Abstract: Aspects of the subject matter described herein relate to language translation. In aspects, a reference to a language translation component is embedded or otherwise inserted into a Web page. When the Web page is rendered, code corresponding to the language translation component may be downloaded and executed. Once executed, the translation component may access other content in the Web page and allow a user to request translation of the Web page. Upon receiving an indication that translation is desired, the translation component may send content in the Web page to a translation service and receive translated content. The translation component may then provide this translated content to a user viewing the Web page.

Abstract: A system, a computer program product, and method for controlling synthesized speech output on a voice-controlled device. One or more sensors are used to detect whether one person or more than one person is within a first settable distance from the voice-controlled device. Next a determination is made whether the audio input received is recognized as speech. In response to only one person being detected within the settable distance, begin outputting synthesized speech based on the audio input without waiting for an attention word to be recognized and otherwise wait for additional criteria before outputting synthesized speech based on the speech input. The additional criteria includes determining that more than one person is detected and recognizing that the attention word is received before outputting synthesized speech based on the audio input.

Abstract: Method and apparatus are provided for reconstructing a noise component of a speech/audio signal. A bitstream, is received and decoded to obtain a speech/audio signal. A first speech/audio signal is determined according to the speech/audio signal. A symbol of each sample value in the first speech/audio signal and an amplitude value of each sample value in the first speech/audio signal is determined. An adaptive normalization length and an adjusted amplitude value of each sample value are determined according to the adaptive normalization length and the amplitude value of each sample value. A second speech/audio signal is determined according to the symbol of each sample value and the adjusted amplitude value of each sample value.

Abstract: A “Facet Recommender” creates conversational recommendations for facets of particular conversational topics, and optionally for things associated with those facets, from consumer reviews or other social media content. The Facet Recommender applies a machine-learned facet model and optional sentiment-model, to identify facets associated with spans or segments of the content and to determine neutral, positive, or negative consumer sentiment associated with those facets and, optionally, things associated with those facets. These facets are selected by the facet model from a list or set of manually defined or machine-learned facets for particular conversational topic types. The Facet Recommender then generates new conversational utterances (i.e., short neutral, positive or negative suggestions) about particular facets based on the sentiments associated with those facets. In various implementations, utterances are fit to one or more predefined conversational frameworks.

Abstract: Processing circuitry of an information processing apparatus obtains a set of identity vectors that are calculated according to voice samples from speakers. The identity vectors are classified into speaker classes respectively corresponding to the speakers. The processing circuitry selects, from the identity vectors, first subsets of interclass neighboring identity vectors respectively corresponding to the identity vectors and second subsets of intraclass neighboring identity vectors respectively corresponding to the identity vectors. The processing circuitry determines an interclass difference based on the first subsets of interclass neighboring identity vectors and the corresponding identity vectors; and determines an intraclass difference based on the second subsets of intraclass neighboring identify vectors and the corresponding identity vectors.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designating certain voice commands as hotwords. The methods, systems, and apparatus include actions of receiving a hotword followed by a voice command. Additional actions include determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, where a voice command that is designated as a hotword is treated as a voice input regardless of whether the voice command is preceded by another hotword. Further actions include, in response to determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, designating the voice command as a hotword.

Abstract: Among other things, requests are received from voice assistant devices expressed in accordance with different corresponding protocols of one or more voice assistant frameworks. Each of the requests represents a voiced input by a user to the corresponding voice assistant device. The received requests are re-expressed in accordance with a common request protocol. Based on the received requests, responses to the requests are expressed in accordance with a common response protocol. Each of the responses is re-expressed according to a protocol of the framework with respect to which the corresponding request was expressed. The responses are sent to the voice assistant devices for presentation to the users.

Abstract: A process of an audio stream on a receiving side is facilitated. Encoding processing is performed on audio data and an audio stream in which an audio frame including audio compression data is continuously arranged is generated. Tag information indicating that the audio compression data of a predetermined sound unit is included is inserted into the audio frame including the audio compression data of the predetermined sound unit. A container stream of a predetermined format including the audio stream into which the tag information is inserted is transmitted.

Abstract: When transcribing large audio files, such as in the case of legal depositions, there are often many transcribers to choose from. Embodiments described herein enable calculation of expected accuracy of transcriptions by transcribers, which can be used to guide the selection of transcribers for specific tasks. In one embodiment, a computer receives a segment of an audio recording that includes speech of a person, and identifies an accent of the person and a topic of the segment. The computer generates feature values based on data that includes the accent and the topic, and utilizes a model to calculate, based on the feature values, an expected accuracy of a transcription of the segment by a certain transcriber. The model is generated based on training data that includes segments of previous audio recordings and values of accuracies of transcriptions, by the certain transcriber, of the segments.

Abstract: An audio firewall system has a microphone that generates audio data. A speech-to-text engine converts the audio data to text data. The text data is parsed for a service wake word and corresponding content data. The service wake word identifies one of a local security system and a remote assistant server. A text-to-speech engine converts the service wake word and the corresponding content data to converted audio data. The converted audio data is provided to the remote assistant server. The content data is provided to the local security system. The audio firewall system receives a response from the remote assistant server or the local security system and outputs an audio signal corresponding to the response.

Abstract: According to some embodiments, a system and method are provided to extract relationships from unstructured text documents. The method comprises receiving a training set of sentences that comprise labeled objects and subjects for creating an initial relationship model. A set of unlabeled sentences may be received. Objects and subjects from the set of unlabeled sentences are determined based on the initial model and the determined objects and subjects from the set of unlabeled sentences are displayed to a user for feedback and approval. An indication of whether the determined objects and subjects from the set of unlabeled sentences are correct is received and the initial relationship model is updated based on the received indication.

Abstract: Embodiments of the present disclosure provide a method and a device for processing a voice message, a terminal and a storage medium. The method includes: receiving a voice message sent by a user, the voice message being obtained based on an unordered version of language interaction; determining a corresponding spectrum of frequency domain feature based on the voice message, and performing a signal processing on the spectrum of frequency domain feature to obtain a first acoustic feature based on frame sequence and corresponding to the spectrum of frequency domain feature; and performing a feature extraction on the first acoustic feature to obtain a second acoustic feature based on an ivector algorithm and a deep convolutional neural network algorithm with residual processing, converting the second acoustic feature into a voiceprint model corresponding to the user, and storing the voiceprint model in a voiceprint model database.

Abstract: Creating and deploying a voice from text-to-speech, with such voice being a new language derived from the original phoneset of a known language, and thus being audio of the new language outputted using a single TTS synthesizer. An end product message is determined in an original language n to be outputted as audio n by a text-to-speech engine, wherein the original language n includes an existing phoneset n including one or more phonemes n. Words and phrases of a new language n+1 are recorded, thereby forming audio file n+1. This new audio file is labeled into unique units, thereby defining one or more phonemes n+1. The new phonemes of the new language are added to the phoneset, thereby forming new phoneset n+1, as a result outputting the end product message as an audio n+1 language different from the original language n.

Abstract: A system determines user intent from text. A conversation element is received. An intent is determined by matching a domain independent relationship and a domain dependent term determined from the received conversation element to an intent included in an intent database that stores a plurality of intents and by inputting the matched intent into a trained classifier that computes a likelihood that the matched intent is the intent of the received conversation element. An action is determined based on the determined intent. A response to the received conversation element is generated based on the determined action and output.

Abstract: A voice recognition apparatus, a voice recognition method, and a non-transitory computer readable recording medium are provided. The voice recognition apparatus includes a storage configured to store a preset threshold value for voice recognition; a voice receiver configured to receive a voice signal of an uttered voice; and a voice recognition processor configured to recognize a voice recognition starting word from the received voice signal, perform the voice recognition on the voice signal in response to a similarity score, which represents a recognition result of the recognized voice recognition starting word, being greater than or equal to the stored preset threshold value, and change the preset threshold value based on the recognition result of the voice recognition starting word.

Abstract: A audio signal encoding method includes: dividing a frequency band of an audio signal into a plurality of sub-bands, and quantifying a sub-band normalization factor of each sub-band; determining signal bandwidth of bit allocation according to the quantified sub-band normalization factor, or according to the quantified sub-band normalization factor and bit rate information; allocating bits for a sub-band within the determined signal bandwidth; and coding a spectrum coefficient of the audio signal according to the bits allocated for each sub-band. According to embodiments of the present disclosure, during coding and decoding, signal bandwidth of bit allocation is determined according to the quantified sub-band normalization factor and bit rate information. In this manner, the determined signal bandwidth is effectively coded and decoded by centralizing the bits, and audio quality is improved.

Abstract: A computer-implemented method includes associating, using a processor, one or more words in an electronic agenda template to at least one agenda item indicative of a point for discussion. The processor captures a real-time interaction comprising speech from one or more participants of a plurality of discussion participants into a digital representation. The processor isolates a portion of the real-time interaction from the digital representation. The portion is associated with a single speaker of the plurality of discussion participants. The processor makes at least one match between an isolated portion of the real-time interaction and the at least one agenda item. The processor determines an intent of the single speaker from the isolated portion and matching the determined intent of the single speaker to the at least one agenda item on the electronic agenda template, and generates discussion minutes output based on the matched intent and agenda item.

Abstract: An approach is provided in which an information handling system sends a request in audio format to a user over a voice channel requesting a user data set. The information handling system receives utterances from the user over the voice channel and determines that the utterances do not provide enough information to complete the requested user data set. In turn, the information handling system establishes a messaging channel with the user and sends a request in digital format to the user over the messaging channel to provide additional data to complete the user data set.

Abstract: [Object] To present a response to a natural sentence in a more suitable aspect even in circumstances in which a natural sentence with ambiguity can be input. [Solution] An information processing device including: an acquisition unit configured to acquire an extraction result of candidates for a response to an input which is based on first information indicating a result of natural language analysis on a natural sentence acquired as the input and second information indicating a state or a situation involved in use of a predetermined device; and a control unit configured to cause a predetermined output unit to present information indicating the candidates for the response in an aspect corresponding to the extraction result of the candidates.

Abstract: A computing device includes an interface configured to interface and communicate with a communication system, a memory that stores operational instructions, and processing circuitry operably coupled to the interface and to the memory that is configured to execute the operational instructions to perform various operations. The computing device processes a message that is provided from a sender and is intended for a recipient associated with another computing device in accordance with topic, emotive content, and/or social content to generate a classification model for the message that includes classification parameter value(s). When appropriate to perform message transformation, the computing device selects a tonal transformation based on the classification parameter value(s) and processes the message in accordance with the tonal transformation to generate a normalized message.