公开(公告)号：US20180040333A1

公开(公告)日：2018-02-08

申请号：US15227885

申请日：2016-08-03

Applicant: Apple Inc.

Inventor： Jason Wung ,  Ramin Pishehvar ,  Daniele Giacobello ,  Joshua D. Atkins

IPC: G10L21/0232 ,  G10L25/87 ,  G10L25/30

CPC classification number: G10L21/0232 ,  G10L25/30 ,  G10L25/87 ,  G10L2021/02082

Abstract: Method for performing speech enhancement using a Deep Neural Network (DNN)-based signal starts with training DNN offline by exciting a microphone using target training signal that includes signal approximation of clean speech. Loudspeaker is driven with a reference signal and outputs loudspeaker signal. Microphone then generates microphone signal based on at least one of: near-end speaker signal, ambient noise signal, or loudspeaker signal. Acoustic-echo-canceller (AEC) generates AEC echo-cancelled signal based on reference signal and microphone signal. Loudspeaker signal estimator generates estimated loudspeaker signal based on microphone signal and AEC echo-cancelled signal. DNN receives microphone signal, reference signal, AEC echo-cancelled signal, and estimated loudspeaker signal and generates a speech reference signal that includes signal statistics for residual echo or for noise. Noise suppressor generates a clean speech signal by suppressing noise or residual echo in the microphone signal based on speech reference signal. Other embodiments are described.

公开(公告)号：US20150112672A1

公开(公告)日：2015-04-23

申请号：US14517700

申请日：2014-10-17

Applicant: Apple Inc.

Inventor： Daniele Giacobello ,  Jason Wung ,  Joshua Atkins ,  Ramin Pichevar ,  Raghavendra Prabhu

IPC: G10L21/0208 ,  G10L15/20

CPC classification number: G10L21/0208 ,  H04M9/082

Abstract: An echo canceller can be arranged to receive an input signal and to receive a reference signal. The echo canceller can subtract a linear component of the reference signal from the input signal. A noise suppressor can suppress non-linear effects of the reference signal in the input signal in correspondence with a large number of selectable parameters. Such suppression can be provided on a frequency-by-frequency basis, with a unique set of tunable parameters selected for each frequency. A degree of suppression provided by the noise suppressor can correspond to an estimate of residual echo remaining after the one or more linear components of the reference signal have been subtracted from the input signal, to an estimated double-talk probability, and to an estimated signal-to-noise ratio of near-end speech in the input signal for each respective frequency. A speech recognizer can receive a processed input signal from the noise suppressor.

Abstract translation: 可以设置回波消除器以接收输入信号并接收参考信号。 回波消除器可以从输入信号中减去参考信号的线性分量。 噪声抑制器可以抑制与大量可选参数对应的输入信号中的参考信号的非线性效应。 可以在频率基础上提供这种抑制，并为每个频率选择一组唯一的可调参数。 由噪声抑制器提供的抑制程度可以对应于从输入信号中减去参考信号的一个或多个线性分量之后剩余回波的估计值，估计的双方通话概率，以及估计的信号 每个相应频率的输入信号中的近端语音的噪声比。 语音识别器可以从噪声抑制器接收经处理的输入信号。

公开(公告)号：US09633671B2

公开(公告)日：2017-04-25

申请号：US14517700

申请日：2014-10-17

Applicant: Apple Inc.

Inventor： Daniele Giacobello ,  Jason Wung ,  Joshua Atkins ,  Ramin Pichevar ,  Raghavendra Prabhu

IPC: G10L15/20 ,  G10L21/0208 ,  H04M9/08

CPC classification number: G10L21/0208 ,  H04M9/082

Abstract: An echo canceller can be arranged to receive an input signal and to receive a reference signal. The echo canceller can subtract a linear component of the reference signal from the input signal. A noise suppressor can suppress non-linear effects of the reference signal in the input signal in correspondence with a large number of selectable parameters. Such suppression can be provided on a frequency-by-frequency basis, with a unique set of tunable parameters selected for each frequency. A degree of suppression provided by the noise suppressor can correspond to an estimate of residual echo remaining after the one or more linear components of the reference signal have been subtracted from the input signal, to an estimated double-talk probability, and to an estimated signal-to-noise ratio of near-end speech in the input signal for each respective frequency. A speech recognizer can receive a processed input signal from the noise suppressor.

公开(公告)号：US10074380B2

公开(公告)日：2018-09-11

申请号：US15227885

申请日：2016-08-03

Applicant: Apple Inc.

Inventor： Jason Wung ,  Ramin Pishehvar ,  Daniele Giacobello ,  Joshua D. Atkins

IPC: G10L21/02 ,  G10L21/0232 ,  G10L25/30 ,  G10L25/87 ,  G10L21/0208

CPC classification number: G10L21/0232 ,  G10L25/30 ,  G10L25/87 ,  G10L2021/02082

Abstract: Method for performing speech enhancement using a Deep Neural Network (DNN)-based signal starts with training DNN offline by exciting a microphone using target training signal that includes signal approximation of clean speech. Loudspeaker is driven with a reference signal and outputs loudspeaker signal. Microphone then generates microphone signal based on at least one of: near-end speaker signal, ambient noise signal, or loudspeaker signal. Acoustic-echo-canceller (AEC) generates AEC echo-cancelled signal based on reference signal and microphone signal. Loudspeaker signal estimator generates estimated loudspeaker signal based on microphone signal and AEC echo-cancelled signal. DNN receives microphone signal, reference signal, AEC echo-cancelled signal, and estimated loudspeaker signal and generates a speech reference signal that includes signal statistics for residual echo or for noise. Noise suppressor generates a clean speech signal by suppressing noise or residual echo in the microphone signal based on speech reference signal. Other embodiments are described.