Abstract: A decoder for generating an audio output signal having one or more audio output channels is provided, having a receiving interface for receiving an audio input signal having a plurality of audio object signals, for receiving loudness information on the audio object signals, and for receiving rendering information indicating whether one or more of the audio object signals shall be amplified or attenuated, further having a signal processor for generating the one or more audio output channels of the audio output signal, configured to determine a loudness compensation value depending on the loudness information and depending on the rendering information, and configured to generate the one or more audio output channels of the audio output signal from the audio input signal depending on the rendering information and depending on the loudness compensation value. One or more by-pass audio object signals are employed for generating the audio output signal. Moreover, an encoder is provided.

Abstract: The present disclosure generally relates to interview training and providing interview feedback. An exemplary method comprises: at an electronic device that is in communication with a display and one or more input devices: receiving, via the one or more input devices, media data corresponding to a user's responses to a plurality of prompts; analyzing the media data; and while displaying, on the display, a media representation of the media data, displaying a plurality of analysis representations overlaid on the media representation, wherein each of the plurality of analysis representations is associated with an analysis of content located at a given time in the media representation and is displayed in coordination with the given time in the media representation.

Abstract: An off-policy reinforcement learning actor-critic neural network system configured to select actions from a continuous action space to be performed by an agent interacting with an environment to perform a task. An observation defines environment state data and reward data. The system has an actor neural network which learns a policy function mapping the state data to action data. A critic neural network learns an action-value (Q) function. A replay buffer stores tuples of the state data, the action data, the reward data and new state data. The replay buffer also includes demonstration transition data comprising a set of the tuples from a demonstration of the task within the environment. The neural network system is configured to train the actor neural network and the critic neural network off-policy using stored tuples from the replay buffer comprising tuples both from operation of the system and from the demonstration transition data.

Abstract: A method for increasing stability of an inter-channel time difference (ICTD) parameter in parametric audio coding, wherein a multi-channel audio input signal comprising at least two channels is received. The method comprises obtaining an ICTD estimate, ICTDest(m), for an audio frame m and a stability estimate of said ICTD estimate, and determining whether the obtained ICTD estimate, ICTDest(m), is valid. If the ICTDest(m) is not found valid, and a determined sufficient number of valid ICTD estimates have been found in preceding frames, a hang-over time is determined using the stability estimate and a previously obtained valid ICTD parameter, ICTD(m?1), is selected as an output parameter, ICTD(m), during the hang-over time. The output parameter, ICTD(m), is set to zero if valid ICTDest(m) is not found during the hang-over time.

Abstract: Computer-implemented method and system for receiving and processing one or more moment-associating elements. For example, the computer-implemented method includes receiving the one or more moment-associating elements, transforming the one or more moment-associating elements into one or more pieces of moment-associating information, and transmitting at least one piece of the one or more pieces of moment-associating information.

Abstract: A system and method establishes a communication connection between a first device of a first user and a second device of a second user. Request data corresponding to a request to establish a communication connection with a second user is received, and a user profile associated with the second user is determined. One or more sensors of the second device receive input data corresponding to the environment of the second device, and an identity of the second user is determined based thereon. The communication connection is established and, based on the identity, the second device tracks movement of the second user in the environment.

Abstract: Virtual assistants intelligently emulate a representative of a service provider by providing variable responses to user queries received via the virtual assistants. These variable responses may take the context of a user's query into account both when identifying an intent of a user's query and when identifying an appropriate response to the user's query.

Abstract: Devices, systems and processes for a dynamic microphone system that enhances the passenger experience in autonomous vehicles are described. One example method for enhancing a passenger experiences includes generating, using an artificial intelligence algorithm, a plurality of filters based on a plurality of stored waveforms previously recorded by each of one or more passengers and a plurality of recordings of one or more noise sources, capturing voice commands from at least one of the one or more passengers inside the autonomous vehicle, generating voice commands with reduced distortion based on processing the voice commands using the plurality of filters, and instructing, based on the voice commands with reduced distortion, the autonomous vehicle to perform one or more actions.

Abstract: Systems and methods for correcting a voice query based on a subsequent voice query with a lower pronunciation rate. In some aspects, the systems and methods calculate first and second pronunciation rates of first and second voice queries. The systems and methods determine that the second pronunciation rate is lower than the first pronunciation rate and determine a first candidate pronunciation time for a first candidate word from the first voice query. The systems and methods determine a second candidate pronunciation time, adjusted to the first pronunciation rate, for the second candidate word from the second voice query. The systems and methods determine that the first candidate pronunciation time matches the second candidate pronunciation time and generate a third voice query based on the first voice query by replacing the first candidate word with the second candidate word.

Abstract: Presented are a combined learning method and device using a transformed loss function and feature enhancement based on a deep neural network for speaker recognition that is robust to a noisy environment.

Abstract: A message playing method includes: receiving a first message, and asking in a voice manner, whether to play the first message; if a first voice of a user does not match a keyword of a positive reply, continuing to detect a voice of the user; if a second voice of the user detected, matches the keyword of the positive reply, playing the first message in the voice manner, and recording a quantity of times of using a text corresponding to the first voice; and when the quantity of times of using the text that corresponds to the first voice and that is recorded is greater than a first threshold, adding the text to the keyword of the positive reply.

Abstract: A system and method includes identifying unstructured conversational dialogue data sourced from communications between a subscriber and a conversational dialogue agent; automatically mapping, via the one or more computers, one or more distinct unstructured data synthetization requests defined in the unstructured conversational dialogue data to a distinct artifact synthetization objective defined within a synthetization objective distillation layer; generating, via the one or more computers, a plurality of artifact synthetization prompts corresponding to the plurality of unstructured synthetization data requests based on the distinct artifact synthetization objective mapped to each of the one or more distinct unstructured synthetization data requests; and generating, by a target machine learning model, a plurality of synthesized digital artifacts based on an input of the plurality of artifact synthetization prompts generated for the plurality of unstructured synthetization data requests.

Abstract: A method for predicting parametric vocoder parameter includes receiving a text utterance having one or more words, each word having one or more syllables, and each syllable having one or more phonemes. The method also includes receiving, as input to a vocoder model, prosodic features that represent an intended prosody for the text utterance and a linguistic specification. The prosodic features include a duration, pitch contour, and energy contour for the text utterance, while the linguistic specification includes sentence-level linguistic features, word-level linguistic features for each word, syllable-level linguistic features for each syllable, and phoneme-level linguistic features for each phoneme. The method also includes predicting vocoder parameters based on the prosodic features and the linguistic specification.

Abstract: The disclosure provides a method and an apparatus for generating a response, an electronic device, and a storage medium. The method includes: obtaining a current user request in a current conversation and historical coreference information in the current conversation; extracting content matching the current user request from the historical coreference information; updating the current user request based on the content to obtain an updated current user request; and generating a response of the current user request based on the updated current user request.

Abstract: An on-vehicle device includes: a plurality of agent function units configured to provide services including causing an output unit to output an audio response in response to an utterance of an occupant of a vehicle; and a content management unit configured to determine whether or not the instructed content is stored in an in-vehicle storage device mounted in the vehicle or a portable storage medium brought into the vehicle when the playback of the content is instructed by the utterance of the occupant, and to cause the playback device to play back the content present in the in-vehicle storage device or the portable storage medium when the instructed content is determined as being stored in the in-vehicle storage device or the portable storage medium.

Abstract: Systems and methods for maintaining voice assistant persistence across multiple network microphone devices are described. In one example, first and second NMDs each identify a wake word based on detected sound, and are each transitioned from an inactive state to an active state in which the NMD captures and transmits sound data over a network interface. The first NMD is selected over the second NMD to output a first response, and both NMDs remain in the active state to further capture and transmit sound data. After further capturing and transmitting of sound data, the second NMD is selected over the first NMD to output a second response. After a predetermined time, one or both of the NMDs are transitioned back to the inactive state. The selection of one NMD over another for outputting a response can be based at least in part on user location information.

Abstract: A method for encoding an input audio stream including the steps of obtaining a first playback stream presentation of the input audio stream intended for reproduction on a first audio reproduction system, obtaining a second playback stream presentation of the input audio stream intended for reproduction on a second audio reproduction system, determining a set of transform parameters suitable for transforming an intermediate playback stream presentation to an approximation of the second playback stream presentation, wherein the transform parameters are determined by minimization of a measure of a difference between the approximation of the second playback stream presentation and the second playback stream presentation, and encoding the first playback stream presentation and the set of transform parameters for transmission to a decoder.

Abstract: Systems and methods for media playback via a media playback system include capturing sound data via a network microphone device and identifying a candidate wake word in the sound data. Based on identification of the candidate wake word in the sound data, the system selects a first wake-word engine from a plurality of wake-word engines. Via the first wake-word engine, the system analyzes the sound data to detect a confirmed wake word, and, in response to detecting the confirmed wake word, transmits a voice utterance of the sound data to one or more remote computing devices associated with a voice assistant service.

Abstract: A method of using voice commands from a mobile device to remotely access and control a computer. The method includes receiving audio data from the mobile device at the computer. The audio data is decoded into a command. A software program that the command was provided for is determined. At least one process is executed at the computer in response to the command. Output data is generated at the computer in response to executing at least one process at the computer. The output data is transmitted to the mobile device.

Abstract: At least selectively utilizing a large language model (LLM) in generating a natural language (NL) based summary to be rendered in response to a query. In some implementations, in generating the NL based summary additional content is processed using the LLM. The additional content is in addition to query content of the query itself and, in generating the NL based summary, can be processed using the LLM and along with the query content—or even independent of the query content. Processing the additional content can, for example, mitigate occurrences of the NL based summary including inaccuracies and/or can mitigate occurrences of the NL based summary being over-specified and/or under-specified.