Abstract: A method to transcribe communications may include selecting a first transcription generation technique from among multiple transcription generation techniques for generating transcriptions of audio of one or more communication sessions that involve a user device and obtaining performances of the multiple transcription generation techniques with respect to generating the transcriptions of the audio. The method may also include monitoring comparisons between the performances of the multiple transcription generation techniques and obtaining input from the user with respect to the comparisons. The method may further include selecting a second transcription generation technique from among the multiple transcription generation techniques based on the input from the user.

Abstract: Methods, systems, and devices for systems and techniques for processing keywords in audio data are described. In some devices configured with a virtual assistant, an audio processing component may support a command-first, keyword-second voice activation procedure. The audio processing component may receive audio data from a microphone and may compress a portion of the audio data and store the compressed audio data in a first buffer and may store a portion of the audio data that is uncompressed in a second buffer. The audio processing component may use the uncompressed audio data to detect the presence of a keyword and use the compressed audio data to identify a command associated with the keyword. Upon detection of the keyword, the audio processing component may decompress the compressed audio data and transmit the decompressed audio data and the uncompressed audio data to a main processor of the device.

Abstract: Systems and methods for creating and editing an inspection plan and directing an inspector through an inspection are provided. An exemplary system, according to one implementation, comprises a mobile computing device and a server computer. The mobile computing device is configured to communicate audible prompts to an inspector and receive audible replies from the inspector. The server computer is configured to store an inspection plan comprising a sequence of inspection steps, translate each of the inspection steps of the inspection plan into audible prompts, transmit the audible prompts to the mobile computing device, receive the audible replies from the mobile computing device, and translate the audible replies into a set of inspection results.

Abstract: A voice recognition device includes a memory and a processor. The processor is configured to store in the memory, digital voice data corresponding to a voice signal input from a voice input unit, recognize a spoken voice utterance from the voice data after a voice input start instruction is received, determine whether to correct the recognition result of the spoken voice utterance based on a time interval from a time when the voice input start instruction is received to a time when the voice signal is input via the voice input unit, and correct the recognition result of the voice utterance based on the time interval.

Abstract: A method and system architecture for automation and alarm systems is provided. According to exemplary embodiments, relatively simple processing tasks are performed at the sensor level, with more complex processing being shifted to the gateway entity or a networked processing device. The gateway entity dynamically allocates processing resources for sensors. If a sensor detects than an event is occurring, or predicts that an event is about to occur, the sensor submits a resources allocation request and a power balancer running on the gateway entity processes the request. In response to the resources allocation request, the gateway entity allocates some processing resources to the requesting sensor and the data is processed in real-time or near-real-time by the gateway entity.

Abstract: In accordance with an embodiment, described herein is a system and method for semantic analysis and use of song lyrics in a media content environment. Semantic analysis is used to identify persons, events, themes, stories, or other meaningful information within a plurality of songs. For each song, a story graph is generated which describes a narrative within that song's lyrics. The story graph is then used to determine a feature vector associated with the song's narrative. In response to receiving an input vector, for example as a search input for a particular song track, the input vector can be matched against feature vectors of the plurality of songs, to determine appropriate tracks. Example use cases include the selection and delivery of media content in response to input searches for songs of a particular nature, or the recommendation or suggestion of media content in social messaging or other environments.

Abstract: A computer hardware-implemented method, system, and/or computer program product prevents a cascading failure in a complex stream computer system causing an untrustworthy output from the complex stream computer system. Multiple upstream subcomponents in a complex stream computer system generate multiple outputs, which are used as inputs to a downstream subcomponent, wherein the multiple upstream subcomponents execute upstream computational processes. An accuracy value is assigned to each of the multiple outputs from the upstream subcomponents, and weighting values are assigned to each of the inputs to the downstream subcomponent. If using the accuracy values and weighting values fails to adjust the downstream subcomponent to meet a predefined trustworthiness level for making a first type of prediction, then a new downstream computational process that produces a different second type of prediction is executed.

Abstract: A method, a system, and a computer program product for providing media to a requester at a particular playback rate associated with the requester. The method includes receiving a request from a requester for a playback session of media that includes a time varying content. In response to receiving the request, a profile associated with the requester is accessed to determine a playback rate of the media for the requester. In response to determining the playback rate of the media for the requester, the media is provided to the requester at the determined playback rate. The method further includes monitoring the playback session of the media for playback changes by the requester and dynamically adapting the playback rate associated with the requester based on the type and frequency of playback changes.

Abstract: A context-aware voice guidance method is provided that interacts with other voice services of a user device. The voice guidance does not provide audible guidance while the user is making a verbal request to any of the voice-activated services. Instead, the voice guidance transcribes its output on the screen while the verbal requests from the user are received. In some embodiments, the voice guidance only provides a short warning sound to get the user's attention while the user is speaking on a phone call or another voice-activated service is providing audible response to the user's inquires. The voice guidance in some embodiments distinguishes between music that can be ducked and spoken words, for example from an audiobook, that the user wants to pause instead of being skipped. The voice guidance ducks music but pauses spoken words of an audio book in order to provide voice guidance to the user.

Abstract: A data input system has a processor which receives user input comprising a sequence of one or more items and a language model which computes candidate next items in the sequence using the user input. A training engine trains the language model using data about a plurality of true words which a user intended to input using the data input system, and for each true word, at least one alternative candidate, being a word computed assuming imperfect entry of the true word to the data input system.

Abstract: An identity verification system enables the identity of an individual to be verified to others using the internet. An initial identification ceremony is recorded in which the user performs instructions that cannot be known in advance, such as reading text that cannot be anticipated. The initial ceremony can be replayed and authenticated by individuals who already personally know the user. Alternatively, the identity of the user in the initial ceremony can be authenticated using other existing techniques such as KBA. A secondary instruction ceremony is subsequently performed when identity verification is required in order to authorize a directive or transaction. In the secondary instruction ceremony the user performs unforeseeable instructions such as reading text that cannot be anticipated and reading aloud an indication of the transaction.

Abstract: A state of an application program can be stored and transferred to a remote system. The remote system attempts to recreate the original state of the application program. If the remote system is unable to do so, an image of the state of the application program is obtained, instead. Assignment of control to various functions of an application program is achieved by associating a function (i.e., modifying a parameter) with a user control at a remote location.

Abstract: A transcription system automates the control of the playback of the audio to accommodate the user's ability to transcribe the words spoken. In some examples, a delay between playback and typed input is estimated by processing the typed words using a wordspotting approach. The estimated delay is used as in input to an automated speed control, for example, to maintain a target or maximum delay between playback and typed input.

Abstract: The apparatus for synchronously processing text data and voice data, comprises: a storing unit for storing text data and voice data; a text data dividing section for dividing the text data; a text data phoneme converting section for phonemically converting the divided text data; a text data phoneme conversion accumulated value calculating section for calculating accumulated values of text data phoneme conversion values; a voice data dividing section for dividing the voice data; a reading data phoneme converting section for phonemically converting the divided voice data; a voice data phoneme conversion accumulated value calculating section for calculating accumulated values of voice data phoneme conversion values; a phrase corresponding data producing section for producing phrase corresponding data; and an output section for synchronously outputting the text data and the divided voice data.

Abstract: According to some aspects, a method for improving understandability of audio corresponding to dictation to assist a transcriptionist in transcribing the dictation is provided. The method comprises presenting a user interface to the transcriptionist, the user interface including at least one control that can be selectively set to one of a plurality of settings, receiving a selection of one of the plurality of settings via the at least one control, and compressing a dynamic range of at least a portion of the audio using at least one parameter value associated with the selected setting.

Abstract: Disclosed are techniques and systems to provide a narration of a text. In some aspects, the techniques and systems described herein include generating a timing file that includes elapsed time information for expected portions of text that provides an elapsed time period from a reference time in an audio recording to each portion of text in recognized portions of text.

Abstract: A system includes a memory operable to store a search index. The system also includes a processor communicatively coupled to the memory. The processor is operable to receive a search request relating to information stored in an ontology. The processor is further operable to parse the search request to determine a search type. The processor is further operable to query, based at least in part on the search type, one or more of the search index and the ontology.

Abstract: Systems and methods that can be utilized to convert a voice communication received over a telecommunication network to text are described. In an illustrative embodiment, a call processing system coupled to a telecommunications network receives a call from a caller intended for a first party, wherein the call is associated with call signaling information. At least a portion of the call signaling information is stored in a computer readable medium. A greeting is played the caller, and a voice communication from the caller is recorded. At least a portion of the voice communication is converted to text, which is analyzed to identify portions that are inferred to be relatively more important to communicate to the first party. A text communication is generated including at least some of the identified portions and including fewer words than the recorded voice communication. At least a portion of the text communication is made available to the first party over a data network.

Abstract: The invention provides a system, method, and business model for an information system and service having business self-promotion, promotion and promotion tracking, loyalty or frequent participant rewards and redemption, audio coupon, ratings, and other features. A business or organization in which consumers call into a service using ordinary telephone, PC, PDA, or other information appliance, and make requests in plain speech for information on goods and/or services, and the service provides responses to the request in plain speech in real-time.

Abstract: A method of providing speech transcription performance indication includes receiving, at a user device data representing text transcribed from an audio stream by an ASR system, and data representing a metric associated with the audio stream; displaying, via the user device, said text; and via the user device, providing, in user-perceptible form, an indicator of said metric. Another method includes displaying, by a user device, text transcribed from an audio stream by an ASR system; and via the user device, providing, in user-perceptible form, an indicator of a level of background noise of the audio stream. Another method includes receiving data representing an audio stream; converting said data representing an audio stream to text via an ASR system; determining a metric associated with the audio stream; transmitting data representing said text to a user device; and transmitting data representing said metric to the user device.

Abstract: Audio data that includes speech may be transcribed using a language model. The transcription may be provided to a user. The user may provide feedback on the transcription, and the language model may be updated based at least in part on the feedback. The feedback may include, for example, an affirmation of the transcription; a disapproval of the transcription; a correction to the transcription; a selection of an alternate transcription result; or any other kind of response.

Abstract: Systems and methods can include converting multi-channel circuit switched voice data to packet-switched voice over internet protocol (VoIP). A multi-channel connection originating from one or more customer premise equipment private branch exchanges can be terminated at a channel to packet gateway device. Call data originating from multiple customer premise equipment telephony devices can be received through the multi-channel connection associated with the one or more customer premise equipment private branch exchanges, and can be processed at the channel to packet gateway device responsive to call control instruction information. The payload data associated with the call data can be packaged according to predetermined packaging rules and transmitted according to VoIP.

Abstract: Audio data that includes speech may be transcribed to text by a speech recognition engine. One or more metrics associated with the audio data and/or the text may be determined. An indicator related to a metric may be provided for a portion of the audio data or the text for which the metric was determined. The indicator may be presented in a user-perceptible format.

Abstract: A method for providing suggestions includes capturing audio that includes speech and receiving textual content from a speech recognition engine. The speech recognition engine performs speech recognition on the audio signal to obtain the textual content, which includes one or more passages. The method also includes receiving a selection of a portion of a first word in a passage in the textual content, wherein the passage includes multiple words, and retrieving a set of suggestions that can potentially replace the first word. At least one suggestion from the set of suggestions provides a multi-word suggestion for potentially replacing the first word. The method further includes displaying, on a display device, the set of suggestions, and highlighting a portion of the textual content, as displayed on the display device, for potentially changing to one of the suggestions from the set of suggestions.

Abstract: A method of providing speech transcription performance indication includes receiving, at a user device data representing text transcribed from an audio stream by an ASR system, and data representing a metric associated with the audio stream; displaying, via the user device, said text; and via the user device, providing, in user-perceptible form, an indicator of said metric. Another method includes displaying, by a user device, text transcribed from an audio stream by an ASR system; and via the user device, providing, in user-perceptible form, an indicator of a level of background noise of the audio stream. Another method includes receiving data representing an audio stream; converting said data representing an audio stream to text via an ASR system; determining a metric associated with the audio stream; transmitting data representing said text to a user device; and transmitting data representing said metric to the user device.

Abstract: Methods and systems for predicting a text are described. In an example, a computing device may be configured to receive one or more typed characters that compose a portion of a text; and receive, a voice input corresponding to a spoken utterance of at least a portion of the text. The computing device may be configured to determine, based on the one or more typed characters and the voice input, one or more candidate texts predicting the text. Further, the computing device may be configured to provide the one or more candidate texts.

Abstract: Speech processing is disclosed for an apparatus having a main processing unit, a memory unit, and one or more co-processors. Memory maintenance and voice recognition result retrievals upon execution are performed with a first main processor thread. Voice detection and initial feature extraction on the raw data are performed with a first co-processor. A second co-processor thread receives feature data derived for one or more features extracted by the first co-processor thread and information for locating probability density functions needed for probability computation by a speech recognition model and computes a probability that the one or more features correspond to a known sub-unit of speech using the probability density functions and the feature data. At least a portion of a path probability that a sequence of sub-units of speech correspond to a known speech unit is computed with a third co-processor thread.

Abstract: A system for selection by a user and delivery to the user over an internetwork transmission channel of selected audio data files at a delivery rate of at least twice the delivery rate for normal, audibly perceptible playback of an audio data file. The user registers the user's selection of audio material with a central library of audio and/or text data files, and a digitized and optionally compressed omnibus file containing the user's selections is prepared and transmitted to the user at a high data transfer rate. The user receives downloads the selected data files to a personal computer or to a portable storage and playback unit (SPU) that may store and play back digitized text or audio data, using a docking station. The user carries this SPU until the user has an opportunity to audio process and play back the text or audio data files in audibly perceptible form.

Abstract: The invention discloses a system and method for filling out a form from a dialog between a caller and a call center agent. The caller and the caller center agent can have the dialog in the form of telephone conversation, instant messaging chat or email exchange. The system and method provides a list of named entities specific to the call center operation and uses a translation and transcription minor to filter relevant elements from the dialog between the caller and the call center agent. The relevant elements filtered from the dialog are subsequently displayed on the call center agent's computer screen to fill out application forms automatically or through drag and drop operations by the call center agent.

Abstract: A computerized method of detecting a target word in a speech signal. A speech recognition engine and a previously constructed phoneme model is provided. The speech signal is input into the speech recognition engine. Based on the phoneme model, the input speech signal is indexed. A time-ordered list is stored representing n-best phoneme candidates of the input speech signal and phonemes of the input speech signal in multiple phoneme frames. The target word is transcribed into a transcription of target phonemes. The time-ordered list of n-best phoneme candidates is searched for a locus of said target phonemes. While searching, scoring is based on the ranking of the phoneme candidates among the n-best phoneme candidates and based on the number of the target phonemes found. A composite score of the probability of an occurrence of the target word is produced. When the composite score is higher than a threshold, start and finish times are output which bound the locus.

Abstract: The present invention can include a speech processing method for providing dictation capabilities to a voice server. The method can include a step of establishing a real-time voice communication session involving a voice interface. Speech for the communication session can be streamed to a remotely located voice server. A real-time stream of text can be received from the voice server. The stream of text can include text that has been speech-to-text converted by the voice server from the streamed speech. The voice server can use a MRCP based non-halting interface to receive the real-time stream of speech and a delivery interface to deliver real-time text to a designated endpoint.

Abstract: A method for providing suggestions includes capturing audio that includes speech and receiving textual content from a speech recognition engine. The speech recognition engine performs speech recognition on the audio signal to obtain the textual content, which includes one or more passages. The method also includes receiving a selection of a portion of a first word in a passage in the textual content, wherein the passage includes multiple words, and retrieving a set of suggestions that can potentially replace the first word. At least one suggestion from the set of suggestions provides a multi-word suggestion for potentially replacing the first word. The method further includes displaying, on a display device, the set of suggestions, and highlighting a portion of the textual content, as displayed on the display device, for potentially changing to one of the suggestions from the set of suggestions.

Abstract: Provided is a method and a telephone-based system with voice-verification capabilities that enable a user to safely and securely conduct transactions with his or her online financial transaction program account over the phone in a convenient and user-friendly fashion, without having to depend on an internet connection.

Abstract: Video data and audio data corresponding to a predetermined attribute is retrieved from the video data and the audio data, each of which is stored in association with an attribute, and the retrieved items of the video data and the audio data are listed in a form showing a correlation between the video data and the audio data. In a case that items of the video data and audio data are selected, wherein said items of the video data and audio data are displayed and correlated with each other, the selected video data and audio data can be synchronized with each other and played-back.

Abstract: An automated transcription system includes an housing on a PC, and a portable electronic device including a mechanism for creating and managing a plurality of predetermined templates with a plurality of headings and sub-headings that are automatically populated in real time as a user speaks an audio message. The portable electronic device further includes a mechanism for converting and displaying the audio message to a text message on the portable electronic device and thereby enabling a user to read, edit and print the text message. Such an audio message converting and displaying mechanism includes an LCD screen, a microphone for receiving the audio message when the user speaks, and a data transfer interface.

Abstract: Knowledge-based information can be captured and processed to create a library of such knowledge. A maintenance worker performing a task for an asset can record audio and/or video information during the performance, and can upload the recording to a maintenance system. The system processes the recording to produce a text file corresponding to any speech during the recording, and generates a search index allowing the text file to be searched by a user. If the task is performed in the context of a work order, for example, information from the work order can be associated with the text file so that a user can search by text search, keyword, task, or other such information. A user then can locate and access the text file and/or the corresponding recording for playback.

Abstract: Disclosed are systems, methods and computer readable media for applying a multi-state barge-in acoustic model in a spoken dialogue system comprising the steps of (1) presenting a prompt to a user from the spoken dialog system. (2) receiving an audio speech input from the user during the presentation of the prompt, (3) accumulating the audio speech input from the user, (4) applying a non-speech component having at least two one-state Hidden Markov Models (HMMs) to the audio speech input from the user, (5) applying a speech component having at least five three-state HMMs to the audio speech input from the user, in which each of the five three-state HMMs represents a different phonetic category, (6) determining whether the audio speech input is a barge-in-speech input from the user, and (7) if the audio speech input is determined to be the barge-in-speech input from the user, terminating the presentation of the prompt.

Abstract: A computer based digital transcription system employs audio and video inputs of a court proceeding, and stores the digital signals in a memory in the form of distinct file segments of a predetermined time length during each recording session. The computer associates a date and time and a third identifier, such as location, with each distinct file segment. Playback of any desired segment may be effected during recording or at any time afterward; and playback does not interfere with the recording of realtime information.

Abstract: A transcription system having linked computer terminals for a court reporter and/or for examining, defending, and/or associate attorneys is disclosed. A method of language translation may be utilized by the transcription system during a testimonial proceeding, for example. The method involves receiving into the transcription system, in real-time, representations of words spoken in a first language during the testimonial proceeding. The representations are translated, in real-time, to text in the first language. The text in the first language is translated, in real-time, to text in a second language, and the text in the second language is communicated to a terminal for real-time display.

Abstract: The method of the present invention allows inventors to orally document innovative concepts. The method reduces the need to have inventors write out details of an invention in an invention disclosure form. The method also assists inventors in quickly and conveniently recording ideas and preparing invention disclosure forms based on the ideas. In some example forms, inventors are able to pick up a telephone and connect to a network. Once connected to the network an inventor can dictate the concepts of the idea over the network. The dictation is converted into text, such as by a voice analysis program, and then inserted into an invention disclosure form which is dated and archived.

Abstract: A method of indexing a speech segment includes identifying at least two alternative word sequences based on the speech segment. For each word in the alternative sequences, information is placed in an entry for the word in the index. The information indicates the position of the word in at least one of the alternative sequences.

Abstract: The system is designed to interface with external devices and services, to transcribe audio that may be stored elsewhere such as a wireless phone'voice mail, or occurring between two or more parties such as a conference call. An audio stream is separated into many audio shreds, each of which has duration of only a few seconds and cannot reveal the context of the conversation. A workforce of geographically distributed transcription agents who transcribe the audio shreds is able to generate transcription in real time, with many agents working in parallel on a single conversation. No one agent (or group of agents) receives a sufficient number of audio shreds to reconstruct the context of any conversation. The use of human transcribers allows the system to overcome limitations typical of computer-based speech recognition and permits accurate transcription of general-quality speech even in acoustically hostile environments.

Abstract: A computer program product resides on a computer-readable medium, and includes computer-readable, computer-executable instructions for causing a computer to analyze a first playback speed history for at least one audio recording recorded by a first speaker and played by a first listener, the playback speed history being indicative of at least one playback speed associated with the at least one audio recording, and to determine from the first playback speed history a speed setting for playback of another audio recording recorded by a second speaker to be played by a second listener.

Abstract: A computer based digital transcription system employs audio and video inputs of a court proceeding, and stores the digital signals in a memory in the form of distinct file segments of a predetermined time length during each recording session. The computer associates a date and time and a third identifier, such as location, with each distinct file segment. Playback of any desired segment may be effected during recording or at any time afterward; and playback does not interfere with the recording of realtime information.

Abstract: A personal medical dictation system that can be easily and conveniently used to capture and preserve audio information. The system includes a specially designed portable, hand-held recording component that is of a small size, but yet is capable of storing at least one hour of actual dictation in compressed form and a cooperating dictation receiver that functions to automatically transfer the recorded data to a central processing area. The hand-held recording component can be expeditiously, mechanically, and electrically transmitted to the central processing area for transcription. The dictation receiver component also automatically recharges the batteries of the hand-held recording unit.

Abstract: A hand microphone and an adaptor module form an assembly which is a peripheral device for a personal computer. The hand microphone is used to control dictation functions to be carried out by the PC. Two separate analog control signal channels are output from the hand microphone and applied, respectively, as X- and Y-axis inputs for the game port on the PC. Control signals carried in the two signal channels are generated by actuating control switches mounted on the hand microphone.

Abstract: A method and device for synchronizing data between analog and digital mediums recorded either simultaneously from a single source or recorded from different sources, which requires synchronization. User-Data recorded in analog medium is referenced with out-of-band unique digital reference-data that is generated by the interface device to mark the position of the user data. The same reference-data communicated by the device and the user-data are stored in the digital medium also and a relationship is computed and established in the form of a table between the reference-data and the positions of the recorded user-data. Whenever there is a manipulation of the User-data on either of the mediums for the purpose of viewing, listening or editing, the Reference-data in that medium is interpreted by the device and the corresponding location of the user-data in the other medium is accessed by using the table.

Abstract: A digital audio transcription system includes at least one source of audio signals to be recorded and a computer for storing digital signals corresponding to the audio signals for allowing the stored digital signals to be subsequently played back. Recording sessions are defined by signaling the start and stopping of the digital signals; and the computer associates a date and time with each file segment stored during a recording session. A playback selection allows a user to select a virtual file entry from file entries corresponding to the periods of time during which the computer has stored at least one recording session, with the computer being responsive to the playback selection to identify file segments stored in memory on the desired entry date from the selected source of audio signals, which collectively represent the selected virtual file entry.

Abstract: In an input unit (3) which can be operated by foot for a computer (1) which forms a dictating machine with foot-operated input means (24) for manually inputting control information (SI) by which information an audio reproduction mode of the dictating machine can be activated (25) or deactivated (26), and which, with activated audio reproduction mode, can deliver audio information (AI) stored in the dictating machine as an analog audio signal (AS3) to headphones (34) or a loudspeaker (32) respectively, and includes connection means (28, 33, 35) for connecting the input unit (3) to the computer (1) while the control information (SI) can be delivered to the computer (1) via the connection means (28, 33, 35), the connection means (28, 33, 35) are arranged for receiving the audio information (AI) as digital audio data from the computer (1) and for delivering the control information (SI) to the computer (1) over a digital data bus link and the headphones (34) or the loudspeaker can be connected to the connection me

Abstract: A portable digital voice recorder is interfaced for data communication with a personal computer. Voice data files are stored in the portable recorder together with header data which indicates the status of the voice data files. In response to a batch upload command, selected ones of the voice data files are uploaded to the personal computer, depending on the status of the voice data files as indicated by the status data. Graphical user interface software running in the PC causes the PC to display icons indicative of voice data files stored in the portable recorder. The header data in the portable recorder may be changed by operation of the PC's graphical user interface. For example, file designations indicated by header data in the portable recorder may be changed by editing corresponding fields displayed by the PC.