Abstract: One embodiment provides a method, including: receiving, at an input device associated with an information handling device, audio input; determining, using a processor, that an audible anomaly exists in the audio input, wherein the audible anomaly corresponds to a deviation from an established speech input pattern of a user; and performing, responsive to determining that the audible anomaly exists in the audio input, a remedial action to address the audible anomaly. Other aspects are described and claimed.

Abstract: System and method for operating an always-on ASR (automatic speech recognition) system by selecting target keywords and continuously detecting the selected target keywords in voice commands in a mobile device are provided. In the mobile device, a processor is configured to collect keyword candidates, collect usage frequency data for keywords in the keyword candidates, collect situational usage frequency data for the keywords in the keyword candidates, select target keywords from the keyword candidates based on the usage frequency data and the situational usage frequency data, and detect one or more of the target keywords in a voice command using continuous detection of the target keywords.

Abstract: A system and method for authenticating sound verbalized or otherwise generated by a live source within a monitored setting for voice-controlled or sound-controlled automation of a responsive process. One or more classifiers each generate a decision value according to values of predetermined signal features extracted from a received digital stream, and a sound type classification is computed according to an aggregate score of a predetermined number of decision values. The actuation of the responsive process is authenticated when the system discriminately indicates the captured sound signals to be verbalized or generated by a live source. The responsive process is thereby suppressed when the sound is instead determined to be reproduced or otherwise previously transduced, for example by a transmission or recording.

Abstract: The present document relates to a method of layered encoding of a compressed sound representation of a sound or sound field. The compressed sound representation comprises a basic compressed sound representation comprising a plurality of components, basic side information for decoding the basic compressed sound representation to a basic reconstructed sound representation of the sound or sound field, and enhancement side information including parameters for improving the basic reconstructed sound representation.

Abstract: [Object] Technology is provided to enable a mobile terminal to function as a digital assistant even when the mobile terminal is in a state where it cannot communicate with a server apparatus. [Solution] When a user terminal 200 receives a query A from a user, user terminal 200 sends query A to a server 100. Server 100 interprets the meaning of query A using a grammar A. Server 100 obtains a response to query A based on the meaning of query A and sends the response to user terminal 200. Server 100 further sends grammar A to user terminal 200. That is, server 100 sends to user terminal 200 a grammar used to interpret the query received from user terminal 200.

Abstract: A computer implemented method classifies an input corresponding to multiple different kinds of input. The method includes obtaining a set of features from the input, providing the set of features to multiple different models to generate state predictions, generating a set of state-dependent predicted weights, and combining the state predictions from the multiple models, based on the state-dependent predicted weights for classification of the set of features.

Abstract: A method for operating a voice trigger is provided. In some implementations, the method is performed at an electronic device including one or more processors and memory storing instructions for execution by the one or more processors. The method includes receiving a sound input. The sound input may correspond to a spoken word or phrase, or a portion thereof. The method includes determining whether at least a portion of the sound input corresponds to a predetermined type of sound, such as a human voice. The method includes, upon a determination that at least a portion of the sound input corresponds to the predetermined type, determining whether the sound input includes predetermined content, such as a predetermined trigger word or phrase. The method also includes, upon a determination that the sound input includes the predetermined content, initiating a speech-based service, such as a voice-based digital assistant.

Abstract: Embodiments of the innovation relate to, in a contact center apparatus, a method for recognizing user intent associated with user interaction with the contact center apparatus.

Abstract: Systems and methods for providing multi-source based knowledge data for Artificial Intelligence (AI) characters are provided. An example method includes providing a plurality of data sources; receiving, from a user, at least one word during a conversation between the user and an AI character; ascertaining a speech style of the AI character; analyzing the at least one word to determine a type of information needed to generate a reply to the user; selecting, based on the type of information, at least one data source from the plurality of data sources; generating, based on the at least one word, one or more queries; sending the one or more queries to the at least one data source; receiving one or more responses from the at least one data source; forming, based on the one or more responses and the speech style of the AI character, the reply for providing to the user.

Abstract: The exemplary embodiments disclose a method, a computer program product, and a computer system for managing user commands. The exemplary embodiments may include a user giving one or more commands to one or more devices, collecting data of the one or more commands, extracting one or more features from the collected data, and determining which one or more of the commands should be executed on which one or more of the devices based on the extracted one or more features and one or more models.

Abstract: Embodiments presented herein describe techniques for generating a linguistic model of input data obtained from a data source (e.g., a video camera). According to one embodiment of the present disclosure, a sequence of symbols is generated based on an ordered stream of normalized vectors generated from the input data. A dictionary of words is generated from combinations of the ordered sequence of symbols based on a frequency at which combinations of symbols appear in the ordered sequence of symbols. A plurality of phrases is generated based an ordered sequence of words from the dictionary observed in the ordered sequence of symbols based on a frequency by which combinations of words in ordered sequence of words appear relative to one another.

Abstract: A system and methods are provided to analyze audio signals from an incoming voice call. The system includes a processor and a computer readable medium operably coupled thereto, to perform voice analysis operations which include receiving a first audio signal comprising a first audio waveform of a first speech between at least two users during the incoming voice call, accessing speech segment parameters for analyzing the audio signals, determining one or more talk-over segments in the first audio waveform using the speech segment parameters, extracting audio features from each of the one or more talk-over segments, determining, using a machine learning (ML) model trained for interruption analysis of the audio signals, whether each of the one or more talk-over segments are a negative interruption or a non-negative interruption based on the audio features, and determining whether to output a first notification for the negative interruption or the non-negative interruption.

Abstract: Systems and methods to translate a spoken command to a selection sequence are disclosed. Exemplary implementations may: obtain audio information representing sounds captured by a client computing platform; analyze the sounds to determine spoken terms; determine whether the spoken terms include one or more of the terms that are correlated with the commands; responsive to determining that the spoken terms are terms that are correlated with a particular command stored in the electronic storage, perform a set of operations that correspond to the particular command; responsive to determine that the spoken terms are not the terms correlated with the commands stored in the electronic storage, determining a selection sequence that causes a result subsequent to the analysis of the sounds; correlate the spoken terms with the selection sequence; store the correlation of the spoken terms with the selection sequence; and perform the selection sequence to cause the result.

Abstract: Devices and techniques are generally described for learning personalized concepts for natural language processing. In various examples, a first natural language input may be received. In some examples, a determination may be made that the first natural language input comprises non-actionable slot data. A dialog session may be initiated with the user. In some examples, first slot data that is indicated by the user during the dialog session may be determined. In various examples, data representing the first slot data may be stored in a database in association with the first natural language input.

Abstract: The present disclosure relates to detecting the use of fake voice command to activate microphones of smart devices. In one embodiment, sound characteristics associated with an audio signal from a microphone of smart device may be compared with other microphones of the smart device in order to detect fake voice commands. In another embodiment, sound characteristics associated with the audio signal from the microphone may be compared with a threshold range of stored sound characteristics in order to detect fake voice commands. In some embodiments, a controller may triangulate a position associated with a source of a sound in order to detect a fake voice command. In a further embodiment, a controller may verify that a user or associated electronic device are near a smart device to authorize a voice command.

Abstract: Methods and systems are provided for assisting operation of a vehicle using speech recognition. One method involves automatically identifying an input element based at least in part on an audio communication with respect to the vehicle, identifying one or more constraints associated with the input element, obtaining a limited command vocabulary for the input element using the one or more constraints, and automatically constructing a contextual speech recognition graph for the input element prior to user selection of the input element using the limited command vocabulary. Thereafter, subsequently received audio input is recognized using the contextual speech recognition graph that was automatically and prospectively generated.

Abstract: Some implementations are directed to adapting a client application on a feature phone based on experiment parameters. Some of those implementations are directed to adapting an assistant client application, where the assistant client application interacts with remote assistant component(s) to provide automated assistant functionalities via the assistant client application of the feature phone. Some implementations are additionally or alternatively directed to determining whether an invocation, of an assistant client application on a feature phone, is a request for transcription of voice data received in conjunction with the invocation, or is instead a request for an assistant response that is responsive to the transcription of the voice data (e.g., includes assistant content that is based on and in addition to the transcription, and that optionally lacks the transcription itself).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for hotword detection on multiple devices are disclosed. In one aspect, a method includes the actions of receiving, by a computing device, audio data that corresponds to an utterance. The actions further include determining a likelihood that the utterance includes a hotword. The actions further include determining a loudness score for the audio data. The actions further include based on the loudness score, determining an amount of delay time. The actions further include, after the amount of delay time has elapsed, transmitting a signal that indicates that the computing device will initiate speech recognition processing on the audio data.

Abstract: The present document relates to a method of layered encoding of a frame of a compressed higher-order Ambisonics, HOA, representation of a sound or sound field. The compressed HOA representation comprises a plurality of transport signals. The method comprises assigning the plurality of transport signals to a plurality of hierarchical layers, the plurality of layers including a base layer and one or more hierarchical enhancement layers, generating, for each layer, a respective HOA extension payload including side information for parametrically enhancing a reconstructed HOA representation obtainable from the transport signals assigned to the respective layer and any layers lower than the respective layer, assigning the generated HOA extension payloads to their respective layers, and signaling the generated HOA extension payloads in an output bitstream.

Abstract: Systems and methods for natural language generation by an edge computing device are disclosed. In one embodiments, a method comprises: receiving, by an edge computing device, event data from an edge event; determining, by the edge computing device, that a network connection to a cloud server is not available; extracting, by the edge computing device, features of the event data; predicting, by a local neural network of the edge computing device, an action for the edge computing device to take based on the features of the event data, wherein the action is associated with a confidence level; and determining, by the edge computing device, whether the confidence level meets a predetermined threshold value.