Abstract: A method includes receiving multiple samples of time-domain data that includes noise, computing a first two-dimensional (2D) time-frequency representation of the time domain data, and processing the first time-frequency representation using a time-frequency noise reduction mask to generate a second, noise-reduced time-frequency representation of the time domain data. The method also includes generating a time domain output based on the noise-reduced time-frequency representation.

Abstract: A device may receive a set of audio data files corresponding to a set of calls, wherein the set of audio data files includes digital representations of one or more segments of respective calls of the set of calls, and wherein the set of calls includes audio data relating to a particular industry. The device may receive a set of transcripts corresponding to the set of audio data files. The device may determine a plurality of text-audio pairs within the set of calls, wherein a text-audio pair, of the plurality of text-audio pairs, comprises: a digital representation of a segment a call of the set of calls, and a corresponding excerpt of text from the set of transcripts. The device may train, using a machine learning process, an industry-specific text-to-speech model, tailored for the particular industry, based on the plurality of text-audio pairs.

Abstract: An electronic device includes an audio input module, an audio output module, and a processor. The processor is configured to provide a first signal and a second signal into which a first audio signal is processed, output the first audio signal through the audio output module, acquire an external audio signal comprising the first audio signal of the electronic device, acquire a first output value through a first input channel of an audio filter, acquire a second output value through a second input channel of the audio filter, and provide a second audio signal, based at least on a first difference value between the magnitude value corresponding to the first frequency of the external audio signal and the first output value and a second difference value between the magnitude value corresponding to the second frequency of the external audio signal and the second output value.

Abstract: A soft decision audio decoding system for preserving audio continuity in a digital wireless audio receiver is provided that deduces the likelihood of errors in a received digital signal, based on generated hard bits and soft bits. The soft bits may be utilized by a soft audio decoder to determine whether the digital signal should be decoded or muted. The soft bits may be generated based on a degree of closeness of a detected phase trajectory to known legal phase trajectories determined from the running the phase trajectory through a soft-output Viterbi algorithm. The value of the soft bits may indicate confidence in the strength of the hard bit generation. The soft decision audio decoding system may infer errors and decode perceptually acceptable audio without requiring error detection, as in conventional systems, as well as have low latency and improved granularity.

Abstract: Particular embodiments described herein provide for an electronic device that can be configured to receive a verbal command to active a device with an unknown label, derive a probable device and a label for the probable device, activate the probable device, determine that the activated probable device is the same device to be activated by the verbal command, and store the label and a description for the device. In some examples, the label is associated with the description.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating discrete latent representations of input audio data. Only the discrete latent representation needs to be transmitted from an encoder system to a decoder system in order for the decoder system to be able to effectively to decode, i.e., reconstruct, the input audio data.

Abstract: The present disclosure provides systems and methods that perform machine-learned natural language processing. A computing system can include a machine-learned natural language processing model that includes an encoder model trained to receive a natural language text body and output a knowledge graph and a programmer model trained to receive a natural language question and output a program. The computing system can include a computer-readable medium storing instructions that, when executed, cause the processor to perform operations. The operations can include obtaining the natural language text body, inputting the natural language text body into the encoder model, receiving, as an output of the encoder model, the knowledge graph, obtaining the natural language question, inputting the natural language question into the programmer model, receiving the program as an output of the programmer model, and executing the program on the knowledge graph to produce an answer to the natural language question.

Abstract: A language proficiency analyzer automatically evaluates a person's language proficiency by analyzing that person's oral communications with another person. The analyzer first enhances the quality of an audio recording of a conversation between the two people using a neural network that automatically detects loss features in the audio and adds those loss features back into the audio. The analyzer then performs a textual and audio analysis on the improved audio. Through textual analysis, the analyzer uses a multi-attention network to determine how focused one person is on the other and how pleased one person is with the other. Through audio analysis, the analyzer uses a neural network to determine how well one person pronounced words during the conversation.

Abstract: In one embodiment, a method includes, by one or more computing systems, receiving, from a client system associated with a first user, a first user input from the first user, identifying one or more entities referenced by the first user input, determining a classification of the first user input based on a machine-learning classifier model, generating several candidate conversational fillers based on the classification of the first user input and the one or more identified entities, wherein each candidate conversational filler references at least one of the one or more identified entities, ranking the candidate conversational fillers based on a relevancy of the candidate conversational filler to the first user input and a decay model hysteresis, and sending instructions for presenting a top-ranked candidate conversational filler as an initial response to the first user.

Abstract: One embodiment provides a method in which an audible command is first received from a user. Subsequent to receipt of the audible command, one or more sensors may detect certain contextual data associated with the user's physical surroundings. An embodiment may then determine, using the data, whether a default output response associated with the audible command is appropriate with respect to the user's physical surroundings. If the default output response is determined not to be appropriate, an embodiment may thereafter provide an alternative output response that is appropriate with respect to the user's surroundings. Other aspects are described and claimed.

Abstract: Systems and method of diarization of audio files use an acoustic voiceprint model. A plurality of audio files are analyzed to arrive at an acoustic voiceprint model associated to an identified speaker. Metadata associate with an audio file is used to select an acoustic voiceprint model. The selected acoustic voiceprint model is applied in a diarization to identify audio data of the identified speaker.

Abstract: Embodiments are disclosed for transliterating text entries across different script systems. A method according to some embodiments includes steps of: receiving an input string in a first script system input using a keyboard; segmenting, using a probabilistic model, the input string into phonemes that correspond to characters or sets of characters in a second script system; converting the phonemes in the first script system into the characters or sets of characters in the second script system, the characters or sets of characters forming a word or a word prefix in the second script system; and outputting the word or the word prefix in the second script system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating responses using task-independent conversational systems are provided. In one example method, a response to a user text input is generated by updating a state of the conversation based on the user text input, generating a conversational (task-independent) output, and determining whether to provide a conversational response based on the conversational output, or to additionally generate a task-specific output.

Abstract: In an example embodiment, text is received at an ecommerce service from a first user, the text in a first language and pertaining to a first listing on the ecommerce service. Contextual information about the first listing may be retrieved. The text may be translated to a second language. Then, a plurality of text objects, in the second language, similar to the translated text may be located in a database, each of the text objects corresponding to a listing. Then, the plurality of text objects similar to the translated text may be ranked based on a comparison of the contextual information about the first listing and contextual information stored in the database for the listings corresponding to the plurality of text objects similar to the translated text. At least one of the ranked plurality of text objects may then be translated to the first language.

Abstract: Systems and methods for training a control panel to recognize user defined and preprogrammed sound patterns are provided. Such systems and methods can include the control panel operating in a learning mode, receiving initial ambient audio from a region, and saving the initial ambient audio as an audio pattern in a memory device of the control panel. Such systems and methods can also include the control panel operating in an active mode, receiving subsequent ambient audio from the region, using an audio classification model to make an initial determination as to whether the subsequent ambient audio matches or is otherwise consistent with the audio pattern, determining whether the initial determination is correct, and when the control panel determines that the initial determination is incorrect, modifying or updating the audio classification model for improving the accuracy in detecting future consistency with the audio pattern.

Abstract: A text entity extraction method, apparatus, and storage medium are provided. The method includes determining candidate text entities in a target text. Portions of the candidate text entities are combined to generate candidate segmentation combinations corresponding to the target text, the candidate text entities in each candidate segmentation combination being different. A combination probability corresponding to each candidate segmentation combination is calculated, where the combination probability is a probability that grammar is correct when the target text uses the candidate segmentation combination. A target segmentation combination corresponding to the target text is determined according to the combination probabilities. A text entity is extracted from the target text according to the target segmentation combination.

Abstract: The technology disclosed relates to retrieving a personal memo from a database. The method includes receiving, by a virtual assistant, a natural language utterance that expresses a request, interpreting the natural language utterance according to a natural language grammar rule for retrieving memo data from the natural language utterance, the natural language grammar rule recognizing query information, responsive to interpreting the natural language utterance, using the query information to query the database for a memo related to the query information, and providing, to a user, a response generated in dependence upon the memo related to the query information.

Abstract: In speech processing systems personalization is added in the Natural Language Understanding (NLU) processor by incorporating external knowledge sources of user information to improve entity recognition performance of the speech processing system. Personalization in the NLU is effected by incorporating one or more dictionaries of entries, or gazetteers, with information personal to a respective user, that provide the user's information to permit disambiguation of semantic interpretation for input utterances to improve quality of speech processing results.

Abstract: Systems and methods for enabling voice-based interactions with electronic devices can include a data processing system maintaining a plurality of device action data sets and a respective identifier for each device action data set. The data processing system can receive, from an electronic device, an audio signal representing a voice query and an identifier. The data processing system can identify, using the identifier, a device action data set. The data processing system can identify a device action from device action data set based on content of the audio signal. The data processing system can then identify, from the device action dataset, a command associated with the device action and send the command to the for execution device for execution.

Abstract: Methods and apparatus for identifying a music service based on a user command. A content type is identified from a received user command and a music service is selected that supports the content type. A selected music service can then transmit audio content associated with the content type for playback.