公开(公告)号：US09754605B1

公开(公告)日：2017-09-05

申请号：US15177624

申请日：2016-06-09

Applicant: Amazon Technologies, Inc.

Inventor： Amit Singh Chhetri

IPC: G10L21/0264 ,  H04R3/04 ,  G10L25/21 ,  H04S7/00 ,  G10L21/0208

CPC classification number: G10L21/0264 ,  G10L25/06 ,  G10L25/21 ,  G10L2021/02082 ,  H04R3/005 ,  H04R3/02 ,  H04R2499/11 ,  H04S7/305

Abstract: A multi-channel acoustic echo cancellation (AEC) system that includes a step-size controller that dynamically determines a step-size value for each channel and each tone index on a frame-by-frame basis. The system determines the step-size value based on a normalized squared cross-correlation (NSCC) between an estimated echo signal and an error signal, allowing the AEC system to converge quickly when an acoustic room response changes while providing stable steady-state error by avoiding misadjustments due to noise sensitivity and/or near-end speech. The step-size value can be determined using fractional weighting that takes into account a signal strength of each channel.

公开(公告)号：US09473646B1

公开(公告)日：2016-10-18

申请号：US14028229

申请日：2013-09-16

Applicant: Amazon Technologies, Inc.

Inventor： Amit Singh Chhetri

IPC: H04M9/08

CPC classification number: H04M9/082

Abstract: An acoustic echo canceller (AEC) system may be configured to reset the coefficients of a transform equation when an estimated echo diverges from actual acoustic echo. Features are disclosed for determining when to reset the coefficients, and for enabling the reset operation to be performed reliably. Additional features are disclosed for detecting other signal conditions besides AEC divergence, for adjusting the rate at which the coefficients are adapted in response to such conditions, and for prioritizing between potentially incompatible adjustments.

Abstract translation: 声学回声消除器（AEC）系统可以被配置为当估计的回波从实际声学回波发散时，重置变换方程的系数。 公开了用于确定何时重置系数的功能，并且使得能够可靠地执行复位操作。 公开了用于检测除了AEC发散之外的其它信号条件的附加特征，用于响应于这些条件来调节系数被适应的速率，以及用于在可能不兼容的调整之间进行优先级排序。

公开(公告)号：US09390723B1

公开(公告)日：2016-07-12

申请号：US14568033

申请日：2014-12-11

Applicant: Amazon Technologies, Inc.

Inventor： John Walter McDonough, Jr. ,  Wai Chung Chu ,  Amit Singh Chhetri ,  Robert Ayrapetian

IPC: H04R3/00 ,  G10L21/02 ,  G10K11/175

CPC classification number: G10K11/175 ,  G10L21/0208 ,  G10L21/0232 ,  G10L2021/02082

Abstract: Features are disclosed for performing efficient dereverberation of speech signals captured with single- and multi-channel sensors in networked audio systems. Such features could be used in applications requiring automatic recognition of speech captured with sensors. Dereverberation is performed in the sub-band domain, and hence provides improved dereverberation performance in terms of signal quality, algorithmic delay, computational efficiency, and speed of convergence.

Abstract translation: 公开了用于对网络音频系统中的单通道和多通道传感器捕获的语音信号进行有效的去混响的特征。 这些特征可以用于需要用传感器捕获的语音自动识别的应用中。 在子带域中执行混频，从而在信号质量，算法延迟，计算效率和收敛速度方面提供改进的去混响性能。

公开(公告)号：US08983057B1

公开(公告)日：2015-03-17

申请号：US14033253

申请日：2013-09-20

Applicant: Amazon Technologies, Inc.

Inventor： Hongyang Deng ,  Amit Singh Chhetri

IPC: H04M9/08

CPC classification number: H04M9/082

Abstract: A step size controller may be used to control the rate of adaptation in an acoustic echo canceller. Step size control based on the values of adaptive coefficients (rather than, e.g., a fixed initial adaptation period) provides improved reliability and resistance to disruption. Accordingly, features are disclosed for controlling step size based on the values of adaptive coefficients.

Abstract translation: 可以使用步长控制器来控制声学回声消除器中的适应速率。 基于自适应系数的值（而不是例如固定的初始适应周期）的步长控制提供了改进的可靠性和抵抗中断的能力。 因此，公开了基于自适应系数的值来控制步长的特征。

公开(公告)号：US11521635B1

公开(公告)日：2022-12-06

申请号：US17108718

申请日：2020-12-01

Applicant: Amazon Technologies, Inc.

Inventor： Amit Singh Chhetri ,  Navin Chatlani

IPC: G10L15/20 ,  G10L25/84 ,  G10L21/0232 ,  G10L15/06 ,  G10L15/22 ,  G10L15/16 ,  G06N3/08 ,  G06N3/04 ,  G10L25/90 ,  G10L21/0216 ,  G10L21/0208

Abstract: A computing device may receive audio data from a microphone representing audio in an environment of the device, which may correspond to an utterance and noise. A model may be trained to process the audio data to cancel noise from the audio data. The model may include an encoder that includes one or more dense layers, one or more recurrent layers, and a decoder that includes one or more dense layers.

公开(公告)号：US11218802B1

公开(公告)日：2022-01-04

申请号：US16141012

申请日：2018-09-25

Applicant: Amazon Technologies, Inc.

Inventor： Srivatsan Kandadai ,  Amit Singh Chhetri ,  Trausti Thor Kristjansson

IPC: H04R3/00 ,  G06F3/16 ,  G10L25/60 ,  G10L25/84 ,  H04R1/40

Abstract: A mobile device capable of capturing voice commands includes a beamformer for determining audio data corresponding to one or more directions and a beam selector for selecting in which direction a source of target audio lies. The device determines, based on data from one or more sensors, an angle through which the device has rotated. Based on the angle and one or more rotation-compensation functions, the device interpolates audio data corresponding to the one or more directions to compensate for the rotation such that the direction corresponding to the source of target audio remains selected.

公开(公告)号：US10452116B1

公开(公告)日：2019-10-22

申请号：US15620782

申请日：2017-06-12

Applicant: Amazon Technologies, Inc.

Inventor： David William Devries ,  Stewart Robin Shearer ,  Amit Singh Chhetri ,  Serkan Hatipoglu ,  Omar Sze Leung ,  Michael Serge Devyver ,  Leo Benedict Baldwin ,  Noam Sorek

IPC: G06F1/32 ,  G06F3/041 ,  G01P15/09 ,  G06F1/3231 ,  G06F3/16 ,  G01P15/097 ,  G06F1/3287

Abstract: A decision engine executing on an electronic device may determine, using sensor data captured by multiple sensors of the device, whether a user is present in an environment that includes the device. If the user is determined to be present in the environment, the device may transition from a first state to a second state. The first state may be a first power state of the device in which the device is powered off or an idle or dormant state in which the device is powered on but a display of the device is powered off. Correspondingly, the second state may be a second power state of the device in which the device and the display are powered on and content is being rendered on the display. If the decision engine cannot make a determination based on the sensor data, a context engine may adjudicate the user presence determination.

公开(公告)号：US10304475B1

公开(公告)日：2019-05-28

申请号：US15676273

申请日：2017-08-14

Applicant: Amazon Technologies, Inc.

Inventor： Rui Wang ,  Amit Singh Chhetri ,  Xiaoxue Li ,  Trausti Thor Kristjansson ,  Philip Ryan Hilmes

IPC: G10L15/00 ,  G10L15/04 ,  G10L15/16 ,  G10L15/20 ,  G10L21/00 ,  G10L21/02 ,  G10L25/00 ,  G06F17/20 ,  G06F17/30 ,  G06F7/00 ,  G10L21/0216 ,  G10L15/22 ,  G10L15/26 ,  G10L15/08

Abstract: An audio capture device that incorporates a beamformer and beam-specific trigger word detection. Audio data from each beam is processed by a low power trigger word detector, such as a neural network or other trained model to detect if audio data (such as an audio frame or feature vector corresponding thereto) likely includes part of a trigger word. The beam that either most strongly represents a trigger word portion or represents a trigger word portion most early in time may be selected for further processing such as speech processing or confirmation by a more robust power intensive trigger word detector.

公开(公告)号：US10237647B1

公开(公告)日：2019-03-19

申请号：US15446557

申请日：2017-03-01

Applicant: Amazon Technologies, Inc.

Inventor： Amit Singh Chhetri

IPC: H04R3/00 ,  H04R1/40

Abstract: A beamformer system that can isolate a desired portion of an audio signal resulting from a microphone array. A combination of beamformers is used to dampen undesired noise, whether diffuse or coherent. A fixed beamformer is used to dampen diffuse noise while an adaptive beamformer is used to cancel directional coherent noise. The adaptive beamformer isolates and weights audio from various directions. The weights may vary depending on the isolated desired audio signal, dynamically adjusting the step-size adjustments to the weights.

公开(公告)号：US10147439B1

公开(公告)日：2018-12-04

申请号：US15474197

申请日：2017-03-30

Applicant: Amazon Technologies, Inc.

Inventor： Trausti Thor Kristjansson ,  Mohamed Mansour ,  Amit Singh Chhetri ,  Ludger Solbach

IPC: G06F3/00 ,  G10L21/0364 ,  G10L15/22 ,  G10L21/0232 ,  G10L13/00 ,  G10L21/0216

Abstract: A speech-capturing device that can modulate its output audio data volume based on environmental sound conditions at the location of a user speaking to the device. The device detects the sound pressure of a spoken utterance at the device location and determines the distance of the user from the device. The device also detects the sound pressure of noise at the device and uses information about the location of the noise source and user to determine the sound pressure of noise at the location of the talker. The device can then adjust the gain for output audio (such as a spoken response to the utterance) to ensure that the output audio is at a certain desired sound pressure when it reaches the location of the user.