公开(公告)号：US20120185246A1

公开(公告)日：2012-07-19

申请号：US13174964

申请日：2011-07-01

Applicant: Xianxian Zhang ,  Jes Thyssen ,  Kwan Young Shin

Inventor： Xianxian Zhang ,  Jes Thyssen ,  Kwan Young Shin

IPC: G10L21/02

CPC classification number: G10L21/0208 ,  G10L2021/02161

Abstract: Techniques are described herein that suppress noise using multiple sensors (e.g., microphones) of a communication device. Noise modeling (e.g., estimation of noise basis vectors and noise weighting vectors) is performed with respect to a noise signal during operation of a communication device to provide a noise model. The noise model includes noise basis vectors and noise coefficients that represent noise provided by audio sources other than a user of the communication device. Speech modeling (e.g., estimation of speech basis vectors and speech weighting) is performed to provide a speech model. The speech model includes speech basis vectors and speech coefficients that represent speech of the user. A noisy speech signal is processed using the noise basis vectors, the noise coefficients, the speech basis vectors, and the speech coefficients to provide a clean speech signal.

Abstract translation: 本文描述了使用通信设备的多个传感器（例如，麦克风）抑制噪声的技术。 在通信设备的操作期间，相对于噪声信号执行噪声建模（例如噪声基矢量和噪声加权矢量的估计）以提供噪声模型。 噪声模型包括噪声基矢量和表示由通信设备的用户以外的音频源提供的噪声的噪声系数。 执行语音建模（例如，语音基本向量的估计和语音加权）以提供语音模型。 语音模型包括表示用户语音的语音基向量和语音系数。 使用噪声基矢量，噪声系数，语音基矢量和语音系数来处理噪声语音信号以提供干净的语音信号。

公开(公告)号：US20120123772A1

公开(公告)日：2012-05-17

申请号：US13295818

申请日：2011-11-14

Applicant: Jes Thyssen ,  Huaiyu Zeng ,  Juin-Hwey Chen ,  Nelson Sollenberger ,  Xianxian Zhang

Inventor： Jes Thyssen ,  Huaiyu Zeng ,  Juin-Hwey Chen ,  Nelson Sollenberger ,  Xianxian Zhang

IPC: G10L21/02

CPC classification number: G10L21/0208 ,  G10L21/0272 ,  G10L2021/02165 ,  H04R1/245 ,  H04R2410/07

Abstract: Multi-channel noise suppression systems and methods are described that omit the traditional delay-and-sum fixed beamformer in devices that include a primary speech microphone and at least one noise reference microphone with the desired speech being in the near-field of the device. The multi-channel noise suppression systems and methods use a blocking matrix (BM) to remove desired speech in the input speech signal received by the noise reference microphone to get a “cleaner” background noise component. Then, an adaptive noise canceler (ANC) is used to remove the background noise in the input speech signal received by the primary speech microphone based on the “cleaner” background noise component to achieve noise suppression. The filters implemented by the BM and ANC are derived using closed-form solutions that require calculation of time-varying statistics of complex frequency domain signals in the noise suppression system.

Abstract translation: 描述了多通道噪声抑制系统和方法，其省略了传统的延迟和总和固定波束形成器，其包括主话音麦克风和至少一个噪声参考麦克风，其中所需语音位于设备的近场。 多声道噪声抑制系统和方法使用阻塞矩阵（BM）去除由噪声参考麦克风接收的输入语音信号中的期望语音以获得“更干净”的背景噪声分量。 然后，使用自适应噪声消除器（ANC）来基于“更干净的”背景噪声分量来消除由主话音麦克风接收的输入语音信号中的背景噪声，以实现噪声抑制。 由BM和ANC实现的滤波器是使用需要在噪声抑制系统中计算复杂频域信号的时变统计的闭式解决方案导出的。

公开(公告)号：US08041562B2

公开(公告)日：2011-10-18

申请号：US12474927

申请日：2009-05-29

Applicant: Jes Thyssen

Inventor： Jes Thyssen

IPC: G10L21/02

CPC classification number: G10L19/0204 ,  G10L19/005 ,  G10L19/04

Abstract: A technique is described herein for reducing audible artifacts in an audio output signal generated by decoding a received frame in a series of frames representing an encoded audio signal in a predictive coding system. In accordance with the technique, it is determined if the received frame is one of a predefined number of received frames that follow a lost frame in the series of the frames. Responsive to determining that the received frame is one of the predefined number of received frames, at least one parameter or signal associated with the decoding of the received frame is altered from a state associated with normal decoding. The received frame is then decoded in accordance with the at least one parameter or signal to generate a decoded audio signal. The audio output signal is then generated based on the decoded audio signal.

Abstract translation: 本文描述了一种技术，用于通过在代表预测编码系统中的编码音频信号的一系列帧中对接收到的帧进行解码而产生的音频输出信号中减少可听见的伪影。 根据该技术，确定接收到的帧是否是在一系列帧中的丢失帧之后的预定数量的接收帧中的一个。 响应于确定接收到的帧是预定数量的接收帧之一，与所接收的帧的解码相关联的至少一个参数或信号从与正常解码相关联的状态改变。 接收的帧然后根据至少一个参数或信号被解码以产生解码的音频信号。 然后基于解码的音频信号产生音频输出信号。

公开(公告)号：US08000960B2

公开(公告)日：2011-08-16

申请号：US11838891

申请日：2007-08-15

Applicant: Juin-Hwey Chen ,  Robert W. Zopf ,  Jes Thyssen

Inventor： Juin-Hwey Chen ,  Robert W. Zopf ,  Jes Thyssen

IPC: G10L19/10

CPC classification number: G10L19/0204 ,  G10L19/005 ,  G10L19/04

Abstract: A technique is described for concealing the effect of a lost frame in a series of frames representing an encoded audio signal in a sub-band predictive coding system. In accordance with the technique, a first synthesized sub-band audio signal is synthesized, wherein synthesizing the first synthesized sub-band audio signal comprises performing waveform extrapolation based on a stored first sub-band decoded audio signal. A second synthesized sub-band audio signal is also synthesized, wherein synthesizing the second synthesized sub-band audio signal comprises performing waveform extrapolation based on the stored second sub-band decoded audio signal. The first synthesized sub-band audio signal and the second synthesized sub-band audio signal are combined to generate a synthesized full-band output audio signal corresponding to a lost frame.

Abstract translation: 描述了一种用于在子带预测编码系统中隐藏表示编码音频信号的一系列帧中的丢失帧的影响的技术。 根据该技术，合成第一合成子带音频信号，其中合成第一合成子带音频信号包括基于存储的第一子带解码音频信号执行波形外推。 还合成了第二合成子带音频信号，其中合成第二合成子带音频信号包括基于所存储的第二子带解码音频信号执行波形外推。 第一合成子带音频信号和第二合成子带音频信号被组合以产生对应于丢失帧的合成全频带输出音频信号。

公开(公告)号：US20100217590A1

公开(公告)日：2010-08-26

申请号：US12391879

申请日：2009-02-24

Applicant: Elias Nemer ,  Jes Thyssen

Inventor： Elias Nemer ,  Jes Thyssen

IPC: G10L15/20

CPC classification number: G01S3/86 ,  G01S3/8006 ,  G10L15/26 ,  G10L2021/02166

Abstract: A system and method for performing speaker localization is described. The system and method utilizes speaker recognition to provide an estimate of the direction of arrival (DOA) of speech sound waves emanating from a desired speaker with respect to a microphone array included in the system. Candidate DOA estimates may be preselected or generated by one or more other DOA estimation techniques. The system and method is suited to support steerable beamforming as well as other applications that utilize or benefit from DOA estimation. The system and method provides robust performance even in systems and devices having small microphone arrays and thus may advantageously be implemented to steer a beamformer in a cellular telephone or other mobile telephony terminal featuring a speakerphone mode.

Abstract translation: 描述用于执行扬声器定位的系统和方法。 系统和方法利用说话者识别来提供相对于包括在系统中的麦克风阵列从期望的扬声器发出的语音声波的到达方向（DOA）的估计。 候选DOA估计可以由一个或多个其它DOA估计技术预先选择或产生。 该系统和方法适用于支持可导向波束形成以及利用或受益于DOA估计的其他应用。 该系统和方法即使在具有小麦克风阵列的系统和设备中也提供了强大的性能，因此可以有利地实现以引导具有扬声器电话模式的蜂窝电话或其他移动电话终端中的波束形成器。

公开(公告)号：US07684521B2

公开(公告)日：2010-03-23

申请号：US11048916

申请日：2005-02-03

Applicant: Jes Thyssen ,  Juin-Hwey Chen ,  Nambi Seshadri

Inventor： Jes Thyssen ,  Juin-Hwey Chen ,  Nambi Seshadri

IPC: H04L27/06 ,  H03M13/00

CPC classification number: H04L1/0045

Abstract: Typical communication systems operate with a single channel decoder, and hence would have to settle for the performance from the single channel decoder regardless of the conditions of the communications channel. The present invention uses a hybrid channel decoder comprising multiple channel decoders, each configured to optimize the quality of the re-constructed signal for different channel conditions. Therefore, the desired decoder can be selected as conditions of the communications channel, or the data signal, change over time, so as to optimize the re-constructed data signal. In embodiments, the data signal is a speech signal.

Abstract translation: 典型的通信系统使用单个信道解码器进行操作，因此无论通信信道的条件如何，都必须从单信道解码器处理性能。 本发明使用包括多个信道解码器的混合信道解码器，每个信道解码器被配置为优化用于不同信道条件的重构信号的质量。 因此，可以选择期望的解码器作为通信信道的条件或数据信号随时间变化，以便优化重构的数据信号。 在实施例中，数据信号是语音信号。

公开(公告)号：US20090281800A1

公开(公告)日：2009-11-12

申请号：US12464517

申请日：2009-05-12

Applicant: Wilfrid LeBlanc ,  Juin-Hwey Chen ,  Jes Thyssen

Inventor： Wilfrid LeBlanc ,  Juin-Hwey Chen ,  Jes Thyssen

IPC: G10L19/14

CPC classification number: G10L21/0208 ,  G10L19/012 ,  G10L21/0232

Abstract: A speech intelligibility enhancement (SIE) system and method is described that improves the intelligibility of a speech signal to be played back by an audio device when the audio device is located in an environment with loud acoustic background noise. In an embodiment, the audio device comprises a near-end telephony terminal and the speech signal comprises a speech signal received over a communication network from a far-end telephony terminal for playback at the near-end telephony terminal.

Abstract translation: 描述了一种语音清晰度增强（SIE）系统和方法，当音频设备位于具有大声背景噪声的环境中时，提高了由音频设备重放的语音信号的可懂度。 在一个实施例中，音频设备包括近端电话终端，并且语音信号包括通过通信网络从远端电话终端接收的用于在近端电话终端回放的语音信号。

公开(公告)号：US20090248405A1

公开(公告)日：2009-10-01

申请号：US12474809

申请日：2009-05-29

Applicant: Juin-Hwey Chen ,  Jes Thyssen ,  Robert W. Zopf

Inventor： Juin-Hwey Chen ,  Jes Thyssen ,  Robert W. Zopf

IPC: G10L19/00

CPC classification number: G10L19/0208 ,  G10L19/005

Abstract: Systems and methods are described for performing packet loss concealment using an extrapolation of an excitation waveform in a sub-band predictive speech coder, such as an ITU-T Recommendation G.722 wideband speech coder. The systems and methods are useful for concealing the quality-degrading effects of packet loss in a sub-band predictive coder and address some sub-band architectural issues when applying excitation extrapolation techniques to such sub-band predictive coders.

Abstract translation: 描述了使用在诸如ITU-T G.722建议书G.722宽带语音编码器的子带预测语音编码器中外推激励波形来执行分组丢失隐藏的系统和方法。 这些系统和方法对于隐藏子带预测编码器中的分组丢失的质量降级效应是有用的，并且当向这种子带预测编码器应用激励外推技术时，解决某些子带架构问题。

公开(公告)号：US20090232228A1

公开(公告)日：2009-09-17

申请号：US12474927

申请日：2009-05-29

Applicant: Jes Thyssen

Inventor： Jes Thyssen

IPC: H04B14/04

CPC classification number: G10L19/0204 ,  G10L19/005 ,  G10L19/04

Abstract: A technique is described herein for reducing audible artifacts in an audio output signal generated by decoding a received frame in a series of frames representing an encoded audio signal in a predictive coding system. In accordance with the technique, it is determined if the received frame is one of a predefined number of received frames that follow a lost frame in the series of the frames. Responsive to determining that the received frame is one of the predefined number of received frames, at least one parameter or signal associated with the decoding of the received frame is altered from a state associated with normal decoding. The received frame is then decoded in accordance with the at least one parameter or signal to generate a decoded audio signal. The audio output signal is then generated based on the decoded audio signal.

Abstract translation: 本文描述了一种技术，用于通过在代表预测编码系统中的编码音频信号的一系列帧中对接收到的帧进行解码而产生的音频输出信号中减少可听见的伪影。 根据该技术，确定接收到的帧是否是在一系列帧中的丢失帧之后的预定数量的接收帧中的一个。 响应于确定接收到的帧是预定数量的接收帧之一，与所接收的帧的解码相关联的至少一个参数或信号从与正常解码相关联的状态改变。 接收的帧然后根据至少一个参数或信号被解码以产生解码的音频信号。 然后基于解码的音频信号产生音频输出信号。

公开(公告)号：US07512535B2

公开(公告)日：2009-03-31

申请号：US10183554

申请日：2002-06-28

Applicant: Juin-Hwey Chen ,  Jes Thyssen

Inventor： Juin-Hwey Chen ,  Jes Thyssen

IPC: G10L21/02 ,  G10L21/00 ,  G10L11/00

CPC classification number: G10L19/26

Abstract: A filter controller processes a decoded speech (DS) signal. The DS signal has a spectral envelope including a first plurality of formant peaks having different respective amplitudes. The controller produces, from the DS signal, a spectrally-flattened DS signal that is a time-domain signal. The spectrally-flattened time-domain DS signal has a spectral envelope including a second plurality of formant peaks. Each of the second plurality of formant peaks approximately coincides in frequency with a respective one of the first plurality of formant peaks. Also, the second plurality of formant peaks have approximately equal respective amplitudes. Next, the controller derives, from the spectrally-flattened time-domain DS signal, a set of filter coefficients representative of a filter response that is to be used to filter the DS signal.

Abstract translation: 滤波器控制器处理解码语音（DS）信号。 DS信号具有包括具有不同相应振幅的第一多个共振峰的频谱包络。 控制器从DS信号产生作为时域信号的频谱平坦化的DS信号。 频谱平坦化的时域DS信号具有包括第二多个共振峰的频谱包络。 所述第二多个共振峰中的每一个峰值与所述第一多个共振峰中的相应一个峰值大致重合。 而且，第二多个共振峰具有大致相等的相应振幅。 接下来，控制器从频谱平坦化的时域DS信号中导出代表要用于滤波DS信号的滤波器响应的一组滤波器系数。