Abstract: The present disclosure provides method and system for voice separation based on DUET algorithm, and the method comprises receiving signals from microphones; performing a Fourier transform on the received signals; calculating a relative attenuation parameter and a relative delay parameter for each data point; selecting a clustering range for the relative delay parameters based on a distance between the microphones and a sampling frequency of the microphones, clustering the data points within the clustering range for the relative delay parameters into subsets, and performing an inverse Fourier transform on each subsets. According to the present disclosure, it is possible to provide an efficient and intelligent solution to deploy DUET on the software and/or hardware.

Abstract: A method and system for compensating a frequency response of a microphone array are provided. The method includes: multiple microphones in a microphone array respectively receiving a compensation signal sent from a calibration speaker and outputting multiple output signals respectively; determining a uniform frequency response of the multiple microphones based on the multiple output signals; calculating a compensation gain for each of the multiple microphones according to the uniform frequency response; and storing the calculated compensation gain of each microphone.

Abstract: Aspects disclosed herein generally relate to a method and a system for improving voice separation by eliminating overlaps or overlapping points. The time-frequency points from the two recorded mixtures are separated by using a Degenerate unmixing estimation technique (DUET) algorithm. The method or system further eliminates the overlapping time-frequency points which belongs to neither of the original resources of sounds.

Abstract: A method and a system for speech enhancement including a time synchronization unit configured to synchronize microphone signals sent from at least two microphones; a source separation unit configured to separate the synchronized microphone signals and output a separated speech signal, which corresponds to a speech source; and a noise reduction unit including a feature extraction unit configured to extract a speech feature of the separated speech signal and a neural network configured to receive the speech feature and output a clean speech feature.

Abstract: Methods and devices for conducting, based on a clock difference, a synchronization process on voice information collected by a plurality of voice collection devices. Then, after the synchronization process is performed on the voice information collected by the plurality of voice collection devices, conducting a voice separation and recognition process on voice information that was collected by the plurality of voice collection devices and synchronized based on the clock difference among the plurality of voice collection devices.

Abstract: A method and system for compensating a frequency response of a microphone array are provided. The method includes: multiple microphones in a microphone array respectively receiving a compensation signal sent from a calibration speaker and outputting multiple output signals respectively; determining a uniform frequency response of the multiple microphones based on the multiple output signals; calculating a compensation gain for each of the multiple microphones according to the uniform frequency response; and storing the calculated compensation gain of each microphone.

Abstract: The present disclosure is directed to a speech separation method and system using a sliding window. The method comprises: acquiring at least one speech from at least one user by at least one microphone and storing the at least one speech as a speech signal in a sound recording module; extracting the speech signal from the sound recording module and processing the extracted speech signal through a sliding window; and transmitting the processed speech signal to a Degenerate Unmixing Estimation Technique (DUET) module for speech separation.

Abstract: The present disclosure provides method and system for voice separation based on DUET algorithm, and the method comprises receiving signals from microphones; performing a Fourier transform on the received signals; calculating a relative attenuation parameter and a relative delay parameter for each data point; selecting a clustering range for the relative delay parameters based on a distance between the microphones and a sampling frequency of the microphones, clustering the data points within the clustering range for the relative delay parameters into subsets, and performing an inverse Fourier transform on each subsets. According to the present disclosure, it is possible to provide an efficient and intelligent solution to deploy DUET on the software and/or hardware.

Abstract: A method and system for compensating a frequency response of a microphone array are provided. The method includes: multiple microphones in a microphone array respectively receiving a compensation signal sent from a calibration speaker and outputting multiple output signals respectively; determining a uniform frequency response of the multiple microphones based on the multiple output signals; calculating a compensation gain for each of the multiple microphones according to the uniform frequency response; and storing the calculated compensation gain of each microphone.

Abstract: A method and a system for speech enhancement including a time synchronization unit configured to synchronize microphone signals sent from at least two microphones; a source separation unit configured to separate the synchronized microphone signals and output a separated speech signal, which corresponds to a speech source; and a noise reduction unit including a feature extraction unit configured to extract a speech feature of the separated speech signal and a neural network configured to receive the speech feature and output a clean speech feature.

Abstract: A method and device for voice recognition are provided. The method for voice recognition includes: receiving voice information collected by a plurality of voice collection devices configured on a plurality of devices connected together, wherein the plurality of devices comprises at least one mobile device; and conducting a voice separation and recognition process on the voice information collected by the plurality of voice collection devices. The device for voice recognition includes: a processor, configured to: obtain a clock difference among a plurality of devices connected together, where each of the plurality of devices is configured with a voice collection device; obtain voice information collected by the voice collection device configured on the each of the plurality of voice collection devices; and conduct a voice separation and recognition process on voice information collected by the voice collection device based on the clock difference among the among a plurality of devices.