Abstract: A method includes receiving audio data encoding an utterance and obtaining a set of bias phrases corresponding to a context of the utterance. Each bias phrase includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio to generate an output from the speech recognition model. The speech recognition model includes a first encoder configured to receive the acoustic features, a first attention module, a bias encoder configured to receive data indicating the obtained set of bias phrases, a bias encoder, and a decoder configured to determine likelihoods of sequences of speech elements based on output of the first attention module and output of the bias attention module. The method also includes determining a transcript for the utterance based on the likelihoods of sequences of speech elements.

Abstract: Technology is provided for causing a computing system to extract conversation features from a multiparty conversation (e.g., between a coach and mentee), apply the conversation features to a machine learning system to generate conversation analysis indicators, and apply a mapping of conversation analysis indicators to actions and inferences to determine actions to take or inferences to make for the multiparty conversation. In various implementations, the actions and inferences can include determining scores for the multiparty conversation such as a score for progress toward a coaching goal, instant scores for various points throughout the conversation, conversation impact score, ownership scores, etc. These scores can be, e.g., surfaced in various user interfaces along with context and benchmark indicators, used to select resources for the coach or mentee, used to update coach/mentee matchings, used to provide real-time alerts to signify how the conversation is going, etc.

Abstract: Disclosed herein are embodiments of systems, methods, and products comprises an analytic server, which automatically manages appointment scheduling. The analytic server receives a customer request to schedule an appointment. The analytic server determines the required data from both customer and service provider for making the appointment. The analytic server retrieves customer data comprising requested service attributes, user preferences, users attributes from internal database and external data source. The analytic server retrieves service providers' data comprising provider service attributes, providers' attributes from internal database and external data sources. The analytic server accesses external data source by web crawling various websites. The analytic server executes an artificial intelligence model to predict user preferences and needs. The analytic server determines potential service providers best matching the customer's input or predicted preferences.

Abstract: A home appliance including a communication device configured to communicate with another home appliance, a microphone configured to receive a voice from a user, and a processor configured to perform signal processing on first voice data obtained from the microphone and perform voice recognition using the signal-processed first voice data. Wherein the processor generates noise data using second voice data received from the other home appliance and performs the signal processing on the first voice data using the generated noise data.

Abstract: Techniques for reducing memory and processing resources used by a dialog system by sharing resources between pipelined processes of the dialog system. An integrated shared dictionary is constructed for concurrent use by automated speech recognition (ASR) and natural language understanding (NLU) subsystems of the dialog system. The integrated shared dictionary comprises multiple entries, with each entry comprising first information that is used by the ASR subsystem, second information used by the NLU subsystem, and information correlating the first information and the second information. The ASR subsystem uses the integrated shared dictionary to identify a dictionary entry containing a set of words corresponding to speech input. The dictionary entry information is communicated to the NLU subsystem, which uses the entry to generate a meaning representation for the speech input.

Abstract: Audiovisual performances, including vocal music, are captured and coordinated with those of other users in ways that create compelling user experiences. In some cases, the vocal performances of individual users are captured (together with performance synchronized video) on mobile devices, television-type display and/or set-top box equipment in the context of karaoke-style presentations of lyrics in correspondence with audible renderings of a backing track. Contributions of multiple vocalists are coordinated and mixed in a manner that selects for visually prominent presentation performance synchronized video of one or more of the contributors. Prominence of particular performance synchronized video may be based, at least in part, on computationally-defined audio features extracted from (or computed over) captured vocal audio.

Abstract: The systems and methods described herein can generate a voice-based interface to increase the accuracy of translations. The voice-based interface can result in fewer input audio signals being transmitted between devices of a network. Reducing the number of redundant translation requests that are sent between the devices of a network can save bandwidth and other computational resources by processing fewer input audio signals.

Abstract: An audio confirmation system includes a voice acquiring section configured to acquire a voice contained in a motion picture; a voice text producing section configured to produce a voice text based on the acquired voice; a determining section configured to determine whether or not the produced voice text and a caption text that is embedded in an image contained in the motion picture correspond to each other; and an outputting section configured to output a result of the determination of the determining section.

Abstract: A computer-implemented method for automatically analyzing a natural language input for information extraction comprises (i) a step of receiving the natural language input; (ii) a step of providing a grammar model comprising: a local grammar model, a set of external functions, and a finite set of read/write shared memory registers used by a parsing engine and the external functions; (iii) a step of applying the grammar model to the natural language input using the parsing engine, and (iv) a step of extracting information from the natural language input using at least one new output of the grammar model, the new output of the grammar model being built based: on at least one return value of the external functions from evaluating the one or more external functions in step (iii), and one or more input labels and/or output labels.

Abstract: A voice recognizing apparatus includes a sound input unit, a voice level calculating unit, a noise level calculating unit, a character converting unit, a reliability calculating unit, and a necessary voice level calculating unit. The sound input unit is configured to receive electrical signals as voice and noise signals converted from voice of a talker and noise in environment, respectively. The voice level calculating unit and the noise level calculating unit are configured to calculate, as voice and noise levels, levels of the voice and noise signals, respectively. The character converting unit is configured to perform conversion of a waveform of the electric signal as the voice signal into a character string. The reliability calculating unit is configured to calculate reliability of the conversion. The necessary voice level calculating unit is configured to calculate a necessary voice level on the basis of the voice and noise levels and the reliability.

Abstract: A method and apparatus are provided for automatically generating and processing first and second concept vector sets extracted, respectively, from a first set of concept sequences and from a second, temporally separated, concept sequences by performing a natural language processing (NLP) analysis of the first concept vector set and second concept vector set to detect changes in the corpus over time by identifying changes for one or more concepts included in the first and/or second set of concept sequences.

Abstract: An apparatus includes processor(s) to: perform preprocessing operations of a segmentation technique including divide speech data set into data chunks representing chunks of speech audio, use an acoustic model with each data chunk to identify pauses in the speech audio, and analyze a length of time of each identified pause to identify a candidate set of likely sentence pauses in the speech audio; and perform speech-to-text operations including divide the speech data set into data segments that each representing segments of the speech audio based on the candidate set of likely sentence pauses, use the acoustic model with each data segment to identify likely speech sounds in the speech audio, analyze the identified likely speech sounds to identify candidate sets of words likely spoken in the speech audio, and generate a transcript of the speech data set based at least on the candidate sets of words likely spoken.

Abstract: A method for generating final transcriptions representing numerical sequences of utterances in a written domain includes receiving audio data for an utterance containing a numeric sequence, and decoding, using a sequence-to-sequence speech recognition model, the audio data for the utterance to generate, as output from the sequence-to-sequence speech recognition model, an intermediate transcription of the utterance. The method also includes processing, using a neural corrector/denormer, the intermediate transcription to generate a final transcription that represents the numeric sequence of the utterance in a written domain. The neural corrector/denormer is trained on a set of training samples, where each training sample includes a speech recognition hypothesis for a training utterance and a ground-truth transcription of the training utterance. The ground-truth transcription of the training utterance is in the written domain.

Abstract: A home appliance including a first microphone that is disposed on a surface of a housing, a second microphone that is disposed on an inside of the housing, and a processor configured to perform signal processing for first voice data that is acquired from the first microphone, and perform voice recognition using the signal-processed first voice data. The processor is further configured to generate noise data using second voice data that is acquired from the second microphone and perform signal processing for the first voice data using the generated noise data.

Abstract: A dialog generation method includes: training a sequence to sequence (seq2seq)-based dialog model using a loss function including topic range constraint information; and generating a dialog using the trained dialog model. With the dialog generation method, topic range constraint information is introduced in the process of dialog model training using a loss function including the topic range constraint information, thus helping to prevent the trained model from producing low-quality meaningless replies.

Abstract: The present disclosure provides methods, computing devices, and storage media for generating a training corpus. The method includes: mining out pieces of data from user behavior logs associated with a target application, each piece of data including a first behavior log and a second behavior log, the first behavior log including a user speech and a corresponding speech recognition result, the second behavior log belonging to the same user as the first behavior log and time-dependent with the first behavior log; and determining the user speech and the corresponding speech recognition result in each piece of data as a positive feedback sample or a negative feedback sample, based on the first behavior log and the second behavior log.

Abstract: A plurality of messages are received and for each respective one of the received plurality of messages: a series of steps can be performed, including identifying language in the respective message, determining at least one heuristic, generating a first modified version of the respective message and generating at least one second modified version of the respective message. The second modified version(s) include language not in the respective message and not in the first modified version of the respective message. Further the at least one second modified version represents the determined at least one other heuristic. A device can be prompted to respond to a survey that includes the respective message(s), at least one first modified version, and/or at least one second modified version. A second survey can be generated of the respective message(s), modified version(s), and/or second modified version(s).

Abstract: An apparatus includes processor(s) to: perform pre-processing operations including derive an audio noise level of speech audio of a speech data set, derive a first relative weighting for first and second segmentation techniques for identifying likely sentence pauses in the speech audio based on the audio noise level, and select likely sentence pauses for a converged set of likely sentence pauses from likely sentence pauses identified by the first and/or second segmentation techniques based on the first relative weighting; and perform speech-to-text processing operations including divide the speech data set into data segments representing speech segments of the speech audio based on the converged set of likely sentence pauses, and derive a second relative weighting based on the audio noise level for selecting words indicated by an acoustic model or by a language model as being most likely spoken in the speech audio for inclusion in a transcript.

Abstract: Embodiments relate to an intelligent computer platform to selectively amend one or more tokens in a document. A first document set is subjected to natural language processing (NLP) and a vector score is identified for two or more documents of the first document set. Upon receipt of a new document, the new document is subjected to NLP and a new document vector score is identified. The new document is analyzed against the first document set, and the identified vector score of the first document set is compared to the vector score of the new document. One or more tokens of the new document are amended responsive to the comparison, and a new document version is created from the selective amendment.

Abstract: The present disclosure is directed to systems, methods and devices for providing real-time translation for group communications. A speech input may be received from a first group communication device associated with a first language. One or more groups to distribute the speech input may be determined, wherein each of the one or more groups comprises at least one group communication device associated with a language that is different than the first language. The received speech input may be translated into a corresponding language for each of the one or more groups, and the translated speech may be sent to each group communication device of the one or more groups in a language corresponding to each of the one or more groups.