Abstract: An example playback device is configured to receive a first stream of audio comprising source audio content to be played back by the playback device and record, via one or more microphones of the playback device, an audio signal output by the playback device based on the playback device playing the source audio content. The playback device is also configured to determine a transfer function between a frequency-domain representation of the first stream of audio and a frequency-domain representation of the recorded audio signal, and then determine an estimated frequency response of the playback device based on a difference between (i) the transfer function and (ii) a self-response of the playback device, where the self-response of the playback device is stored in a memory of the playback device. Based on the estimated frequency response, the playback device is configured to determine an acoustic calibration adjustment and implement the acoustic calibration adjustment.

Abstract: A method includes receiving a sequence of acoustic frames characterizing a source speech utterance including semantic information and source speech characteristics, obtaining a latent speaker embedding representing target speech characteristics, and generating, at each of a plurality of output steps, using a content encoder of a voice conversion model, a soft speech representation for a corresponding acoustic frame. The method also includes determining, at each of the plurality of output steps, an acoustic estimation for the corresponding acoustic frame, and generating, at each of the plurality of output steps, using a decoder of the voice conversion model, a synthetic speech representation for the corresponding acoustic frame based on the soft speech representation and the acoustic estimation. The synthetic speech representation includes the semantic information of the source speech utterance and the target speech characteristics of the latent speaker embedding.

Abstract: A method includes receiving target audio data captured by a first audio input device, the target audio data comprising a target audio signal and a first version of an interfering audio signal, and receiving reference audio data captured by a second audio input device different from the first audio input device, the reference audio data comprising a second version of the interfering audio signal. The method also includes processing, using a trained neural network, the target audio data and the reference audio data to generate enhanced audio data, the neural network attenuating the interfering audio signal in the enhanced audio data.

Abstract: In at least one embodiment, an apparatus for providing a virtual sound effect in a listening environment is provided. The apparatus includes at least one controller and an audio playback device. The audio playback device includes the at least one controller that is programmed to receive an audio input signal from an audio input source and to apply a head related transfer function (HRTF) to the audio input signal. The at least one controller is further programmed to apply crosstalk cancellation to the audio input signal and to generate an audio output signal after applying the HRTF and the crosstalk cancellation to the audio input signal for playback by at least one loudspeaker that is invisible to a listener in the listening environment.

Abstract: An example playback device is configured to receive a first stream of audio comprising source audio content to be played back by the playback device and record, via one or more microphones of the playback device, an audio signal output by the playback device based on the playback device playing the source audio content. The playback device is also configured to determine a transfer function between a frequency-domain representation of the first stream of audio and a frequency-domain representation of the recorded audio signal, and then determine an estimated frequency response of the playback device based on a difference between (i) the transfer function and (ii) a self-response of the playback device, where the self-response of the playback device is stored in a memory of the playback device. Based on the estimated frequency response, the playback device is configured to determine an acoustic calibration adjustment and implement the acoustic calibration adjustment.

Abstract: In one aspect, a first playback device is configured to (i) receive a set of voice signals, (ii) process the set of voice signals using a first set of audio processing algorithms, (iii) identify, from the set of voice signals, at least two voice signals that are to be further processed, (iv) determine that the first playback device does not have a threshold amount of computational power available, (v) receive an indication of an available amount of computational power of a second playback device, (vi) send the at least two voice signals to the second playback device, (vii) cause the second playback device to process the at least two voice signals using a second set of audio processing algorithms, (viii) receive, from the second playback device, the processed at least two voice signals, and (ix) combine the processed at least two voice signals into a combined voice signal.

Abstract: In at least one embodiment, an apparatus for providing a virtual sound effect in a listening environment is provided. The apparatus includes at least one controller and an audio playback device. The audio playback device includes the at least one controller that is programmed to receive an audio input signal from an audio input source and to apply a head related transfer function (HRTF) to the audio input signal. The at least one controller is further programmed to apply crosstalk cancellation to the audio input signal and to generate an audio output signal after applying the HRTF and the crosstalk cancellation to the audio input signal for playback by at least one loudspeaker that is invisible to a listener in the listening environment.

Abstract: The present disclosure provides an adaptive equalization method and system for an audio system. The method includes: detecting an input signal and estimating signal energy of the input signal; determining an audio signal status based on the estimated signal energy; performing dynamic gain calculation based on the estimated signal energy and the audio signal status; and using the calculated gain to adjust equalization parameters, thereby performing dynamic equalization processing on the input signal.

Abstract: The disclosure relates to a method and system of virtualized spatial audio. The method may use a motion sensor to track a listener's movement, obtain location information associated with the listener's movement, and produce virtual sound adaptively based on the location information associated with the listener's movement. The location information may include distance information and direction information regarding the listener relative to the motion sensor.

Abstract: The disclosure describes a method for a multi-channel speaker system including N speakers. The method may comprise obtaining N! permutations of channel sequence for the N speakers; determining, for each permutation, a voting score that represents the matching degree between the channel sequence indicated in a permutation and a correct channel assignment sequence of the N speakers; selecting the permutation with the highest voting score; and assigning input source channels to the N speakers in the order of the channel sequence indicated in the selected permutation.

Abstract: A novel orally administrable dosage form including a drug/active layer for loading a therapeutic agent and an extension layer for retaining the API or drug dosage form in the stomach of a subject in need thereof. Also disclosed is a method of treating diseases with the dosage form.

Abstract: An expanded spinal fusion cage is provided and includes: an outer frame; a sliding block set with a middle sliding block located within the outer frame, and the middle sliding block is located between two outer sliding blocks; a screw rod penetrating through and combined with the outer frame, and the screw rod is screwed with the middle sliding block, so that the middle sliding block is moved in translation in the outer frame and simultaneously expands the two outer sliding blocks by rotating the screw rod; two curved surface elements located outside the outer frame and combined with the two outer sliding blocks respectively, each of the curved surface elements has a wing plate; and two vertebral arch screws penetrating through and combined with the two wing plates.

Abstract: In one aspect, a first playback device is configured to (i) receive a set of voice signals, (ii) process the set of voice signals using a first set of audio processing algorithms, (iii) identify, from the set of voice signals, at least two voice signals that are to be further processed, (iv) determine that the first playback device does not have a threshold amount of computational power available, (v) receive an indication of an available amount of computational power of a second playback device, (vi) send the at least two voice signals to the second playback device, (vii) cause the second playback device to process the at least two voice signals using a second set of audio processing algorithms, (viii) receive, from the second playback device, the processed at least two voice signals, and (ix) combine the processed at least two voice signals into a combined voice signal.

Abstract: A method of room calibration comprises measuring a plurality of impulse responses at a plurality of measurement points in a room for each speaker of a plurality of speakers. The method also comprises determining a plurality of transfer functions at the plurality of measurement points for each speaker based on the plurality of impulse responses. Furthermore, the method also comprises weighting and summing the transfer functions to obtain a weighted and summed sound curve for each speaker.

Abstract: A method includes receiving, as input, reference audio data representing a reference audio signal captured by an audio input device. The method also includes receiving, as input, from a beamformer, spatially-filtered audio data representing an output of the beamformer, the beamformer configured to spatially filter, based on additional audio data captured by one or more additional audio input devices, the reference audio data to attenuate one or more interfering signals in the spatially-filtered audio data. The method processes, using a trained guided speech-enhancement network, the reference audio data and the spatially-filtered audio data to generate, as output, enhanced audio data, the guided speech-enhancement network processing the reference audio data and the spatially-filtered audio data to further attenuate, in the enhanced audio data, the one or more interfering signals attenuated by the beamformer.

Abstract: An audio playback device comprises a microphone, a speaker, and a processor. The processor is arranged to output by the speaker first audio content and receive by the microphone an indication of the first audio content. A first acoustic response of a room in which the audio playback device is located is determined based on the received indication of first audio content. A mapping is applied to the first acoustic response to determine a second acoustic response. The second acoustic response is indicative of an approximated acoustic response of the room at a spatial location different from a spatial location of the microphone. The second audio content output by the speaker is adjusted based on the second response.

Abstract: Yes A keyword spotting method based on a neural network (NN) acoustic model is provided to tolerate for dynamically adding and deleting keywords by remapping new keywords as individual Acoustic Model Sequences. The method compares sequence matching in the phoneme spaces instead of directly in predetermined acoustic space. Therefore, the acoustic model cross comparison model is relaxed from global optimization to local minimum distance to each distribution.

Abstract: An integrated antenna structure includes a grounding element, a plurality of antenna elements, and a plurality of isolators. The grounding element has an opening. The antenna elements and the isolators are disposed at an outer edge of the grounding element and an edge of the opening. The isolators are located between the antenna elements. Orthogonal projections of the antenna elements and the isolators completely overlap with the grounding element.

Abstract: The disclosure provides a method and a system for head-related transfer function (HRTF) adaptation. The method includes performing a system identification. The system identification includes a pinna identification and a shadowing identification.

Abstract: A display panel repair device includes a supporting part, a picking part, and a debonding part. The supporting part is configured to carry a display panel and control a movement of the display panel. The display panel includes a display substrate and a plurality of light-emitting diodes (LEDs) fixing on the display substrate by binding material. The picking part is located on a side of the supporting part used to carry the display panel and configured for removing a LED from the display substrate and placing a new LED on the display substrate. The debonding part is configured to heat the display substrate from a side of the display substrate away from the picking part to melt the binding material.