Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into a neural-network pipeline, the neural-network pipeline including a convolutional network that receives the audio signal and provides a first output of the convolutional network to an enhancer, the enhancer including a deep complex convolutional recurrent network that receives the first output along with a mel spectrogram of the audio signal and outputs a second output to at least one of a vocoder and a decoder, and control an output of an enhanced audio signal from the at least one of the vocoder and the decoder.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio input, implement, by semantic voice activity detection (VAD) of a dialog manager (DM) trained as a first large language model (LLM), any of a user state detection and a user intention analysis, determine, by the DM and depending on a result of any of the user state detection and the user intention analysis, whether to activate at least part of a core dialog engine (CDE) to process the audio input, and control, by the CDE in response to being activated by the DM, output of a response to any of the query spoken by the user and the other verbal input as a result of the DM determining that the any of the query spoken by the user and the other verbal input requires response from the SDS.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into a neural-network based AHS model, train the neural-network based AHS model based on input signals which are recursively generated from the audio signal during training of the AHS model, and output an AHS signal from the neural-network based AHS model in which AHS is applied to the audio signal, wherein the AHS signal is a version of the audio signal in which acoustic howling noise of the audio signal is suppressed and target audio of the audio signal is sustained.

Abstract: Method, apparatus, and non-transitory storage medium for training a deep neural-network model jointly for acoustic echo suppression and acoustic howling suppression are provided. The method may include generating a teacher speech signal for training the deep neural-network model based on a input speech from a speech system and at least one reference signal. The deep neural-network model is trained jointly for both acoustic echo suppression and acoustic howling suppression by using the teacher speech signal and a correlation loss. During training of the deep neural-network model, the training task formulates a recurrent feedback suppression process as an instantaneous speech separation task using the teacher-forced training strategy.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into frequency-domain Kalman filter (FDKF), input the audio signal and an output from the FDKF into a neural network, estimate, based on the audio signal and the output from the FDKF, and removing feedback signals from the audio signal by the neural network, recover, by a codec receiving an output from the neural network, vocal quality of a target vocal signal, and output a version of the audio signal in which the target vocal signal is enhanced by removal of the feedback signals from the audio signal by the neural network and by recovery of the vocal quality by the codec.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive multiple audio signals obtained from ones of a plurality of microphones of a microphone array, implement an audio zooming based on the audio signals by selectively focusing and enhancing first ones of the audio signals and by attenuating other ones of the audio signals, and control an output of audio based on the audio zooming, and the audio zooming includes a consolidating of a plurality of directional features of the first ones of the audio signals within a field around the microphone array and a countering based on determining directional aspects of the other ones of the audio signals from outside of the field.

Abstract: A method includes receiving a mixture signal comprising at least a first speaker, a second speaker, and background noise, the first speaker having a first distance to a microphone that outputs the mixture signal, the second speaker having a second distance to the microphone; training one or more neural networks to output a target channel and an interference channel by: inputting, into the one or more neural networks, the mixture signal and a task ID associated with one of the first speaker and the second speaker as a target speaker; determining a loss function based on the first distance and the second distance; and updating the neural network based on the loss function.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into a neural-network pipeline, the neural-network pipeline including a convolutional network that receives the audio signal and provides a first output of the convolutional network to an enhancer, the enhancer including a deep complex convolutional recurrent network that receives the first output along with a mel spectrogram of the audio signal and outputs a second output to at least one of a vocoder and a decoder, and control an output of an enhanced audio signal from the at least one of the vocoder and the decoder.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive multiple audio signals obtained from ones of a plurality of microphones of a microphone array, implement an audio zooming based on the audio signals by selectively focusing and enhancing first ones of the audio signals and by attenuating other ones of the audio signals, and control an output of audio based on the audio zooming, and the audio zooming includes a consolidating of a plurality of directional features of the first ones of the audio signals within a field around the microphone array and a countering based on determining directional aspects of the other ones of the audio signals from outside of the field.

Abstract: A method includes inputting a mixture signal comprising speech, background noise, and reverberation into a denoising stage that generates a denoised mixture signal with reduced noise compared to the mixture signal; fusing the denoised mixture signal with the mixture signal to generate a fused signal; and inputting the fused signal into a dereverberation and restoration (DR) stage to generate an output signal with reduced reverberation compared to the mixture signal.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to obtain an audio signal from a channel of at least one microphone of a plurality of microphones, estimate a Kalman gain based on the audio signal, share the Kalman gain to a plurality of channels of other ones of the plurality of microphones, and output an AHS signal from the channel and the plurality of channels, wherein the AHS signal is a version of the audio signal in which acoustic howling noise of the audio signal is suppressed and target audio of the audio signal is sustained.

Abstract: Method, apparatus, and non-transitory storage medium for hybrid acoustic howling suppression based on a frequency filter model and a deep neural network are provided. The method may include receiving a speech signal, the speech signal including target speech, feedback, and noise, and inputting the speech signal into a trained hybrid neural-network based howling suppression model, wherein the trained hybrid neural-network based howling suppression model is trained using training speech signal and pre-processed acoustic feedback from a first frequency filter model. The method may also include generating an enhanced speech signal with suppressed howling as an output of the trained hybrid neural-network based howling suppression model, wherein the enhanced speech signal is used to update parameters of the first frequency filter model.

Abstract: A roadside photovoltaic site selection method considering glare impact includes: obtaining latitudes, longitudes, ground elevations, and contour vertex eights of photovoltaic modules above ground of a photovoltaic array; calculating a sun movement path, determining an installation manner based on predicted solar radiation on inclined surfaces of the photovoltaic array; calculating an elevation angle and an azimuth angle of each reflected light caused by the inclined surfaces; arranging observation points and conducting a glare impact analysis for each observation point; determining whether to adjust an installation tilt angle of each photovoltaic module or distances between adjacent photovoltaic modules; comparing solar radiation received on the inclined surfaces between the critical tilt angle and the optimal tilt angle; in response to a difference in the solar radiation less than a set value, accepting the critical tilt angle; and completing the glare impact analysis for each observation point, and providin

Abstract: A roadside photovoltaic site selection method considering glare impact includes: obtaining latitudes, longitudes, ground elevations, and contour vertex eights of photovoltaic modules above ground of a photovoltaic array; calculating a sun movement path, determining an installation manner based on predicted solar radiation on inclined surfaces of the photovoltaic array; calculating an elevation angle and an azimuth angle of each reflected light caused by the inclined surfaces; arranging observation points and conducting a glare impact analysis for each observation point; determining whether to adjust an installation tilt angle of each photovoltaic module or distances between adjacent photovoltaic modules; comparing solar radiation received on the inclined surfaces between the critical tilt angle and the optimal tilt angle; in response to a difference in the solar radiation less than a set value, accepting the critical tilt angle; and completing the glare impact analysis for each observation point, and providin

Abstract: Method, apparatus, and non-transitory storage medium for training a deep neural-network model jointly for acoustic echo suppression and acoustic howling suppression are provided. The method may include generating a teacher speech signal for training the deep neural-network model based on a input speech from a speech system and at least one reference signal. The deep neural-network model is trained jointly for both acoustic echo suppression and acoustic howling suppression by using the teacher speech signal and a correlation loss. During training of the deep neural-network model, the training task formulates a recurrent feedback suppression process as an instantaneous speech separation task using the teacher-forced training strategy.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to obtain an audio signal from a channel of at least one microphone of a plurality of microphones, estimate a Kalman gain based on the audio signal, share the Kalman gain to a plurality of channels of other ones of the plurality of microphones, and output an AHS signal from the channel and the plurality of channels, wherein the AHS signal is a version of the audio signal in which acoustic howling noise of the audio signal is suppressed and target audio of the audio signal is sustained.

Abstract: Method, apparatus, and non-transitory storage medium for hybrid acoustic howling suppression based on a frequency filter model and a deep neural network are provided. The method may include receiving a speech signal, the speech signal including target speech, feedback, and noise, and inputting the speech signal into a trained hybrid neural-network based howling suppression model, wherein the trained hybrid neural-network based howling suppression model is trained using training speech signal and pre-processed acoustic feedback from a first frequency filter model. The method may also include generating an enhanced speech signal with suppressed howling as an output of the trained hybrid neural-network based howling suppression model, wherein the enhanced speech signal is used to update parameters of the first frequency filter model.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into frequency-domain Kalman filter (FDKF), input the audio signal and an output from the FDKF into a neural network, estimate, based on the audio signal and the output from the FDKF, and removing feedback signals from the audio signal by the neural network, and output a version of the audio signal in which a target vocal signal is enhanced by removal of the feedback signals from the audio signal by the neural network.

Abstract: A method and apparatus comprising computer code configured to cause a processor or processors to receive an audio signal obtained from a microphone, input the audio signal into frequency-domain Kalman filter (FDKF), input the audio signal and an output from the FDKF into a neural network, estimate, based on the audio signal and the output from the FDKF, and removing feedback signals from the audio signal by the neural network, recover, by a codec receiving an output from the neural network, vocal quality of a target vocal signal, and output a version of the audio signal in which the target vocal signal is enhanced by removal of the feedback signals from the audio signal by the neural network and by recovery of the vocal quality by the codec.

Abstract: A method includes receiving a mixture signal comprising at least a first speaker, a second speaker, and background noise, the first speaker having a first distance to a microphone that outputs the mixture signal, the second speaker having a second distance to the microphone; training one or more neural networks to output a target channel and an interference channel by: inputting, into the one or more neural networks, the mixture signal and a task ID associated with one of the first speaker and the second speaker as a target speaker; determining a loss function based on the first distance and the second distance; and updating the neural network based on the loss function.