Abstract: A method for acquiring the number of modified frames for active sound, and a method and apparatus for voice activity detection are disclosed. Firstly, a first voice activity detection decision result and a second voice activity detection decision result are obtained (501), the number of hangover frames for active sound is obtained (502), and the number of background noise updates is obtained (503), and then the number of modified frames for active sound is calculated according to the first voice activity detection decision result, the number of background noise updates and the number of hangover frames for active sound (504), and finally, a voice activity detection decision result of a current frame is calculated according to the number of modified frames for active sound and the second voice activity detection decision result (505).

Abstract: A system is provided to determine teaching technique based upon sound amplitude comprising: processor; and a memory device holding an instruction set executable on the processor to cause the computer system to perform operations comprising: sampling amplitude of sound at a sampling rate; assigning a respective sound amplitude and a respective amplitude variation to the respective sound sample; and classifying the sound samples based upon the assigned sound amplitude and sound sample variation.

Abstract: An example apparatus for generating dialogue includes an audio receiver to receive audio data including speech. The apparatus also includes a verification score generator to generate a verification score based on the audio data. The apparatus further includes a user detector to detect that the verification score exceeds a lower threshold but does not exceed a higher threshold. The apparatus includes a dialogue generator to generate dialogue to solicit additional audio data to be used to generate an updated verification score in response to detecting that the verification score exceeds a lower threshold but does not exceed a higher threshold.

Abstract: Techniques for using a dynamic wakeword detection threshold are described. A server(s) may receive audio data corresponding to an utterance from a device in response to the device detecting a wakeword using a wakeword detection threshold. The server(s) may then determine the device should use a lower wakeword detection threshold for a duration of time. In addition to sending the device output data responsive to the utterance, the server(s) may send the device an instruction to use the lower wakeword detection threshold for the duration of time. Alternatively, the server(s) may train a machine learning model to determine when the device should use a lower wakeword detection threshold. The server(s) may send the trained machine learned model to the device for use at runtime.

Abstract: Provided herein is an integrated circuit including, in some embodiments, a special-purpose host processor, a neuromorphic co-processor, and a communications interface between the host processor and the co-processor configured to transmit information therebetween. The special-purpose host processor is operable as a stand-alone host processor. The neuromorphic co-processor includes an artificial neural network. The co-processor is configured to enhance special-purpose processing of the host processor through the artificial neural network. In such embodiments, the host processor is a keyword identifier processor configured to transmit one or more detected words to the co-processor over the communications interface. The co-processor is configured to transmit recognized words, or other sounds, to the host processor.

Abstract: A system and method to receive a spoken utterance and convert the spoken utterance into a recognized speech results through multiple automatic speech recognition modules. Multiple conversation modules interpret the recognized speech results. The system and method assign an affinity status to one or more of the multiple automatic speech recognition modules. An affinity status restricts the conversion of a subsequent spoken utterance to a selected automatic speech recognition module or modules.

Abstract: Provided is a system for a robot to eliminate its own sound source, comprising a sound pickup unit (1) used for acquiring a digital signal having mixed left and right sound channels and outputting same; a sound-card driver unit (2) used for mixing the received reference signal of the robot sound source and the digital signal having mixed left and right sound channels, to form a mixed voice signal, and outputting same; a signal separation unit (3) used for separating the mixed voice signal into a digital signal having mixed left and right sound channels and a reference signal, and determining whether the digital signal having the mixed left and right sound channels contains an interference signal, obtaining a playback signal volume, and outputting same.

Abstract: A language model may be used in a variety of natural language processing tasks, such as speech recognition, machine translation, sentence completion, part-of-speech tagging, parsing, handwriting recognition, or information retrieval. A natural language processing task may use a vocabulary of words, and a word hash vector may be created for each word in the vocabulary. A sequence of input words may be received, and a hash vector may be obtained for each word in the sequence. A language model may process the hash vectors for the sequence of input words to generate an output hash vector that describes words that are likely to follow the sequence of input words. One or words may then be selected using the output word hash vector and used for a natural language processing task.

Abstract: Vector Quantizer and method therein for vector quantization, e.g. in a transform audio codec. The method comprises comparing an input target vector with four centroids C0, C1, C0,flip and C1,flip, wherein centroid C0,flip is a flipped version of centroid C0 and centroid C1,flip is a flipped version of centroid C1, each centroid representing a respective class of codevectors. A starting point for a search related to the input target vector in the codebook is determined, based on the comparison. A search is performed in the codebook, starting at the determined starting point, and a codevector is identified to represent the input target vector. A number of input target vectors per block or time segment is variable. A search space is dynamically adjusted to the number of input target vectors. The codevectors are sorted according to a distortion measure reflecting the distance between each codevector and the centroids C0 and C1.

Abstract: A method for text-to-speech (TTS) pronunciation correction includes initiating a correction of a pronunciation error produced by a TTS system; receiving an input from a user for the correction of the pronunciation error; and updating the pronunciation based on the input provided by the user. The user input can include providing one or more voice recording entries, a manual entry from a keyboard, and/or choosing an entry from a predetermined list of phonetic strings. The method can also include pronouncing the corrected pronunciation entry back to the user. The user input can be further used for correction of similar syllables, words, and/or phrases.

Abstract: A mechanism is described for facilitating simultaneous recognition and processing of multiple speeches from multiple users according to one embodiment. A method of embodiments, as described herein, includes facilitating a first microphone to detect a first speech from a first speaker, and a second microphone to detect a second speech from a second speaker. The method may further include facilitating a first beam-former to receive and process the first speech, and a second beam-former to receive and process the second speech, where the first and second speeches are at least received or processed simultaneously. The method may further include communicating a first output associated with the first speech and a second output associated with the second speech to the first speaker and the second speaker, respectively, using at least one of one or more speaker devices and one or more display devices.

Abstract: A device and method for adjusting speech intelligibility at an audio device is provided. The device comprises a microphone, a transmitter and a controller. The controller is configured to: determine a noise level at the microphone; select a voice tag, of a plurality of voice tags, based on the noise level, each of the plurality of voice tags associated with respective noise levels; determine an intelligibility rating of a mix of the voice tag and noise received at the microphone; and when the intelligibility rating is below a threshold intelligibility rating, enhance speech received the microphone based on the intelligibility rating prior to transmitting, at the transmitter, a signal representing intelligibility enhanced speech.

Abstract: Systems and processes are disclosed for discovering trending terms in automatic speech recognition. Candidate terms (e.g., words, phrases, etc.) not yet found in a speech recognizer vocabulary or having low language model probability can be identified based on trending usage in a variety of electronic data sources (e.g., social network feeds, news sources, search queries, etc.). When candidate terms are identified, archives of live or recent speech traffic can be searched to determine whether users are uttering the candidate terms in dictation or speech requests. Such searching can be done using open vocabulary spoken term detection to find phonetic matches in the audio archives. As the candidate terms are found in the speech traffic, notifications can be generated that identify the candidate terms, provide relevant usage statistics, identify the context in which the terms are used, and the like.

Abstract: Techniques described herein relate to allowing users to employ voice-based human-to-computer dialog to program automated assistants with customized routines, or “dialog routines,” that can later be invoked to accomplish task(s). In various implementations, a first free form natural language input—that identifies a command to be mapped to a task and slot(s) required to be filled with values to fulfill the task—may be received from a user. A dialog routine may be stored that includes a mapping between the command and the task, and which accepts, as input, value(s) to fill the slot(s). Subsequent free form natural language input may be received from the user to (i) invoke the dialog routine based on the mapping, and/or (ii) to identify value(s) to fill the slot(s). Data indicative of at least the value(s) may be transmitted to a remote computing device for fulfillment of the task.

Abstract: A method and an apparatus for synthesizing an audio signal are described. A spectral tilt is applied to the code of a codebook used for synthesizing a current frame of the audio signal. The spectral tilt is based on the spectral tilt of the current frame of the audio signal. Further, an audio decoder operating in accordance with the inventive approach is described.

Abstract: A character input device includes a memory that stores one or more conversion candidates, and a processor that executes a process including receiving input of a first character of a word and the number of characters of the word, extracting a conversion candidate including the first character from the one or more conversion candidates stored in the memory according to the number of characters of the word, and displaying the extracted conversion candidate.

Abstract: An intention acquisition method, an electronic device and a computer-readable computer storage are provided.

Abstract: In an audio processing system (300), a filtering section (350, 400): receives subband signals (410, 420, 430) corresponding to audio content of a reference signal (301) in respective frequency subbands; receives subband signals (411, 421, 431) corresponding to audio content of a response signal (304) in the respective subbands; and forms filtered inband references (412, 422, 432) by applying respective filters (413, 423, 433) to the subband signals of the reference signal.

Abstract: Techniques for determining a command or intent likely to be subsequently invoked by a user of a system are described. A user inputs a command (either via a spoken utterance or textual input) to a system. The system determines content responsive to the command. The system also determines a second command or corresponding intent likely to be invoked by the user subsequent to the previous command. Such determination may involve analyzing pairs of intents, with each pair being associated with a probability that one intent of the pair will be invoked by a user subsequent to a second intent of the pair. The system then outputs first content responsive to the first command and second content soliciting the user as to whether the system to execute the second command.

Abstract: In an aspect of the present disclosure, a method for analyzing the speech delivery of a user is disclosed including presenting to the user a plurality of speech delivery analysis criteria, receiving from the user a selection of at least one of the speech delivery analysis criterion, receiving, from at least one sensing device, speech data captured by the at least one sensing device during the delivery of a speech by the user, transmitting the speech data and the selected at least one speech delivery analysis criterion to an analysis engine for analysis based on the selected at least one speech delivery analysis criteria, receiving, from the analysis engine an analysis report for the speech data, the analysis report comprising an analysis of the speech data performed by the analysis engine based on the selected at least one criterion, and presenting to the user the analysis report.