Abstract: The present document relates to methods and apparatus for rendering input audio for playback in a playback environment. The input audio includes at least one audio object and associated metadata, and the associated metadata indicates at least a location of the audio object. A method for rendering input audio including divergence metadata for playback in a playback environment comprises creating two additional audio objects associated with the audio object such that respective locations of the two additional audio objects are evenly spaced from the location of the audio object, on opposite sides of the location of the audio object when seen from an intended listener's position in the playback environment, determining respective weight factors for application to the audio object and the two additional audio objects, and rendering the audio object and the two additional audio objects to one or more speaker feeds in accordance with the determined weight factors.

Abstract: Automatic control of media presentation parameters is provided by using one or more of real-time audio playback measurement data from microphones and audience facial and body expression interpretation from video and infrared cameras, in conjunction with artificial intelligence for interpretation and evaluation of facial and body expression and predetermined perceptual audio models. Media presentation parameters can include, for example, speaker volume, audio equalization, feedback elimination, play/pause, and other audio content-related aspects of presentation. In some embodiments, additional environmental parameters can be modified to enhance audience experience, such as, for example, temperature, lighting, and the like, in response to audience facial and body expression.

Abstract: An information providing device includes circuitry configured to: acquire an uttered word which is uttered by a user and an utterance time at which the Littered word is uttered by the user; control output of offer information associated with the uttered word to the user; and restrict output of the offer information associated with the uttered word within a predetermined masking period from the utterance time of the uttered word.

Abstract: A data processing unit includes a processing circuit that is configured to process data based on a value of a first parameter, a first operator that is selectively set to one of a first state and a second state that are physically identified, a second operator that is set to a physical state indicating the value of the first parameter, and a processor that is configured to set the value of the first parameter indicated by the physical state of the second operator in the processing circuit in a case where the first operator is in the first state at a time of activating the data processing unit, and set a value of the first parameter supplied from the information processing device in the processing circuit in a case where the first operator is in the second state at the time of activating the data processing unit.

Abstract: An agent automation system includes a memory configured to store a reasoning agent/behavior engine (RA/BE) including a first persona and a current context and a processor configured to execute instructions of the RA/BE to cause the first persona to perform actions comprising: receiving intents/entities of a first user utterance; recognizing a context overlay cue in the intents/entities of the first user utterance, wherein the context overlay cue defines a time period; updating the current context of the RA/BE by overlaying context information from at least one stored episode associated with the time period; and performing at least one action based on the intents/entities of the first user utterance and the current context of the RA/BE.

Abstract: [Object] To provide an information processing apparatus by which sound can be smoothly re-listened to. [Solution] There is provided an information processing apparatus including: a reproduction processing unit that performs reproduction of a recorded sound on a basis of a reproduction start instruction for starting re-listening of the recorded sound from a position tracking back a predetermined time from a reproduction start time, at which the reproduction start instruction is input, to a position of a present time.

Abstract: Embodiments of the present systems and methods may provide techniques that provide the capability to automatically generate allowance intervals for voice personas that meet desired requirements for realism and fidelity. For example, a method for voice persona generation may be implemented in a computer system comprising a processor, memory accessible by the processor, and computer program instructions stored in the memory and executable by the processor, the method comprising: displaying to a user, a plurality of user-selectable voice persona parameters that control features of a synthesized voice signal, and displaying, in conjunction with each of at least some of plurality of user-selectable voice persona parameters, voice transformation allowance intervals of the voice persona parameters, accepting from a user, a selection of at least one user-selectable voice persona parameter, and generating a synthesized voice signal based on the selected at least one user-selectable voice persona parameter.

Abstract: The disclosed technology teaches a deep end-to-end speech recognition model, including using multi-objective learning criteria to train a deep end-to-end speech recognition model on training data comprising speech samples temporally labeled with ground truth transcriptions.

Abstract: A first zone player engages in synchronous playback of given audio content by obtaining the given audio content, generating and placing representative audio frames into a buffer, and transmitting the audio frames to a second zone player to play the given audio content in synchrony with the second zone player. After receiving a command to pause the synchronous playback, the first zone player prepares for a fast-resume by identifying a given audio frame and retaining at least some of the audio frames in the buffer for use during the fast-resume. The first zone player then initiates the fast-resume by determining a future resume time, transmitting an instruction to the second zone player to resume playback at the future resume time, and at the future resume time, resuming use of the audio frames in the buffer, starting with the given audio frame, to play the given audio content in synchrony.

Abstract: A device for detecting an audio interface includes a processing unit, a first audio interface transmitting circuit, and a second audio interface transmitting circuit. The processing unit is utilized to generate a clock signal and a word select (WS) signal. The first audio interface transmitting circuit is utilized to generate a first audio data according to the clock signal. The second audio interface transmitting circuit is utilized to generate a second audio data according to the clock signal and the WS signal. The processing unit switches to the first audio interface transmitting circuit if a voltage potential of the WS signal remains at a high voltage level or remains at a low voltage level longer than a predetermined period. The processing unit switches to the second audio interface transmitting circuit if the voltage potential of the WS signal changes during the predetermined period.

Abstract: Described herein are systems and methods for compressing or otherwise reducing the memory requirements for storing and computing the model parameters in recurrent neural language models. Embodiments include space compression methodologies that share the structured parameters at the input embedding layer, the output embedding layers, or both of a recurrent neural language model to significantly reduce the size of model parameters, but still compactly represent the original input and output embedding layers. Embodiments of the methodology are easy to implement and tune. Experiments on several data sets show that embodiments achieved similar perplexity and BLEU score results while only using a fraction of the parameters.

Abstract: A method and device for sound effect processing, and a non-transitory storage medium. The method includes the following actions. A task manager is traversed to determine whether a sound effect service process for a sound effect service exists. Responsive to determining that the sound effect service process exists, whether the sound effect service process is a system process of a system, is determined. When the sound effect service process is not a system process of the system, the sound effect service process is set to be a system process of the system.

Abstract: A device and method for backward compatibility for virtual reality (VR), mixed reality (MR), augmented reality (AR), computer vision, and graphics systems. The device and method enable rendering audio data with more degrees of freedom on devices that support fewer degrees of freedom. The device includes memory configured to store audio data representative of a soundfield captured at a plurality of capture locations, metadata that enables the audio data to be rendered to support N degrees of freedom, and adaptation metadata that enables the audio data to be rendered to support M degrees of freedom. The device also includes one or more processors coupled to the memory, and configured to adapt, based on the adaptation metadata, the audio data to provide the M degrees of freedom, and generate speaker feeds based on the adapted audio data.

Abstract: A system and method executed by audio processing software on one or more electronic devices in a computer system to process digital audio signals.

Abstract: Embodiments of the present invention provide an earphone volume adjustment method and apparatus. The method includes: when it is detected that an intensity of external environmental noise is greater than a preset threshold, obtaining position information and/or a motion status of a user; determining a time window according to the position information and/or the motion status; and adjusting earphone volume according to an intensity of external environmental noise in the time window. According to the embodiments of the present invention, an inappropriate phenomenon such as turning up or turning down the earphone volume in a short time can be avoided, and the earphone volume can be appropriately adjusted by comprehensively considering the intensity of the external environmental noise and the position information and/or the motion status of the user. Therefore, user experience is improved.

Abstract: A hearing device includes: an antenna for receiving a first wireless input signal from an external device and providing an antenna output signal; a transceiver configured to provide a transceiver input signal; an input module for provision of a first input signal, the input module comprising a first microphone; a processor; a receiver configured to provide an audio output signal; a pre-processor for provision of a pre-processor output signal based on the first input signal; and a controller comprising a speech intelligibility estimator for determining a speech intelligibility indicator indicative of speech intelligibility based on the transceiver input signal and a first controller input signal, wherein the controller is configured to provide a controller output signal based on the speech intelligibility indicator; wherein the pre-processor is configured to apply, based on the controller output signal, a pre-processing scheme to the first input signal and/or the transceiver input signal.

Abstract: Example techniques may involve guest access to a media playback system. A guest may use a guest control device, such as a smartphone or tablet, to control aspects of a host's media playback system. In addition, the guest may temporarily register their user account of a streaming audio service with the host's media playback system, which enables playback of audio content from that service by one or more playback devices of the media playback system. When the guest control device de-registers from the host's media playback system, retrieval of audio content from the streaming audio service is disabled.

Abstract: Methods, systems, and computer programs are presented for generating location-based playlists. The disclosed method includes providing a music service for generating playlists for a location, identifying users having respective user devices within the defined boundaries of the location, and aggregating music preferences of the identified users. Each of the user devices have access to the music service and the aggregated music preferences of the identified users identify a plurality of music tracks. The disclosed method further includes generating a playlist having the plurality of music tracks based on the aggregated music preferences and providing an access to the generated playlist to the identified users at the location. The plurality of music tracks of the playlist is provided for listening by the music service to one or more of the user devices.

Abstract: A method of playing media files includes accessing a media library, identifying a plurality of media files in the media library, indexing metadata of the plurality of media files, targeting a first subset of media files in the plurality of media files based on metadata of a selected media file, and playing at least one media file of the first subset of media files.

Abstract: A CS CTC model may be initialed from a major language CTC model by keeping network hidden weights and replacing output tokens with a union of major and secondary language output tokens. The initialized model may be trained by updating parameters with training data from both languages, and a LID model may also be trained with the data. During a decoding process for each of a series of audio frames, if silence dominates a current frame then a silence output token may be emitted. If silence does not dominate the frame, then a major language output token posterior vector from the CS CTC model may be multiplied with the LID major language probability to create a probability vector from the major language. A similar step is performed for the secondary language, and the system may emit an output token associated with the highest probability across all tokens from both languages.