Abstract: A component management server computer (“server”) and processing methods are disclosed. In some embodiments, the server is programmed to continuously receive input data regarding what is happening in the physical room from one or more input devices. The server is programmed to then detect an utterance of a spoken word from the input data and generate one or more sound metrics based on the input data. Based on the sound metrics as applied to certain criteria, the server is programmed to activate a component, such as an input device, variable, software system, or output device, and cause one or more output devices to execute an action that alerts a user of the activated component. The server can also be programmed to turn on, off, up, or down any of the components based on the activated component.

Abstract: Determining whether, upon cessation of a second automated assistant session that interrupted and supplanted a prior first automated assistant session: (1) to automatically resume the prior first automated assistant session, or (2) to transition to an alternative automated assistant state in which the prior first session is not automatically resumed. Implementations further relate to selectively causing, based on the determining and upon cessation of the second automated assistant session, either the automatic resumption of the prior first automated assistant session that was interrupted, or the transition to the state in which the first session is not automatically resumed.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for modulating language model biasing. In some implementations, context data is received. A likely context associated with a user is determined based on at least a portion of the context data. One or more language model biasing parameters based at least on the likely context associated with the user is selected. A context confidence score associated with the likely context based on at least a portion of the context data is determined. One or more language model biasing parameters based at least on the context confidence score is adjusted. A baseline language model based at least on the one or more of the adjusted language model biasing parameters is biased. The baseline language model is provided for use by an automated speech recognizer (ASR).

Abstract: The present disclosure describes a system and method for providing dynamic remediation of a pluggable streaming device issue, such as a customer premises equipment (CPE) device. Sometimes, various features of the CPE device to begin to fail. For example, synchronization of the audio and video streams may drift, media rental purchases may time out, or playback may throttle to low quality. Such failures can be caused by device or network issues. The present disclosure describes a CPE remediation system that operates to identify a failure associated with playing media streamed by the CPE device. The CPE remediation system may further determine a solution to remediate an observed CPE device-related failure. In some examples, the CPE remediation process may further provide or perform one or more actions included in the determined solution. In some examples, the solution may include a warm or a cold reboot.

Abstract: A tag characterizing a portion of a multi-view interactive digital media representation (MVIDMR) may be determined by applying a grammar to natural language data. The MVIDMR may include images of an object and may be navigable in one or more dimensions. An object model location for the tag identifying a location within a three-dimensional object model may be determined by applying the grammar to the natural language data. The tag may then be applied to the MVIDMR by associating it with two or more of the images at positions determined based on the object model location.

Abstract: In an approach for enabling a user to visualize a transcript of a discussion via a head-mounted AR device and to selectively copy one or more parts of the transcript that is contextually relevant for inclusion in a new document file or a previously created document file, a processor captures audio of a spoken content of a first participant of a discussion via an AR device worn by a user. A processor analyzes the audio of the spoken content of the first participant. A processor converts the audio of the spoken content of the first participant to text to create a transcript. A processor creates a visualization of the transcript. A processor presents the visualization of the transcript to the user via the AR device. A processor enables the user to copy one or more parts of the transcript into a document file via a selection support.

Abstract: The present invention relates to an image output device mounted on a vehicle to implement augmented reality. One or more of an autonomous driving vehicle, a user terminal, and a server of the present invention can be linked to an artificial intelligence module, a drone (unmanned aerial vehicle, UAV), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to 5G services, etc.

Abstract: A method includes training a set of teacher models. Training the set of teacher models includes, for each individual teacher model of the set of teacher models, training the individual teacher model to transcribe unlabeled audio samples and predict a pseudo labeled dataset having multiple labels. At least some of the unlabeled audio samples contain named entity (NE) audio data. At least some of the labels include transcribed NE labels corresponding to the NE audio data. The method also includes correcting at least some of the transcribed NE labels using user-specific NE textual data. The method further includes retraining the set of teacher models based on the pseudo labeled dataset from a selected one of the teacher models, where the selected one of the teacher models predicts the pseudo labeled dataset more accurately than other teacher models of the set of teacher models.

Abstract: A method for detecting business intent from a business intent corpus by employing an Intent Detection via Multi-hop Unified Syntactic Graph (IDMG) is presented. The method includes parsing each text sample representing a business need description to extract syntactic information including at least tokens and words, tokenizing the words of the syntactic information to generate sub-words for each of the words by employing a multi-lingual pre-trained language model, aligning the generated sub-words to the tokens of the syntactic information to match ground-truth intent actions and objects to the tokenized sub-words, generating a unified syntactic graph, encoding, via a multi-hop unified syntactic graph encoder, the unified syntactic graph to generate an output, and predicting an intent action and object from the output.

Abstract: A method of detecting live speech comprises: receiving a signal containing speech; obtaining a first component of the received signal in a first frequency band, wherein the first frequency band includes audio frequencies; and obtaining a second component of the received signal in a second frequency band higher than the first frequency band. Then, modulation of the first component of the received signal is detected; modulation of the second component of the received signal is detected; and the modulation of the first component of the received signal and the modulation of the second component of the received signal are compared. It may then be determined that the speech may not be live speech, if the modulation of the first component of the received signal differs from the modulation of the second component of the received signal.

Abstract: A mobile communication environment (100) can include a mobile device (160) to measure and send sound pressure level data. The mobile device (160) can initiate the collection of audio information responsive to detecting a trigger event. Mobile device (160) can measure or calculate the sound pressure level from the audio information. Metadata including time information and geographic location information can be captured with the collected audio information. Mobile device (160) can send the sound pressure level data and metadata through a wired or wireless communication path to a database (614).

Abstract: Embodiments provide a computer-executable method, computer system and non-transitory computer-readable medium for programmatically analyzing a consumer review. The method includes programmatically accessing, via a network device, one or more consumer reviews for a commercial entity or a commercial object. The method also includes executing a consumer review processing engine to programmatically identify an attribute descriptor in the one or more consumer reviews, and executing the consumer review processing engine to programmatically generate a sentiment score associated with the one or more consumer reviews. The method further includes storing, on a non-transitory computer-readable storage device, the attribute descriptor and the sentiment score in association with the commercial entity or the commercial object.

Abstract: Methods, apparatuses and computer program products for providing a conversational data-to-text system are described herein. An example method may include receiving a first natural language query from a client device; generating a first analytic operation instruction associated with a multi-dimensional dataset based at least in part on the first natural language query; determining a first multi-dimensional data object based at least in part on the first analytic operation instruction and the multi-dimensional dataset; generating a first natural language response to the first natural language query based at least in part on the first multi-dimensional data object; and transmitting the first natural language response to the client device.

Abstract: The unique attributes of handheld devices and how they are used—particularly multi-screen devices—are leveraged to define rules for automatically triggering workflows. By monitoring signals from various device sensors, the device can anticipate a user's intention to perform an action, such as capturing a quick thought. A workflow for performing the action (or actions) may be automatically triggered based on rules for evaluating the sensor signals. By anticipating the user's intentions, the device can automatically perform many of the underlying actions behind the scenes, thereby minimizing the actions performed by the user and improving the user experience. In this way, cumbersome, multi-step user inputs and interactions are avoided by anticipating user intentions and automatically triggering workflows.

Abstract: An ASR model includes a first encoder configured to receive a sequence of acoustic frames and generate a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The ASR model also includes a second encoder configured to receive the first higher order feature representation generated by the first encoder at each of the plurality of output steps and generate a second higher order feature representation for a corresponding first higher order feature frame. The ASR model also includes a decoder configured to receive the second higher order feature representation generated by the second encoder at each of the plurality of output steps and generate a first probability distribution over possible speech recognition hypothesis. The ASR model also includes a language model configured to receive the first probability distribution over possible speech hypothesis and generate a rescored probability distribution.

Abstract: A system for implicit authentication for a mobile device associated with a user, wherein the implicit authentication is behavioral, biometric and task-based and includes at least one authentication task selected so as to leverage the user's muscle memory. The mobile device comprises a touchscreen; a transaction authentication information unit; one or more sensors coupled to the transaction authentication information unit; and an anomaly detector coupled to the transaction authentication information unit. The sensors comprise one or more touchscreen sensors coupled to the touchscreen, an accelerometer, and a gyroscope, and are used to obtain and transmit one or more sets of data to the transaction authentication information unit. The sets of data are associated with one or more performances of the authentication task by the user. The anomaly detector generates an authentication model using the one or more data sets transmitted to the transaction authentication information unit.

Abstract: A time scaler for providing a time scaled version of an input audio signal is configured to compute or estimate a quality of a time scaled version of the input audio signal obtainable by a time scaling of the input audio signal. The time scaler is configured to perform the time scaling of the input audio signal in dependence on the computation or estimation of the quality of the time scaled version of the input audio signal obtainable by the time scaling. An audio decoder has such a time scaler.

Abstract: Techniques are described for determining a likelihood that a customer communication is fraudulent using one or more machine learning models. For example, a computing system includes a memory and one or more processors in communication with the memory. The one or more processors are configured to: receive a set of emotion factor values for communication data of a current communication associated with a customer, wherein each emotion factor value indicates a measure of a particular emotion factor in the current communication; classify, using an emotion variance model running on the one or more processors, the current communication into an emotional fraud category based on the set of emotion factor values for the current communication associated with the customer; and determine a risk score for the current communication indicative of a probability that the current communication is fraudulent based on at least the emotional fraud category for the current communication.

Abstract: An example computer-implemented method for contextual prediction of content for a structured resource is provided. The example method includes providing, to an endpoint device, an input interface for inputting natural language content. The example method includes receiving, from the endpoint device via the input interface, an initial input of natural language content. The example method includes receiving context data associated with the endpoint device. The example method includes obtaining, using a machine-learned content generation model, and based on the initial input and the context data, a suggested portion of natural language content. The example method includes providing, to the endpoint device, the suggested portion.

Abstract: Methods, systems, and computer-storage media are provided for utilizing multiple Picture Archiving Communication Systems (PACS) to view one more medical images by storing one or more PACS at a database within the system. Requests are received from one or more users at one or more facilities to utilize one or more PACS to view one or more medical images. After accessing the database to determine one or more PACS authorized for each facility from which a request is received, one or more users are provided with one or more PACS to view medical images associated with radiological exams and provide the necessary assessments and reports for treatment.

Abstract: Techniques for training a natural language understanding (NLU) engine may include generating a first annotation of free-form text documenting a healthcare patient encounter and a link between the first annotation and a corresponding portion of the text, using the NLU engine. A second annotation of the text and a link between the second annotation and a corresponding portion of the text may be received from a human user. The first annotation and its corresponding link may be merged with the second annotation and its corresponding link. Training data may be provided to the engine in the form of the text and the merged annotations and links.

Abstract: A route viewing system includes a computing system that receives information associated with one or more routes through a network, and identifies the routes that are associated with at least one illicit user computer used by an illicit user. The computing system then obtains a source location of a source address of the routes and a destination location of a destination address of the routes, and displays the routes on a geographical display at the source location of the source address and the destination location of the destination address of each of the routes.

Abstract: A method for generating a voice announcement as feedback to a handwritten user input is disclosed in which a user enters on a control device. A list of possible whole words which can be entered by the user input is provided together with a corresponding transcription and a predetermined word end, which comprises one or more characters of a whole word of the whole words, is removed from the end of said whole word in accordance with a predetermined shortening rule and corresponding to this, a transcription end corresponding to the word end is determined based on a predetermined assignment rule and is removed from the corresponding transcription of the whole word for generating a partial word and an associated partial transcription. The partial word and the partial transcription are added to another list.

Abstract: A method implemented by a server communicably coupled to at least two devices, each device including camera(s), the devices being present within same real-world environment. The method includes: receiving, from the devices(s), images captured by respective cameras of the devices; identifying one of the devices whose camera has camera parameter(s) better than camera parameter(s) of camera of another of the devices; training neural network using images captured by camera of one of the devices as ground truth material and using images captured by camera of another of the devices as training material; generating correction information to correct images captured by camera of another of the devices using trained neural network; and correcting the images captured by the camera of the another of the device(s) by utilising the correction information at the server, or sending correction information to another of the devices for correcting the images.

Abstract: A speech-processing system may provide access to one or more virtual assistants via a voice-controlled device. A user may leverage a first virtual assistant to translate a natural language command from a first language into a second language, which the device can forward to a second virtual assistant for processing. The device may receive a command from a user and send input data representing the command to a first speech-processing system representing the first virtual assistant. The device may receive a response in the form of a first natural language output from the first speech-processing system along with an indication that the first natural language output should be directed to a second speech-processing system representing the second virtual assistant. For example, the command may be in the first language, and the first natural language output may be in the second language, which is understandable by the second speech-processing system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage medium, for training a neural network, wherein the neural network is configured to receive an input data item and to process the input data item to generate a respective score for each label in a predetermined set of multiple labels. The method includes actions of obtaining a set of training data that includes a plurality of training items, wherein each training item is associated with a respective label from the predetermined set of multiple labels; and modifying the training data to generate regularizing training data, comprising: for each training item, determining whether to modify the label associated with the training item, and changing the label associated with the training item to a different label from the predetermined set of labels, and training the neural network on the regularizing data.

Abstract: A multi-channel transformer acoustic model that processes a plurality of audio signals output by microphones of a microphone array and outputs probabilities for acoustic units of an utterance represented in the audio signals. The audio signals represent the individual microphones' respective capturing of the utterance. The multi-channel model may perform self-attention on embeddings of the audio signals and then cross-channel attention across the attended audio signals. The cross-channel attention may involve processing of signals relative to each other to model the relationships across channels within and across time frames. The multi-channel model may include a transducer to perform processing frame-by-frame.

Abstract: Disclosed herein are techniques for using a generative adversarial network (GAN) to train a semantic parser of a dialog system. A method described herein involves accessing seed data that includes seed tuples. Each seed tuple includes a respective seed utterance and a respective seed logical form corresponding to the respective seed utterance. The method further includes training a semantic parser and a discriminator in a GAN. The semantic parser learns to map utterances to logical forms based on output from the discriminator, and the discriminator learns to recognize authentic logical forms based on output from the semantic parser. The semantic parser may then be integrated into a dialog system.

Abstract: A multi-factor identification system is provided in which enrolled user authentication information is updated in the course of an authorization request based upon at least one of a confidence level of a match between a request first factor identifier, produced based upon first unique user identifying information received with the authentication request, and a respective matching enrolled first factor identifier and a confidence level of a match between a request second factor identifier, produced based upon second unique user identifying information received with the authentication request, and a respective matching enrolled second factor identifier.

Abstract: The present disclosure provides systems and methods to obtain feedback descriptive of autonomous vehicle failures. In particular, the systems and methods of the present disclosure can detect that a vehicle failure event occurred at an autonomous vehicle and, in response, provide an interactive user interface that enables a human located within the autonomous vehicle to enter feedback that describes the vehicle failure event. Thus, the systems and methods of the present disclosure can actively prompt and/or enable entry of feedback in response to a particular instance of a vehicle failure event, thereby enabling improved and streamlined collection of information about autonomous vehicle failures.

Abstract: Some implementations relate to methods, systems, and computer-readable media to generate text tags for games. In some implementations, a computer-implemented method to generate one or more text tags includes obtaining a plurality of chat transcripts, each chat transcript associated with a respective gameplay session of a respective game of a plurality of games. Each chat transcript includes content provided by participants in the gameplay session. The method further includes programmatically analyzing the plurality of chat transcripts to determine one or more characteristics for each game of the plurality of games, and generating a text tag for at least one game of the plurality of games based on the one or more characteristics of the at least one game.

Abstract: Various embodiments of the present disclosure performing conversation sentiment monitoring for a conversation data object. In various embodiments, a text block that can be resized is identified within a conversation data object and successive regularized sentiment profile generation iterations are performed until a regularized sentiment score of the block exceeds a regularized sentiment score threshold. A current regularized sentiment profile generation iteration involves determining a regularized sentiment score for the block based on an initial sentiment score, a subjectivity probability value, and, optionally, a stage-wise penalty factor. A determination is then made as to whether the score exceeds the threshold. If so, then a regularized sentiment profile of the conversation data object is updated based on the regularized sentiment score. If not, then the text block is resized and a subsequent regularized sentiment profile generation iteration is performed based on the resized block.

Abstract: A natural language processing system may select a synthesized speech quality using user profile data. The system may receive a natural language input and determine responsive output data. The system may, based at least in part on user profile data associated with the input, determine response configuration data corresponding to a quality of synthesized speech. The system may then determine further output data for presentation using the responsive output data and response configuration data.

Abstract: Systems/methods for intelligent commissioning of an HVAC system provide a control node and at least a first network node coupled to communicate with the control node, the first network node configured to retrieve via a user interface objects configured at the control node, configure at least a second network node using the retrieved objects, and report the configuration of the second network node at the control node. A user interface of a first network node can access the objects at the control node. The first network node can apply the accessed objects to configure a second network node using a commissioning tool. The commissioning tool can be activated specifically for certain authorized HVAC personas or roles. The first network node can report the configuring at the control node. The commissioning tool can be voice-enabled to allow a single user to configure the HVAC system via voice commands.

Abstract: In an approach to digital twin enabled equipment diagnostics based on acoustic modeling, a real-time audio input of an asset is received from a mobile device. The real-time audio input is analyzed using one or more acoustic modeling algorithms to establish a deviation from a baseline, where the baseline is associated with a digital twin of the asset. Responsive to determining the deviation from the baseline exceeds a predetermined threshold, the user is iteratively directed to move the mobile device until a stopping criteria is met.

Abstract: The present disclosure describes a system and method for providing dynamic remediation of a pluggable streaming device issue, such as a customer premises equipment (CPE) device. Sometimes, various features of the CPE device to begin to fail. For example, synchronization of the audio and video streams may drift, media rental purchases may time out, or playback may throttle to low quality. Such failures can be caused by device or network issues. The present disclosure describes a CPE remediation system that operates to identify a failure associated with playing media streamed by the CPE device. The CPE remediation system may further determine a solution to remediate an observed CPE device-related failure. In some examples, the CPE remediation process may further provide or perform one or more actions included in the determined solution. In some examples, the solution may include a warm or a cold reboot.

Abstract: A method, computer program product, and computing system for synchronizing machine vision and audio is executed on a computing device and includes obtaining encounter information of a patient encounter, wherein the encounter information includes machine vision encounter information and audio encounter information. The machine vision encounter information and the audio encounter information are temporally-aligned to produce a temporarily-aligned encounter recording.

Abstract: The present disclosure relates to a virtual counseling system in which a user can virtually receive counseling by inputting query information into a system. A virtual counseling system according to an embodiment of the present disclosure may include an input unit obtaining audio information from a user and generating audio data; a determination unit receiving the audio data through the input unit, determining a type of the audio data, and generating type information on the audio data; and a text data generation unit generating object data by receiving the type information from the determination unit, converting content of the audio data into first text data, and combining the object data and the first text data to generate second text data.

Abstract: A voice recognition method includes the following steps. An audio and a correct result are received. The audio is recognized, and a text file corresponding to the audio is output. The word error rate is determined by comparing the text file to the correct result. The word error rate is adjusted according to the weight of at least one important word, in order to calculate a professional score that corresponds to the text file. A determination is made as to whether the professional score is higher than a score threshold. In response to the professional score is higher than the score threshold, the text file, the audio, or the correct result corresponding to the professional score is sent to an engine training module for training.

Abstract: A method performed by an audio system comprising a headset. The method sends a playback signal containing user-desired audio content to drive a speaker of the headset that is being worn by a user, receives a microphone signal from a microphone that is arranged to capture sounds within an ambient environment in which the user is located, performs a speech detection algorithm upon the microphone signal to detect speech contained therein, in response to a detection of speech, determines that the user intends to engage in a conversation with a person who is located within the ambient environment, and, in response to determining that the user intends to engage in the conversation, adjusts the playback signal based on the user-desired audio content.

Abstract: Systems and methods for detecting demographic bias in automatic speech recognition (ASR) systems. Corpuses of transcriptions from different demographic groups are analyzed, where one of the groups is known to be susceptible to bias and another group is known not to be susceptible to bias. A difference between the transcription accuracy for the first group and a transcription accuracy for a second group is measured. ASR accuracy for each group is measured and compared to each other using both statistics-based and practicality-based methodologies to determine whether a given ASR system or model exhibits a meaningful level of bias. Based on the statistical significance and the practical significance, an alert including a recommendation to adjust the ASR model is generated.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identity management are disclosed. In one aspect, a method includes the actions of receiving, from a first computing device, first audio data that includes representations of one or more words in a first voice. The actions further include generating second audio data that includes representations of the one or more words in a second voice. The actions further include providing, for output to a second computing device, the second audio data.

Abstract: A system is provided to categorize voice prints during a voice authentication. The system includes a processor and a computer readable medium operably coupled thereto, to perform voice authentication operations which include receiving an enrollment of a user in the biometric authentication system, requesting a first voice print comprising a sample of a voice of the user, receiving the first voice print of the user during the enrollment, accessing a plurality of categorizations of the voice prints for the voice authentication, wherein each of the plurality of categorizations comprises a portion of the voice prints based on a plurality of similarity scores of distinct voice prints in the portion to a plurality of other voice prints, determining, using a hidden layer of a neural network, one of the plurality of categorizations for the first voice print, and encoding the first voice print with the one of the plurality of categorizations.

Abstract: Aspects of the disclosure generally relate to computing devices and may be generally directed to devices, systems, methods, and/or applications for learning conversations among two or more conversation participants, storing this knowledge in a knowledgebase (i.e. neural network, graph, sequences, etc.), and enabling a user to simulate a conversation with an artificially intelligent conversation participant.

Abstract: Contact centers strive to provide a positive and productive customer-agent interaction to successfully resolve the issue for a call. While audio content, such as music or messages, on hold are commonplace, selecting audio enhancements to be inserted into, and concurrently with, the customer-agent provides the customer and/or agent with cues and motivations to promote the successful completion of the call. Cues may be provided to announce the arrival or departure of an agent, virtually take a customer from one location to another for a different portion of the interaction, add excitement and anticipation to an upcoming event by providing an audio experience foreshadowing of the actual event, calm frayed nerves, or other purpose.

Abstract: Various techniques are disclosed, including receiving at a multiplatform management system a natural language request from a computing device, the multiplatform management system interfacing with multiple disparate platforms including a natural language processing platform, determining an event type based on the natural language request, identifying a user-requested action based on data associated with the natural language processing platform in data communication with the multiplatform management system, selecting a cloud platform to perform the user-requested action, formatting data representing the user-requested action into a formatted user-requested action, and performing the action.

Abstract: Provided is a display apparatus for displaying an operation menu associated with a data processing performed on handwritten data, wherein the operation menu includes information related to the data processing according to a display position of the operation menu.

Abstract: A sound association system identifies one or more aurally active words in digital text. Aurally active words refer to words that denote particular sounds. Context-based sounds corresponding to the one or more aurally active words are also identified. Each context-based sound is anchored to or associated with the corresponding one or more aurally active words and is played back when the digital text is played back or read, providing context-based background sounds associated with the one or more aurally active words. For example, a context-based sound can be played back at a higher volume when the one or more aurally active words are played back or read, and at a lower volume when other words of the digital text are played back or read.

Abstract: A wrist terminal includes a communicator, timer, and at least one processor. The communicator receives a beacon ID transmitted from a beacon transmitter installed in a workplace. The timer obtains date-and-time information on a date and time at which the communicator receives the beacon ID. The processor performs a determining process and recording process. In the determining process, the processor determines whether a work status in the workplace is a work start or a work end, based on a state of the wrist terminal when the communicator receives the beacon ID. In the recording process, the processor records, in a storage, log information that includes the date-and-time information obtained by the timer and work status information on the work status determined in the determining process, the date-and-time information and the work status information being associated with each other.

Abstract: A method of detecting live speech comprises: receiving a signal containing speech; obtaining a first component of the received signal in a first frequency band, wherein the first frequency band includes audio frequencies; and obtaining a second component of the received signal in a second frequency band higher than the first frequency band. Then, modulation of the first component of the received signal is detected; modulation of the second component of the received signal is detected; and the modulation of the first component of the received signal and the modulation of the second component of the received signal are compared. It may then be determined that the speech may not be live speech, if the modulation of the first component of the received signal differs from the modulation of the second component of the received signal.