公开(公告)号：US08254588B2

公开(公告)日：2012-08-28

申请号：US12288922

申请日：2008-10-24

Applicant: Muralidhar Karthik ,  George Sapna ,  Anoop Kumar Krishna

Inventor： Muralidhar Karthik ,  George Sapna ,  Anoop Kumar Krishna

IPC: H04B3/20

CPC classification number: H04M3/002 ,  H04B3/234 ,  H04B3/238 ,  H04M9/082

Abstract: A system and method for Acoustic Echo Cancellation. The system and method include a subband affine projection filter and a variable step size controller configured to cancel an estimated echo from a near-end signal. The system and method also include a divergence detector adapted to reset the subband affine projection filter in response to determining a divergence is occurring. Additionally, the system and method include a double talk detector adapted to transmit a signal to mask an output signal when double talk is detected.

Abstract translation: 声学回波消除的系统和方法。 该系统和方法包括子带仿射投影滤波器和可变步长控制器，其被配置为从近端信号中消除估计的回波。 所述系统和方法还包括发散检测器，其适于响应于确定发生发散而重新设置子带仿射投影滤波器。 此外，该系统和方法包括双重通话检测器，适用于在检测到双方通话时发送信号以屏蔽输出信号。

公开(公告)号：US20090123002A1

公开(公告)日：2009-05-14

申请号：US12288922

申请日：2008-10-24

Applicant: Muralidhar Karthik ,  George Sapna ,  Anoop Kumar Krishna

Inventor： Muralidhar Karthik ,  George Sapna ,  Anoop Kumar Krishna

IPC: H04B3/20

CPC classification number: H04M3/002 ,  H04B3/234 ,  H04B3/238 ,  H04M9/082

Abstract: A system and method for Acoustic Echo Cancellation. The system and method include a subband affine projection filter and a variable step size controller configured to cancel an estimated echo from a near-end signal. The system and method also include a divergence detector adapted to reset the subband affine projection filter in response to determining a divergence is occurring. Additionally, the system and method include a double talk detector adapted to transmit a signal to mask an output signal when double talk is detected.

Abstract translation: 声学回波消除的系统和方法。 该系统和方法包括子带仿射投影滤波器和可变步长控制器，其被配置为从近端信号中消除估计的回波。 所述系统和方法还包括发散检测器，其适于响应于确定发生发散而重新设置子带仿射投影滤波器。 此外，该系统和方法包括双重通话检测器，适用于在检测到双方通话时发送信号以屏蔽输出信号。

公开(公告)号：US08775168B2

公开(公告)日：2014-07-08

申请号：US11890268

申请日：2007-08-03

Applicant: Karthik Muralidhar ,  Anoop Kumar Krishna

Inventor： Karthik Muralidhar ,  Anoop Kumar Krishna

IPC: G10L25/00 ,  G10L21/00 ,  G10L21/02 ,  G10L15/00

CPC classification number: G10L25/87

Abstract: A Yule-Walker based, low-complexity voice activity detector (VAD) is disclosed. An input signal is typically noisy speech (i.e., corrupted with, for example, babble noise). In one embodiment, a first initialization stage of the VAD computes an occurrence of a silent period within the input signal and the AR parameters. The VAD could accordingly compute a tentative adaptive threshold and output hypothesis H1 (which means speech is present) during this stage. During the second initialization stage, the VAD generally builds a database of associated values and computes the adaptive threshold accordingly. The second initialization stage could also output tentative VAD decisions based on the tentative threshold computed in the first initialization stage. Finally, the VAD periodically retrains or updates AR parameters, threshold values and/or the database and outputs VAD decisions accordingly.

Abstract translation: 公开了一种基于Yule-Walker的低复杂度语音活动检测器（VAD）。 输入信号通常是有噪声的语音（例如，由于混音噪声而损坏）。 在一个实施例中，VAD的第一初始化阶段计算输入信号和AR参数内的无声期的出现。 在此阶段，VAD可以相应地计算出暂定自适应阈值和输出假设H1（这意味着语音存在）。 在第二初始化阶段期间，VAD通常构建相关值的数据库并相应地计算自适应阈值。 第二初始化阶段还可以基于在第一初始化阶段中计算的暂定阈值来输出暂定VAD决定。 最后，VAD定期重新训练或更新AR参数，阈值和/或数据库，并相应地输出VAD决定。

公开(公告)号：US08355334B2

公开(公告)日：2013-01-15

申请号：US12072817

申请日：2008-02-28

Applicant: Tanu Malhotra ,  Anoop Kumar Krishna

Inventor： Tanu Malhotra ,  Anoop Kumar Krishna

IPC: H04J1/16

CPC classification number: H04L65/1069 ,  H04L65/1006 ,  H04L65/1009 ,  H04L65/80 ,  H04L69/24 ,  H04M7/0072

Abstract: A method of improving voice quality in a packet based network. The method includes receiving an incoming call from a first endpoint and matching capabilities between the first endpoint and the second endpoint. The method also includes completing the incoming call if the capabilities match and tracking the packet loss associated with the network. The method also includes negotiating the voice quality based on the tracking and the capabilities. Also described is a devices and system for a similar method.

Abstract translation: 一种改善基于分组的网络中的语音质量的方法。 该方法包括从第一端点接收来话呼叫并在第一端点与第二端点之间匹配能力。 该方法还包括如果能力匹配并跟踪与网络相关联的分组丢失，则完成来话呼叫。 该方法还包括基于跟踪和能力来协商语音质量。 还描述了用于类似方法的装置和系统。

公开(公告)号：US20080219244A1

公开(公告)日：2008-09-11

申请号：US12072817

申请日：2008-02-28

Applicant: Tanu Malhotra ,  Anoop Kumar Krishna

Inventor： Tanu Malhotra ,  Anoop Kumar Krishna

IPC: H04L12/66

CPC classification number: H04L65/1069 ,  H04L65/1006 ,  H04L65/1009 ,  H04L65/80 ,  H04L69/24 ,  H04M7/0072

Abstract: A method of improving voice quality in a packet based network. The method includes receiving an incoming call from a first endpoint and matching capabilities between the first endpoint and the second endpoint. The method also includes completing the incoming call if the capabilities match and tracking the packet loss associated with the network. The method also includes negotiating the voice quality based on the tracking and the capabilities. Also described is a devices and system for a similar method.

Abstract translation: 一种改善基于分组的网络中的语音质量的方法。 该方法包括从第一端点接收来话呼叫并在第一端点与第二端点之间匹配能力。 该方法还包括如果能力匹配并跟踪与网络相关联的分组丢失，则完成来话呼叫。 该方法还包括基于跟踪和能力来协商语音质量。 还描述了用于类似方法的装置和系统。