公开(公告)号：US10482896B2

公开(公告)日：2019-11-19

申请号：US15991811

申请日：2018-05-29

Applicant: Retune DSP ApS

Inventor： Ulrik Kjems ,  Thomas Krogh Andersen

IPC: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

Abstract: The present invention relates to a multi-band noise reduction system for digital audio signals producing a noise reduced digital audio output signal from a digital audio signal. The digital audio signal comprises a target signal and a noise signal, i.e. a noisy digital audio signal. The multi-band noise reduction system operates on a plurality of sub-band signals derived from the digital audio signal and comprises a second or adaptive signal-to-noise ratio estimator which is configured for filtering a plurality of first signal-to-noise ratio estimates of the plurality of sub-band signals with respective time-varying low-pass filters to produce respective second signal-to-noise ratio estimates of the plurality of sub-band signals. A low-pass cut-off frequency of each of the time-varying low-pass filters is adaptable in accordance with a first signal-to-noise ratio estimate determined by a first signal-to-noise ratio estimator and/or the second signal-to-noise ratio estimate of the sub-band signal.

公开(公告)号：US10109290B2

公开(公告)日：2018-10-23

申请号：US15318046

申请日：2015-06-10

Applicant: Retune DSP ApS

Inventor： Ulrik Kjems ,  Thomas Krogh Andersen

IPC: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

Abstract: The present invention relates to a multi-band noise reduction system for digital audio signals producing a noise reduced digital audio output signal from a digital audio signal. The digital audio signal comprises a target signal and a noise signal, i.e. a noisy digital audio signal. The multi-band noise reduction system operates on a plurality of sub-band signals derived from the digital audio signal and comprises a second or adaptive signal-to-noise ratio estimator which is configured for filtering a plurality of first signal-to-noise ratio estimates of the plurality of sub-band signals with respective time-varying low-pass filters to produce respective second signal-to-noise ratio estimates of the plurality of sub-band signals. A low-pass cut-off frequency of each of the time-varying low-pass filters is adaptable in accordance with a first signal-to-noise ratio estimate determined by a first signal-to-noise ratio estimator and/or the second signal-to-noise ratio estimate of the sub-band signal.

公开(公告)号：US20180277139A1

公开(公告)日：2018-09-27

申请号：US15991811

申请日：2018-05-29

Applicant: Retune DSP ApS

Inventor： Ulrik Kjems ,  Thomas Krogh Andersen

IPC: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

CPC classification number: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

Abstract: The present invention relates to a multi-band noise reduction system for digital audio signals producing a noise reduced digital audio output signal from a digital audio signal. The digital audio signal comprises a target signal and a noise signal, i.e. a noisy digital audio signal. The multi-band noise reduction system operates on a plurality of sub-band signals derived from the digital audio signal and comprises a second or adaptive signal-to-noise ratio estimator which is configured for filtering a plurality of first signal-to-noise ratio estimates of the plurality of sub-band signals with respective time-varying low-pass filters to produce respective second signal-to-noise ratio estimates of the plurality of sub-band signals. A low-pass cut-off frequency of each of the time-varying low-pass filters is adaptable in accordance with a first signal-to-noise ratio estimate determined by a first signal-to-noise ratio estimator and/or the second signal-to-noise ratio estimate of the sub-band signal.

公开(公告)号：US20170125033A1

公开(公告)日：2017-05-04

申请号：US15318046

申请日：2015-06-10

Applicant: Retune DSP ApS

Inventor： Ulrik KJEMS ,  Thomas Krogh ANDERSEN

IPC: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

CPC classification number: G10L21/0232 ,  G10L21/0316 ,  G10L21/038

Abstract: The present invention relates to a multi-band noise reduction system for digital audio signals producing a noise reduced digital audio output signal from a digital audio signal. The digital audio signal comprises a target signal and a noise signal, i.e. a noisy digital audio signal. The multi-band noise reduction system operates on a plurality of sub-band signals derived from the digital audio signal and comprises a second or adaptive signal-to-noise ratio estimator which is configured for filtering a plurality of first signal-to-noise ratio estimates of the plurality of sub-band signals with respective time-varying low-pass filters to produce respective second signal-to-noise ratio estimates of the plurality of sub-band signals. A low-pass cut-off frequency of each of the time-varying low-pass filters is adaptable in accordance with a first signal-to-noise ratio estimate determined by a first signal-to-noise ratio estimator and/or the second signal-to-noise ratio estimate of the sub-band signal.

公开(公告)号：US10412488B2

公开(公告)日：2019-09-10

申请号：US15753139

申请日：2016-07-18

Applicant: Retune DSP ApS

Inventor： Ulrik Kjems ,  Thomas Krogh Andersen

IPC: G10L15/20 ,  H04R3/00 ,  G10L21/0216 ,  G10L25/84 ,  G10L21/0232 ,  G10L15/08

Abstract: The present invention relates in one aspect to a microphone array signal processing system comprising a digital buffer coupled to a signal input and configured to store first and second digital audio signals. A beamformer analyzer is configured to, in response to a first voice trigger, determine noise statistics based on the first signal segment of the first digital audio signal and a first signal segment of the second digital audio signal. A coefficients calculator is configured to calculate a first set of fixed beamformer coefficients of a beamforming algorithm. The beamforming algorithm is configured for applying the first set of fixed beamformer coefficients to the first signal segments of the first and second digital audio signals retrieved from the digital buffer to produce a noise reduced digital audio signal.

公开(公告)号：US20180249246A1

公开(公告)日：2018-08-30

申请号：US15753139

申请日：2016-07-18

Applicant: Retune DSP ApS

Inventor： Ulrik KJEMS ,  Thomas Krogh ANDERSEN

IPC: H04R3/00 ,  G10L21/0232 ,  G10L25/84 ,  G10L15/20

CPC classification number: H04R3/005 ,  G10L15/20 ,  G10L21/0216 ,  G10L21/0232 ,  G10L25/84 ,  G10L2015/088 ,  G10L2021/02166 ,  H04R2499/11

Abstract: The present invention relates in one aspect to a microphone array signal processing system comprising a digital buffer coupled to a signal input and configured to store first and second digital audio signals. A beamformer analyser is configured to, in response to a first voice trigger, determine noise statistics based on the first signal segment of the first digital audio signal and a first signal segment of the second digital audio signal. A coefficients calculator is configured to calculate a first set of fixed beamformer coefficients of a beamforming algorithm. The beamforming algorithm is configured for applying the first set of fixed beamformer coefficients to the first signal segments of the first and second digital audio signals retrieved from the digital buffer to produce a noise reduced digital audio signal.

公开(公告)号：US09224393B2

公开(公告)日：2015-12-29

申请号：US13967055

申请日：2013-08-14

Applicant: OTICON A/S ,  Retune DSP ApS

Inventor： Ulrik Kjems ,  Jesper Jensen

IPC: G10L21/0208 ,  G10L15/20 ,  H04R3/00 ,  G10L21/0232 ,  H04M9/08 ,  G10L21/0216 ,  H04R1/10 ,  H04R3/02

CPC classification number: G10L15/20 ,  G10L21/0232 ,  G10L2021/02166 ,  H04M9/082 ,  H04R1/1083 ,  H04R3/005 ,  H04R3/02 ,  H04R2225/43 ,  H04R2227/009

Abstract: A method comprises processing M subband communication signals and N target-cancelled signals in each subband with a set of beamformer coefficients to obtain an inverse target-cancelled covariance matrix of order N in each band; using a target absence signal to obtain an initial estimate of the noise power in a beamformer output signal averaged over recent frames with target absence in each subband; multiplying the initial noise estimate with a noise correction factor to obtain a refined estimate of the power of the beamformer output noise signal component in each subband; processing the refined estimate with the magnitude of the beamformer output to obtain a postfilter gain value in each subband; processing the beamformer output signal with the postfilter gain value to obtain a postfilter output signal in each subband; and processing the postfilter output subband signals to obtain an enhanced beamformed output signal.

Abstract translation: 一种方法包括利用一组波束形成器系数来处理每个子带中的M个子带通信信号和N个目标消除信号，以获得每个频带中N阶的逆目标消除协方差矩阵; 使用目标缺失信号来获得在每个子带中目标不存在的最近帧上平均的波束形成器输出信号中的噪声功率的初始估计; 将初始噪声估计与噪声校正因子相乘以获得每个子带中波束形成器输出噪声信号分量的功率的精确估计; 用波束形成器输出的幅度处理精细估计，以获得每个子带中的后置滤波器增益值; 用后置滤波器增益值处理波束形成器输出信号以获得每个子带中的后滤波器输出信号; 以及处理后置滤波器输出子带信号以获得增强的波束形成输出信号。

公开(公告)号：US20140056435A1

公开(公告)日：2014-02-27

申请号：US13967055

申请日：2013-08-14

Applicant: Retune DSP ApS ,  OTICON A/S

Inventor： Ulrik KJEMS ,  Jesper JENSEN

IPC: G10L15/20

CPC classification number: G10L15/20 ,  G10L21/0232 ,  G10L2021/02166 ,  H04M9/082 ,  H04R1/1083 ,  H04R3/005 ,  H04R3/02 ,  H04R2225/43 ,  H04R2227/009

Abstract: A method comprises processing M subband communication signals and N target-cancelled signals in each subband with a set of beamformer coefficients to obtain an inverse target-cancelled covariance matrix of order N in each band; using a target absence signal to obtain an initial estimate of the noise power in a beamformer output signal averaged over recent frames with target absence in each subband; multiplying the initial noise estimate with a noise correction factor to obtain a refined estimate of the power of the beamformer output noise signal component in each subband; processing the refined estimate with the magnitude of the beamformer output to obtain a postfilter gain value in each subband; processing the beamformer output signal with the postfilter gain value to obtain a postfilter output signal in each subband; and processing the postfilter output subband signals to obtain an enhanced beamformed output signal.

Abstract translation: 一种方法包括利用一组波束形成器系数处理每个子带中的M个子带通信信号和N个目标消除信号，以获得每个频带中的N阶逆目标消除协方差矩阵; 使用目标缺失信号来获得在每个子带中目标不存在的最近帧上平均的波束形成器输出信号中的噪声功率的初始估计; 将初始噪声估计与噪声校正因子相乘以获得每个子带中波束形成器输出噪声信号分量的功率的精确估计; 用波束形成器输出的幅度处理精细估计，以获得每个子带中的后置滤波器增益值; 用后置滤波器增益值处理波束形成器输出信号以获得每个子带中的后滤波器输出信号; 以及处理后置滤波器输出子带信号以获得增强的波束形成输出信号。