Abstract: Adapters for neural machine translation systems. A method includes determining a set of similar n-grams that are similar to a source n-gram, and each similar n-gram and the source n-gram is in a first language; determining, for each n-gram in the set of similar n-grams, a target n-gram is a translation of the similar n-gram in the first language to the target n-gram in the second language; generating a source encoding of the source n-gram, and, for each target n-gram determined from the set of similar n-grams determined for the source n-gram, a target encoding of the target n-gram and a conditional source target memory that is an encoding of each of the target encodings; providing, as input to a first prediction model, the source encoding and the condition source target memory; and generating a predicted translation of the source n-gram from the first language to the second language.

Abstract: Methods, apparatuses, and systems are described for dynamically navigating interactive communication systems. An example method may comprise: receiving, from a user device, sound waves or audio information, the sound waves or audio information indicative of a request to initiate an interactive communication session with a communication system of a biller or merchant; interpreting, based on the sound waves or audio information, an intent of the communication session and an identity of the biller or merchant; retrieving a predetermined interaction coding associated with the biller or merchant; and initiating the interactive communication session with the communication system of the biller or merchant based on the predetermined interaction coding.

Abstract: A method, computer program product, and computing system for receiving a plurality of predefined beams associated with a microphone array. A plurality of predefined nulls associated with the microphone array may be received. One or more predefined beams from the plurality of predefined beams or one or more predefined nulls from the plurality of predefined nulls may be selected. A microphone array may obtain audio encounter information, via the microphone array, using at least one of the one or more selected beams and the one or more selected nulls.

Abstract: Automated methods and systems are provided for preserving the perception of interparticipant directional gesture and gaze across a grid or mesh of remote video feeds from a plurality of participants engaged in a videoconferencing session. The perception of directional gesture and gaze is accomplished through mirroring the video feeds in a gallery view displayed to those participants who are interacting. The perception of interparticipant directional gesture and gaze of those being mirrored by those not mirrored is accomplished by un-mirroring (or double-mirroring) those interacting.

Abstract: Embodiments of the present disclosure provide a wearable electronic device for providing audio output and capturing a visual media. The wearable electronic device includes a neckband including a pair of arms coupled by a central portion therebetween, and at least one image sensor disposed in the neckband. Further, the wearable electronic device includes a processor operatively coupled to a communication interface, and is configured to at least receive a control command through an application, for capturing the image data of a surrounding environment of the user. The processor is configured to trigger the at least one image sensor for capturing the image data of the surrounding environment of the user in real-time. Further, the processor transmits the image data being captured by the at least one image sensor in real-time, to the user device of the user for enabling the user to view the visual media of the surrounding environment.

Abstract: Embodiments include a microphone assembly comprising an array microphone and a housing configured to support the array microphone and sized and shaped to be mountable in a drop ceiling in place of at least one of a plurality of ceiling tiles included in the drop ceiling. A front face of the housing includes a sound-permeable screen having a size and shape that is substantially similar to the at least one of the plurality of ceiling tiles. Embodiments also include an array microphone system comprising a plurality of microphones arranged, on a substrate, in a number of concentric, nested rings of varying sizes around a central point of the substrate. Each ring comprises a subset of the plurality of microphones positioned at predetermined intervals along a circumference of the ring.

Abstract: An intelligent noise cancellation process for audio or video conference calls. Different levels of deep learning model classifiers are leveraged to determine, in real-time, the presence of noise data and voice data in audio input signals being received at numerous audio input devices. In response, appropriate action is taken to prevent the noise data from being included in the subsequent audio communication. Specifically, a lightweight neural network-based model classifier is initially used to identify noise data and/or the presence of predetermined trigger words or phrases in audio input signals. In the event that the lightweight model is unable to identify the presence of the triggering words/phrases, a heavyweight neural network-based model classifier is called upon, whereby the audio signals are attempted to be converted to a human-understandable language format (i.e., a text format) as a means of positively identifying voice data in audio input signals.

Abstract: Disclosed are a speech recognition device and a speech recognition method, which perform speech recognition by executing an artificial intelligence (AI) algorithm and/or a machine learning algorithm provided therein, and which can communicate with other electronic devices and an external server in a 5G communication environment.

Abstract: In one aspect, a networked microphone device is configured to (i) receive sound data, (ii) determine, via the wake-word engine, that a first portion of the sound data is representative of a wake word, (iii) determine that a second networked microphone device was added to a media playback system, (iv) transmit the first portion of the sound data to a second networked microphone device, (v) begin determining a command to be performed by the first networked microphone device, (vi) receive an indication of whether the first portion of the sound data is representative of the wake word, and (vii) output a response indicative of whether the first portion of the sound data is representative of the wake word.

Abstract: A language translation assembly includes a housing that is wearable on a user's ear. A control circuit is positioned within the housing and the control circuit stores a language translation program. A retainer is coupled to the housing and the retainer is positionable around the user's ear for retaining the housing on the user's ear. A microphone is coupled to the housing to sense audile sounds. A speaker is coupled to the housing to emit words translated into the native language of the user when the microphone senses spoken words in a non-native language with respect to the user. The operational software selects an appropriate response in the user's native language from the language database. Additionally, the speaker emits the appropriate response to instruct the user to speak in the non-native language.

Abstract: Interleaved conversation concept flow enhancement can include detecting a topic of an on-going conversation over an electronic communication system. In response to detecting a topic newly introduced into the on-going conversation, a topic-divergence metric (TDM) with respect to the original topic and the newly introduced topic can be determined. A topic-divergence response action can be initiated in response to the newly introduced topic diverging from the original topic by more than a predetermined threshold based on the TDM.

Abstract: An audio device with a body configured to be worn on or abutting an outer ear of a user, wherein the body is configured to contact at least one of the outer ear and the portion of the head that abuts the outer ear, at two separate spaced contact locations, and wherein the body is compliant at a body portion that defines one of the contact locations. The device also has an acoustic module carried by the body and configured to locate a sound-emitting opening anteriorly of and proximate the user's ear canal opening when the body is worn on or abutting the ear of the user.

Abstract: Systems and methods for improved acoustic echo cancellation are provided. In various embodiments, a microphone located in the loudspeaker enclosure provides a first signal that is used to estimate the loudspeaker displacement which is proportional to the sound pressure level (SPL) inside the enclosure. A second signal is then derived by mapping the displacement to the loudspeaker's force factor (Bl(x)) and then modulating this by a measured current to a voice coil inside the speaker to provide an estimate of the force acting on the moving mass of the loudspeaker. The first signal is highly correlated with the echo signal for low frequencies and the second signal is highly correlated with the echo signal for high frequencies. The two signals are then combined to provide a single improved AEC reference signal.

Abstract: In a refrigerator, when a door switch detects opening of a door of the refrigerator, a voice recognition control unit determines whether voice data acquired by a voice acquisition unit satisfies a predetermined voice recognition start condition. In a case where the voice recognition start condition is satisfied, the recognition control unit controls the voice recognition unit to perform voice recognition. However, in a case where the voice recognition start condition is not satisfied, the recognition control unit controls the voice recognition unit not to perform voice recognition.

Abstract: A system includes: a server including first circuitry and a memory that stores, for each event, voice data recorded during the event, text data converted from the voice data, and time information indicating a time when the text data was generated; and a display control apparatus communicably connected with the server, including second circuitry to based on information on the event stored in the memory, control a display to display text data in an order according to the time when the text data was generated, and a graphical control region that sets playback position in a total playback time of the voice data, and in response to selection of particular text data from the text data being displayed, control the display to display the graphical control region at a location determined based on a time when the particular text data was generated.

Abstract: Disclosed herein is a web-based videoconference system that allows for video avatars to navigate within the virtual environment. The system has a presented mode that allows for a presentation stream to be texture mapped to a presenter screen situated within the virtual environment. The relative left-right sound is adjusted to provide sense of an avatar's position in a virtual space. The sound is further adjusted based on the area where the avatar is located and where the virtual camera is located. Video stream quality is adjusted based on relative position in a virtual space. Three-dimensional modeling is available inside the virtual video conferencing environment.

Abstract: Methods and systems are provided for participant-individualized audio volume control and host-customized audio volume control of streaming audio for a plurality of participants who are each receiving the streaming audio from a host within a videoconferencing platform, and who are also simultaneously engaged in remote audio communications with each other within the same videoconferencing platform.

Abstract: A computing device receives a video feed. The video feed is divided into a sequence of video segments. For each video segment, visual features of the video segment are extracted. A predicted spectrogram is generated based on the extracted visual features. A synthetic audio waveform is generated from the predicted spectrogram. All synthetic audio waveforms of the video feed are concatenated to generate a synthetic soundtrack that is synchronized with the video feed.

Abstract: A speech recognition device of the present disclosure recognizes an uttered speech of a user, and includes a controller configured to control output of any disturbance sound according to whether uttered content requested to the user is content desired not to be heard by a third party, and stop the output of the disturbance sound in response to end of an utterance of the user.

Abstract: The present disclosure relates generally to systems and methods for facilitating two-way communication sessions using serverless cloud-based functions configured in a function-as-a-service (FaaS) system. One example includes accessing a template configured to execute a response based on an event, facilitating a two-way communication session with a user device, and processing data of the two-way communication session to identify an event trigger corresponding to the template. Execution of a serverless cloud-based function associated with the event trigger is requested, and one or more outputs of the serverless cloud-based function associated with the event trigger are integrated into the two-way communication session.