Abstract: A voice synthesis method according to an embodiment includes altering a series of synthesis spectra in a partial period of a synthesis voice based on a series of amplitude spectrum envelope contours of a voice expression to obtain a series of altered spectra to which the voice expression has been imparted, and synthesizing a series of voice samples to which the voice expression has been imparted, based on the series of altered spectra.

Abstract: An embodiment of the inventive concept includes a communication device, a voice input receiving unit, a memory, and a processor. In addition, various embodiments recognized through the specification are possible.

Abstract: In one embodiment, a method includes receiving, from a client system associated with a first user, a first audio input. The method includes generating multiple transcriptions corresponding to the first audio input based on multiple automatic speech recognition (ASR) engines. Each ASR engine is associated with a respective domain out of multiple domains. The method includes determining, for each transcription, a combination of one or more intents and one or more slots to be associated with the transcription. The method includes selecting, by a meta-speech engine, one or more combinations of intents and slots from the multiple combinations to be associated with the first user input. The method includes generating a response to the first audio input based on the selected combinations and sending, to the client system, instructions for presenting the response to the first audio input.

Abstract: A communication system includes a first terminal device, a second terminal device, and an advice providing device. The first terminal device is operated by an operator. The second terminal device is operated by a guest. The second terminal device communicates with the first terminal device through a network. The advice providing device includes circuitry that determines advice for the operator based on voice data including first voice data that is related to the operator and transmitted from the first terminal device and second voice data that is related to the guest and transmitted from the second terminal device. The circuitry of the advice providing device further transmits the advice to the first terminal device. The first terminal device receives the advice and displays, on a display, the advice.

Abstract: An image forming apparatus may include: a storage that stores information in which a job type is associated with speech patterns for processings related to the job type; and a hardware processor that may: be inputted with a speech; acquire a job type; use speech patterns associated with a job type, which is acquired by the hardware processor and is being executed, to analyze a speech inputted during execution of the job; and execute the processings based on an analysis result by the hardware processor.

Abstract: A learning apparatus comprises a learning part that learns an error correction model by a set of a speech recognition result candidate and a correct text of speech recognition for given audio data, wherein the speech recognition result candidate includes a speech recognition result candidate which is different from the correct text, and the error correction model is a model that receives a word sequence of the speech recognition result candidate as input and outputs an error correction score indicating likelihood of the word sequence of the speech recognition result candidate in consideration of a speech recognition error.

Abstract: The present invention discloses a speech interaction method and apparatus, and pertains to the field of speech processing technologies. The method includes: acquiring speech data of a user; performing user attribute recognition on the speech data to obtain a first user attribute recognition result; performing content recognition on the speech data to obtain a content recognition result of the speech data; and performing a corresponding operation according to at least the first user attribute recognition result and the content recognition result, so as to respond to the speech data. According to the present invention, after speech data is acquired, user attribute recognition and content recognition are separately performed on the speech data to obtain a first user attribute recognition result and a content recognition result, and a corresponding operation is performed according to at least the first user attribute recognition result and the content recognition result.

Abstract: One implementation is for a device. The device includes a controller having a memory protection unit, a legally relevant memory portion capable of interacting with the controller, a legally non-relevant memory portion capable of interacting with the controller, an interface in the memory protection unit that allows a privileged application to access the legally relevant memory portion and disallows an unprivileged application to access to the legally relevant portion, and an interrupt system, wherein when the unprivileged application makes an attempt to interact with the legally relevant memory portion, the memory protection unit takes an action associated with the unprivileged application.

Abstract: A system for determining context and intent in a conversation using machine learning (ML) based artificial intelligence (AI) in omnichannel data communications is disclosed. The system may comprise a data store to store and manage data within a network, a server to facilitate operations using information from the one or more data stores, and a ML-based AI subsystem to communicate with the server and the data store in the network. The ML-based AI subsystem may comprise a data access interface to receive data associated with a conversation with a user via a communication channel. The ML-based AI subsystem may comprise a processor to provide a proactive, adaptive, and intelligent conversation by applying hierarchical multi-intent data labeling framework, training at least one model with training data, and generating and deploying a production-ready model based on the trained and retained at least one model.

Abstract: A method includes displaying a first set of text content characterized by a first difficulty level. The method includes obtaining speech data associated with the first set of text content. The method includes determining linguistic feature(s) within the speech data. The method includes in response to completion of the speech data, determining a reading proficiency value associated with the first set of text content and based on the linguistic feature(s). The method includes in accordance with determining the reading proficiency value satisfies change criteria, changing a difficulty level for a second set of text content. After changing the difficulty level, the second set of text content corresponds to a second difficulty level different from the first difficulty level. The method includes in accordance with determining the reading proficiency value does not satisfy the change criteria, maintaining the second set of text content at the first difficulty level.

Abstract: Accommodation for color or visual impairments may be implemented by selective color substitution. A color accommodation module receives an image frame from a host system and generates a color-adapted version of the image frame. The color accommodation module may include a rule based filter that substitutes one or more colors within the image frame with one or more corresponding alternative colors.

Abstract: Aspects of the subject disclosure may include, for example, obtaining a natural language instruction, interpreting the instruction to obtain a machine interpretation, and analyzing the machine interpretation to obtain an intent of the natural language instruction. An actionable command adapted to cause a digital manipulation tool to digitally manipulate a content item is determined according to the intent of the natural language instruction. The actionable command is provided to the digital manipulation tool to obtain the manipulated content item. Other embodiments are disclosed.

Abstract: A dense passage retrieval machine learning model having a first encoder for resources and a second encoder for messages can automatically match relevant resources to computers or sessions based on analysis of a series of messages of an online chat conversation. Continuous re-training is supported based on feedback from a moderator computer and/or user computers.

Abstract: In one embodiment, a method includes receiving, from a client system associated with a first user, a first audio input. The method includes generating multiple transcriptions corresponding to the first audio input based on multiple automatic speech recognition (ASR) engines. Each ASR engine is associated with a respective domain out of multiple domains. The method includes determining, for each transcription, a combination of one or more intents and one or more slots to be associated with the transcription. The method includes selecting, by a meta-speech engine, one or more combinations of intents and slots from the multiple combinations to be associated with the first user input. The method includes generating a response to the first audio input based on the selected combinations and sending, to the client system, instructions for presenting the response to the first audio input.

Abstract: There is provided a speech dialog system that includes a first microphone, a second microphone, a processor and a memory. The first microphone captures first audio from a first spatial zone, and produces a first audio signal. The second microphone captures second audio from a second spatial zone, and produces a second audio signal. The processor receives the first audio signal and the second audio signal, and the memory contains instructions that control the processor to perform operations of a speech enhancement module, an automatic speech recognition module, and a speech dialog module that performs a zone-dedicated speech dialog.

Abstract: A method of providing a platform for configuring device-specific speech recognition is provided. The method includes providing a user interface for developers to select a set of at least two acoustic models appropriate for a specific type of a device, receiving, from a developer, a selection of the set of the at least two acoustic models, and configuring a speech recognition system to perform device-specific speech recognition by using one acoustic model selected from the at least two acoustic models of the set.

Abstract: The present disclosure discloses a method and apparatus for providing a speech service. A specific embodiment of the method comprises: receiving request information sent by a device, the request information comprising first event information and speech information, the first event information used for indicating a first event occurring on the device when the device sends the request information, wherein the first event information comprises speech input event information used for instructing a user to input the speech information; generating response information comprising an operation instruction for a targeted device on the basis of the first event information and the speech information; and sending the response information to the targeted device for the targeted device to perform an operation indicated by the operation instruction. The embodiment improves the efficiency of providing a speech service.

Abstract: Disclosed is a multi-device control method including: performing a voice recognition operation on a voice command generated from a sound source; identifying distances between each of the plurality of devices and the sound source; assigning response rankings to the devices by combining a context-specific correction score of each device corresponding to the voice command and the distances; and selecting a device to respond to the voice command from among the devices according to the response rankings.

Abstract: Embodiments of the present disclosure provide a method and apparatus for man-machine conversion, and an electronic device. The method includes: outputting question information to a user based on a first task of a first conversation scenario; judging, in response to receiving reply information returned by the user, whether to trigger a second conversation scenario based on the reply information; generating, in response to determining the second conversation scenario being triggered based on the reply information, response information corresponding to the reply information based on the second conversation scenario; and outputting the response information to the user.

Abstract: Aspects of the invention include detecting and processing sections spanning processed document partitions by caching a document partition. The document partition includes metadata indicating that the document partition is a portion of a whole document. Aspects also include pairing a candidate paragraph from the document partition with a cached paragraph segment and determining, using a coherence model, a probability that the candidate paragraph and the cached paragraph segment constitute a semantically coherent paragraph. Aspects further include discarding the cached paragraph segment and processing the candidate paragraph and the cached paragraph segment separately based on a determination that the probability is less than a threshold level and processing the candidate paragraph and the cached paragraph segment together as a cross-partition paragraph based on a determination that the probability is greater than the threshold level.

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: Methods and systems are provided for implementing source separation techniques, and more specifically performing source separation on mixed source single-channel and multi-channel audio signals enhanced by inputting lip motion information from captured image data, including selecting a target speaker facial image from a plurality of facial images captured over a period of interest; computing a motion vector based on facial features of the target speaker facial image; and separating, based on at least the motion vector, audio corresponding to a constituent source from a mixed source audio signal captured over the period of interest. The mixed source audio signal may be captured from single-channel or multi-channel audio capture devices. Separating audio from the audio signal may be performed by a fusion learning model comprising a plurality of learning sub-models. Separating the audio from the audio signal may be performed by a blind source separation (“BSS”) learning model.

Abstract: An electronic device and a method for controlling the electronic device is provided. The electronic device includes a microphone, a memory configured to include at least one instruction, and a processor configured to execute the at least one instruction. The processor is configured to control the electronic device to perform voice recognition for an inquiry based on receiving input of a user inquiry through the microphone, and acquire a text for the inquiry, generate a plurality of inquiries for acquiring response data for the inquiry from a plurality of databases using a relation graph indicating a relation between the acquired text and data stored in the plurality of databases, acquire response data corresponding to each of the plurality of inquiries from each of the plurality of databases, and generate a response for the inquiry based on the response data acquired from each of the plurality of databases and output the response.

Abstract: The present teaching relates to generating a conversational agent. In one example, a plurality of input utterances may be received from a developer. A paraphrase model is obtained. The paraphrase model is generated based on machine translation. For each of the plurality of input utterances, one or more paraphrases of the input utterance are generated based on the paraphrase model. For each of the plurality of input utterances, at least one of the one or more paraphrases is selected based on an instruction from the developer to generate selected paraphrases. The conversational agent is generated based on the plurality of input utterances and the selected paraphrases.

Abstract: A keyboard is described that determines, using a first decoder and based on a selection of keys of a graphical keyboard, text. Responsive to determining that a characteristic of the text satisfies a threshold, a model of the keyboard identifies the target language of the text, and determines whether the target language is different than a language associated with the first decoder. If the target language of the text is not different than the language associated with the first decoder, the keyboard outputs, for display, an indication of first candidate words determined by the first decoder from the text. If the target language of the text is different: the keyboard enables a second decoder, where a language associated with the second decoder matches the target language of the text, and outputs, for display, an indication of second candidate words determined by the second decoder from the text.

Abstract: A system and method are disclosed for generating a teleconference space for two or more communication devices using a computer coupled with a database and comprising a processor and memory. The computer generates a teleconference space and transmits requests to join the teleconference space to the two or more communication devices. The computer stores in memory identification information, and audiovisual data associated with one or more users, for each of the two or more communication devices. The computer stores audio transcription data, transmitted to the computer by each of the two or more communication devices and associated with one or more communication device users, in the computer memory. The computer merges the audio transcription data from each of the two or more communication devices into a master audio transcript, and transmits the master audio transcript to each of the two or more communication devices.

Abstract: A method of disambiguating user queries in a multi-turn dialogue include a set of user utterances includes using a predefined language model to recognize an ambiguous entity in an unresolved user utterance from the multi-turn dialogue. The method further includes outputting a clarifying question about the ambiguous entity, and receiving a clarifying user utterance. The method further includes identifying a disambiguating entity in the clarifying user utterance. The method further includes rewriting the unresolved user utterance as a rewritten utterance that replaces the ambiguous entity with the disambiguating entity, and outputting the rewritten utterance to one or more query answering machines.

Abstract: This disclosure describes, in part, techniques implemented by a speech-processing system for providing an extensible skill-interface component to facilitate voice-control of third-party developer devices. The speech-processing system may provide the skill-interface component to third-party device developers using a web-based portal through which the skill interfaces may be created to voice-enable third-party devices having unique capabilities. For instance, a skill interface may define events, such as voice commands of a user, which map to directives configured to cause the third-party devices to perform an operation that is responsive to the event. In this way, the speech-processing system may receive audio data representing a voice command of a user in an environment of a third-party device, and return a directive to cause the third-party device to perform an operation responsive to the voice command.

Abstract: A system and method for dictation using a peripheral device includes a voice recognition mouse. The voice recognition mouse includes a microphone, a first button, a processor coupled to the microphone and the first button, and a memory coupled to the processor. The memory stores instructions that, when executed by the processor, cause the processor to detect actuation of the first button and in response to detecting actuation of the first button, invoke the microphone for capturing audio speech from a user. The captured audio speech is streamed to a first module. The first module is configured to invoke a second module for converting the captured audio speech into text and forward the text to the first module for providing to an application expecting the text, the application being configured to display the text on a display device.

Abstract: There is a method and system for translating speech to text, the speech having been received by a client device. A user utterance corresponding to the speech is received. A first predicted text corresponding to the user utterance and a first confidence score corresponding to the first predicted text are determined using a local graph. The user utterance is transmitted to a server. A second predicted text corresponding to the user utterance and a second confidence score corresponding to the second predicted text is received from the server. If the first confidence score is higher than the second confidence score, the first predicted text is output.

Abstract: A system that outputs information generated by summarizing contents of voices and images as texts. A CPU of the system performs, according to a program stored in a memory, recording voice data and captured image data, generating first text information by performing speech recognition on the acquired voice data, generating second text information by performing character recognition on the acquired image data, and generating summary text information smaller in the number of characters than the first text information and the second text information, based on the first text information and the second text information, according to a predetermined criterion.

Abstract: A model-pair is selected to recognize spoken words in a speech signal generated from a speech, which includes an acoustic model and a language model. A degree of disjointedness between the acoustic model and the language model is computed relative to the speech by comparing a first recognition output produced from the acoustic model and a second recognition output produced from the language model. When the acoustic model incorrectly recognizes a portion of the speech signal as a first word and the language model correctly recognizes the portion of the speech signal as a second word, a textual representation of the second word is determined and associated with a set of sound descriptors to generate a training speech pattern. Using the training speech pattern, the acoustic model is trained to recognize the portion of the speech signal as the second word.

Abstract: Approaches for interpretable counting for visual question answering include a digital image processor, a language processor, and a counter. The digital image processor identifies objects in an image, maps the identified objects into an embedding space, generates bounding boxes for each of the identified objects, and outputs the embedded objects paired with their bounding boxes. The language processor embeds a question into the embedding space. The scorer determines scores for the identified objects. Each respective score determines how well a corresponding one of the identified objects is responsive to the question. The counter determines a count of the objects in the digital image that are responsive to the question based on the scores. The count and a corresponding bounding box for each object included in the count are output. In some embodiments, the counter determines the count interactively based on interactions between counted and uncounted objects.

Abstract: A system and method for decorrelating audio data. A method includes determining a plurality of propagation vectors for each of a plurality of sound sources based on audio data captured by a plurality of sound capturing devices and a location of each of the plurality of sound sources, wherein the plurality of sound sources and the plurality of sound capturing devices are deployed in a space, wherein the audio data is captured by the plurality of sound capturing devices based on sounds emitted by the plurality of sound sources in the space; determining a plurality of beam former outputs, wherein each beam former output is determined for one of the plurality of sound sources; determining a decoupling matrix based on the plurality of beam former outputs and the propagation vectors; and decorrelating audio data captured by the plurality of sound capturing devices based on the decoupling matrix.

Abstract: A voice identification method, device, apparatus, and a storage medium are provided. The method includes: receiving voice data; and performing a voice identification on the voice data, to obtain first text data associated with the voice data; determining common text data in a preset fixed data table, wherein a similarity between a pronunciation of the determined common text data and a pronunciation of the first text data meets a preset condition, wherein the determined common text data is a voice identification result with an occurrence number larger than a first preset threshold; and replacing the first text data with the determined common text data.

Abstract: Systems and methods for processing speech are described. In certain examples, image data is used to generate visual feature tensors and audio data is used to generate audio feature tensors. The visual feature tensors and the audio feature tensors are used by a linguistic model to determine linguistic features that are usable to parse an utterance of a user. The generation of the feature tensors may be jointly configured with the linguistic model. Systems may be provided in a client-server architecture.

Abstract: A processing device is to: identify, using digital interview data of interviewees captured during interviews, a subset of the interviewees that have a disability; label a first group of the interviewees as disabled and a second group of the interviewees as not disabled with reference to the disability; identify features from the digital interview data for the first group that correlate with the disability; formulate a digital fingerprint of the features that identifies how the first group differs from the second group with reference to the disability; map the digital fingerprint of the features onto a dataset of an interviewee belonging to the second group of the interviewees, to generate a mapped dataset; and determine, from the mapped dataset, effects of the digital fingerprint on a job performance score for the interviewee.

Abstract: A method, computer program product, and computing system for monitoring a plurality of conversations within a monitored space to generate a conversation data set; processing the conversation data set using machine learning to: define a system-directed command for an ACI system, and associate one or more conversational contexts with the system-directed command; detecting the occurrence of a specific conversational context within the monitored space, wherein the specific conversational context is included in the one or more conversational contexts associated with the system-directed command; and executing, in whole or in part, functionality associated with the system-directed command in response to detecting the occurrence of the specific conversational context without requiring the utterance of the system-directed command and/or a wake-up word/phrase.

Abstract: This speech processing device is provided with: a contribution degree estimation means which calculates a contribution degree representing a quality of a segment of the speech signal; and a speaker feature calculation means which calculates a feature from the speech signal, for recognizing attribute information of the speech signal, using the contribution degree as a weight of the segment of the speech signal.

Abstract: Method, apparatus, and computer program product are provided for customizing an automatic closed captioning system. In some embodiments, at a data use (DU) location, an automatic closed captioning system that includes a base model is provided, search criteria are defined to request from one or more data collection (DC) locations, a search request based on the search criteria is sent to the one or more DC locations, relevant closed caption data from the one or more DC locations are received responsive to the search request, the received relevant closed caption data are processed by computing a confidence score for each of a plurality of data sub-sets of the received relevant closed caption data and selecting one or more of the data sub-sets based on the confidence scores, and the automatic closed captioning system is customized by using the selected one or more data sub-sets to train the base model.

Abstract: Technology of the disclosure may facilitate user discovery of various voice-based action queries that can be spoken to initiate computer-based actions, such as voice-based action queries that can be provided as spoken input to a computing device to initiate computer-based actions that are particularized to content being viewed or otherwise consumed by the user on the computing device. Some implementations are generally directed to determining, in view of content recently viewed by a user on a computing device, at least one suggested voice-based action query for presentation via the computing device. Some implementations are additionally or alternatively generally directed to receiving at least one suggested voice-based action query at a computing device and providing the suggested voice-based action query as a suggestion in response to input to initiate providing of a voice-based query via the computing device.

Abstract: A system, method and non-transitory computer readable medium for providing call quality driven communication management wherein an audio data stream of a communication session having one or more utterances is processed to generate a transcript of the communication session. The generated transcript is analyzed to determine whether a quality of the audio data stream, and one or more quality improvement measures when one or more audio artifacts are determined to be present in the audio data stream.

Abstract: An electronic device for performing a voice recognition and a controlling method are provided.

Abstract: Systems and processes for operating an intelligent automated assistant to provide extension of digital assistant services are provided. An example method includes, at an electronic device having one or more processors, receiving, from a first user, a first speech input representing a user request. The method further includes obtaining an identity of the first user; and in accordance with the user identity, providing a representation of the user request to at least one of a second electronic device or a third electronic device. The method further includes receiving, based on a determination of whether the second electronic device or the third electronic device, or both, is to provide the response to the first electronic device, the response to the user request from the second electronic device or the third electronic device. The method further includes providing a representation of the response to the first user.

Abstract: Disclosed herein is a method for intelligently recognizing voice by a voice recognizing apparatus in various noise environments. The method includes acquiring a first noise level for an environment in which the voice recognizing apparatus is located, inputting the first noise level into a previously learned noise-sensitivity model to acquire a first optimum sensitivity, and recognizing a user's voice based on the first optimum sensitivity. The noise-sensitivity model is learned in a plurality of noise environments acquiring different noise levels, so that it is possible to accurately acquire an optimum sensitivity corresponding to a noise level depending on an operating state when an IoT device (voice recognizing apparatus) is in operation.

Abstract: According to one embodiment, an electronic device includes a transceiver and a hardware processor. The transceiver establishes a wired connection or a wireless connection between the electronic device and a wearable device wearable by a user. The hardware processor receives voice data generated by collecting speech of the user from the wearable device. The hardware processor performs a process to recognize one of one or more first commands from the voice data when a first mode is set as a voice recognition mode, performs a process to recognize one of second commands from the voice data when a second mode is set as the voice recognition mode, and repeatedly performs a process to recognize one of third commands from the voice data when a third mode is set as the voice recognition mode.

Abstract: A method and system for modifying an audible command is provided. The method includes continuously receiving audible commands associated with a context of interactions between a user and individuals. The audible commands are analyzed with respect to associated actions and user attributes of the audible commands are identified. Specified information required for executing each command of the audible commands and portions of the specified information associated with specified individuals of the individuals are determined. Digital audio samples of the user are retrieved and assigned to the portions of the specified information with respect to each command. The associated actions are modified with respect to the specified individuals and self-learning software code comprising the modified actions is generated and executed such that the commands are executed with respect to the modified actions.

Abstract: Mechanisms, in a Question Answering (QA) system comprising a processor and a memory, for evaluating a hypothetical link in an ontology are provided. An initial analysis of the ontology is performed to identify a set of information concept entities and links between information concept entities in the ontology. The hypothetical link between a first information concept entity and a second information concept entity in the ontology is generated based on the initial analysis of the ontology. Natural language questions corresponding to the hypothetical link are processed to generate answer results directed to a plurality of links between a plurality of information concept entities. The answer results are aggregated across the plurality of links to determine an aggregate answer result for the hypothetical link. An indication of whether or not the hypothetical link is a valid link is output based on the aggregate answer result for the hypothetical link.

Abstract: Techniques for labeling user inputs for updating user recognition voice profiles are described. A system may leverage various signals, generated during or after processing of a user input, to retroactively determine which user spoke the user input. For example, after the system receives the user input, the user may provide the system with non-spoken user verification information. Based on such user verification information, the system may label the previously spoken user input as originating from the particular user. The system may also or alternatively use system usage history to retroactively label user inputs.

Abstract: Systems, methods, and computer-readable storage media for providing for intelligent switching of languages and/or pronunciations in a text-to-speech system. As the system receives text, the text is analyzed to identify portions which should have speech constructed using a pronunciation distinct from the remaining portions of the text. The text-to-speech system uses multiple pronunciation dictionaries to generate and produce speech corresponding to the text, where the identified portions of the text are in a different language or have a different accent from the remainder of the text. Having generated speech corresponding to the text in multiple languages, accents, or dialects, the system combines the portions, then communicates the speech to the text recipient.