, a strap that is connected to the apparatus body to make the apparatus body hung from a neck of a wearer, a first voice acquisition unit that acquires a voice of a speaker and is disposed in either a left or right strap when viewed from the wearer, a second voice acquisition unit that acquires the voice of the speaker and is disposed in the opposite strap in which the first voice acquisition unit is disposed, and an arrangement recognition unit that recognizes arrangements of the first and second voice acquisition units, when viewed from the wearer, by comparing a voice signal of the voice acquired by the first voice acquisition unit with sound pressure of a heart sound of the wearer acquired by the second voice acquisition unit.

Abstract: Various embodiments of the present invention for validating the authenticity of a website are provided. An example of a method according to the present invention comprises providing a website having an artifact, receiving a communication from a user, at a service provider, for validating the website associated with a service provider, inquiring from the user a description of the artifact comparing the artifact on the website with the description of the artifact from the user and generating a indication to the user based upon the comparing. The communication is over a first communication channel and the website is accessed over a second communication channel. The first communication channel is different than the second. The artifact can be displayed after a user session is identified.

Abstract: In one embodiment, a method includes receiving an acoustic input signal at a speech recognizer. A user is identified that is speaking based on the acoustic input signal. The method then determines speaker-specific information previously stored for the user and a set of responses based on the recognized acoustic input signal and the speaker-specific information for the user. It is determined if the response should be output and the response is outputted if it is determined the response should be output.

Abstract: An audio analysis apparatus includes the following components. A strap has an end portion connected to a main body and is used to hang the main body from a user's neck. A first audio acquisition device is at the end portion or in the main body. Second and third audio acquisition devices are at positions separate from the end portion by substantially the same predetermined distances, on the respective sides of the strap extending from the user's neck. An analysis unit discriminates whether an acquired sound is an uttered voice of the user or another person by comparing audio signals of acquired by the first and second or third audio acquisition devices and detects an orientation of the user's face by comparing the audio signals acquired by the second and third audio acquisition devices. A transmission unit transmits the analysis result to an external apparatus.

Abstract: An audio analysis apparatus includes the following components. A main body includes a discrimination unit and a transmission unit. A strap is used for hanging the main body from a user's neck. A first audio acquisition device is provided to the strap or the main body. A second audio acquisition device is provided to the strap at a position where a distance between the second audio acquisition device and the user's mouth is smaller than the distance between the first audio acquisition device and the user's in a state where the strap is worn around the user's neck. The discrimination unit discriminates whether an acquired sound is an uttered voice of the user or of another person by comparing audio signals of the sound acquired by the first and second audio acquisition devices. The transmission unit transmits information including the discrimination result to an external apparatus.

Abstract: A system for confirming that a subject is the source of spoken audio and the identity of the subject providing the spoken audio is described. The system includes at least one motion sensor operable to capture physical motion of at least one articulator that contributes to the production of speech, at least one acoustic signal sensor to receive acoustic signals, and a processing device comprising a memory and communicatively coupled to the at least one motion sensor and the at least one acoustic signal sensor. The processing device is programmed to correlate physical motion data with acoustical signal data to uniquely characterize the subject for purposes of verifying the subject is the source of the acoustical signal data and the identity of the subject.

Abstract: A voice recognition apparatus, a voice recognition server, a voice recognition system, and a voice recognition method, in which a general-purpose voice recognition engine may accurately recognize a limited number of words used in a specific area.

Abstract: Systems, methods, and apparatus for using at least one mobile device to receive a representation of at least one audio signal. In some embodiments, the at least one audio signal comprises speech of at least one of a plurality of first participants of a meeting, the plurality of first participants participating in the meeting from a first location, and the at least one audio signal may be audibly rendered to at least one second participant of the meeting at a second location different from the first location. In some embodiments, the at least one mobile device may further receive an indication of an identity of a leading speaker of the speech in the at least one audio signal, the leading speaker being identified from among the plurality of first participants, and may render the identity of the leading speaker to the at least one second participant.

Abstract: An electronic device and a method for controlling an electronic device are disclosed. The electronic device includes: a display unit; a voice input unit; and a controller displaying a plurality of contents on the display unit, receiving a voice command for controlling any one of the plurality of contents through the voice input unit, and controlling content corresponding to the received voice command. Multitasking performed by the electronic device can be effectively controlled through a voice command.

Abstract: A method (700) and system (900) for authenticating a user is provided. The method can include receiving one or more spoken utterances from a user (702), recognizing a phrase corresponding to one or more spoken utterances (704), identifying a biometric voice print of the user from one or more spoken utterances of the phrase (706), determining a device identifier associated with the device (708), and authenticating the user based on the phrase, the biometric voice print, and the device identifier (710). A location of the handset or the user can be employed as criteria for granting access to one or more resources (712).

Abstract: Techniques are provided to improve identification of a person using speaker recognition. In one embodiment, a unique social graph may be associated with each of a plurality of defined contexts. The social graph may indicate speakers likely to be present in a particular context. Thus, an audio signal including a speech signal may be collected and processed. A context may be inferred, and a corresponding social graph may be identified. A set of potential speakers may be determined based on the social graph. The processed signal may then be compared to a restricted set of speech models, each speech model being associated with a potential speaker. By limiting the set of potential speakers, speakers may be more accurately identified.

Abstract: Methods and systems for providing a voice-based digital signature service are disclosed. The method includes a first user sending a document to a second user for signature, and the first user also sending a PIN to the second user and to a voice verification authority. The second user sending, to the voice verification authority, a voice recording comprising the PIN along with consent of the second user to the PIN. The voice verification authority comparing the voice recording with a predefined voice sample of the second user, and the PIN received from the first user with the PIN received from the second user. The voice verification authority then sending a notification to a signing entity based on the comparison. The signing entity signing the PIN of the document with a private key associated with the second user and sends an acknowledgement to the first user and the second user.

Abstract: An identity authentication method is applied on a system. The system is connected to an external storage device storing a first voice model. The system includes an information server and a terminal. The information server includes a database. The information server executes the following steps. First, receiving the first voice model transmitted by the terminal. Second, determining whether the first voice model matches one second voice model, and transmitting the verification result to the terminal. The terminal executes the following steps. First, generating a prompt to prompt the user to input voice signals. Second, receiving the input voice signals. Third, extracting voice features from the input voice signals. Fourth, determining whether the extracted voice features matches the first voice model. Fifth, determining the verification result is successful when matches, and determining the identity authentication is success only when two verification results are both successful.

Abstract: Speaker content generated in an audio conference is visually represented in accordance with a method. Speaker content from a plurality of audio conference participants is monitored using a computer with a tangible non-transitory processor and memory. The speaker content from each of the plurality of audio conference participants is monitored. A visual representation of speaker content for each of the plurality of audio conference participants is generated based on the analysis of the speaker content from each of the plurality of audio conference participant. The visual representation of speaker content is displayed.

Abstract: Event audio data that is based on verbal utterances associated with a medical event associated with a patient is received. A list of a plurality of candidate text strings that match interpretations of the event audio data is obtained, based on information included in a medical speech repository, information included in a speech accent repository, and a matching function. A selection of at least one of the candidate text strings included in the list is obtained. A population of at least one field of an electronic medical form is initiated, based on the obtained selection.

Abstract: A touch sensor assembly having a selectable sensitivity level allows for a user to activate a touch sensor while wearing hand covers. The touch sensor assembly includes a touch sensor, a comparison unit, and a control unit. The touch sensor has a selectable sensitivity level, and detects a value corresponding to a capacitance of the touch sensor. The comparison unit has a predetermined threshold stored therein, and determines activation of the touch sensor if the detected value exceeds the predetermined value. The control unit is operable to select a sensitivity level of the touch sensor by varying the predetermined threshold by an amount unrelated to an environmental effect on the touch sensor.

Abstract: Functionality is described herein for recognizing speakers in an energy-efficient manner. The functionality employs a heterogeneous architecture that comprises at least a first processing unit and a second processing unit. The first processing unit handles a first set of audio processing tasks (associated with the detection of speech) while the second processing unit handles a second set of audio processing tasks (associated with the identification of speakers), where the first set of tasks consumes less power than the second set of tasks. The functionality also provides unobtrusive techniques for collecting audio segments for training purposes. The functionality also encompasses new applications which may be invoked in response to the recognition of speakers.

Abstract: Embodiments of the present invention provide a method, system and article of manufacture for adjusting a language model within a voice recognition system, based on text received from an external application. The external application may supply text representing the words of one participant to a text-based conversation. n such a case, changes may be made to a language model by analyzing the external text received from the external application.

Abstract: A speaker-verification digital signature system is disclosed that provides greater confidence in communications having digital signatures because a signing party may be prompted to speak a text-phrase that may be different for each digital signature, thus making it difficult for anyone other than the legitimate signing party to provide a valid signature.

Abstract: An embodiment of the present invention provides a speech recognition engine that utilizes portable voice profiles for converting recorded speech to text. Each portable voice profile includes speaker-dependent data, and is configured to be accessible to a plurality of speech recognition engines through a common interface. A voice profile manager receives the portable voice profiles from other users who have agreed to share their voice profiles. The speech recognition engine includes speaker identification logic to dynamically select a particular portable voice profile, in real-time, from a group of portable voice profiles. The speaker-dependent data included with the portable voice profile enhances the accuracy with which speech recognition engines recognize spoken words in recorded speech from a speaker associated with a portable voice profile.

Abstract: In an aspect, in general, a method for computer assisted speaker authentication in a voice communication session includes establishing a voice communication session between a first speaker and an agent, accepting a first voice signal from the first speaker, determining a voice characteristic measure of the first voice signal, including characterizing a similarity of the first voice signal to each of one or more stored characterizations of voice signals previously acquired from one or more known speakers, and providing an interface to the agent during the voice communication session between the agent and the first speaker, including presenting an indicator based on the determined voice characteristic measure to the agent.