公开(公告)号：US06671667B1

公开(公告)日：2003-12-30

申请号：US09536583

申请日：2000-03-28

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L1902

CPC分类号： G10L25/78 ,  G10L21/0208 ,  G10L21/0232 ,  G10L21/0264 ,  G10L2025/783

摘要： In order to enhance the quality of a communication signal derived from speech and noise, the likelihood that the communication signals result from at least some speech is determined. A calculator calculates a first power signal representing the power of at least a portion of the communication signals estimated over a first time period and calculates a second power signal representing the power of at least a portion of the communication signals estimated over a second time period longer than the first time period. The calculator also generates a comparison signal having a value related to the likelihood that the portion of the communication signals result from at least some speech by comparing a first expression involving the first power signal with a second expression involving the second power signal. The calculator also generates a speech likelihood signal having a value representing a first likelihood that the communication signal results from at least some speech in the event that the comparison signal value falls within a first range and having a second value representing a second likelihood that the communication signal results from at least some speech in the event that the comparison signal value falls within a second range. The second likelihood is different from the first likelihood.

摘要翻译： 为了提高从语音和噪声导出的通信信号的质量，确定了至少一些语音产生通信信号的可能性。 计算器计算表示在第一时间周期估计的通信信号的至少一部分的功率的第一功率信号，并且计算表示在第二时间段内估计的至少一部分通信信号的功率的第二功率信号 比第一个时期。 计算器还生成比较信号，其具有与通过将包含第一功率信号的第一表达与涉及第二功率信号的第二表达进行比较，从至少一些语音产生通信信号的部分的可能性的值。 计算器还产生语音似然信号，该语音似然信号具有表示在比较信号值落在第一范围内并且具有表示第二可能性的第二值的通信信号从至少一些语音产生的第一可能性的值， 在比较信号值落在第二范围的情况下，来自至少一些语音的信号。 第二种可能性与第一种可能性不同。

公开(公告)号：US06839666B2

公开(公告)日：2005-01-04

申请号：US10316776

申请日：2002-12-11

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L11/02 ,  G10L21/02 ,  G10L19/02

CPC分类号： G10L25/78 ,  G10L21/0208 ,  G10L21/0232 ,  G10L21/0264 ,  G10L2025/783

摘要： The spectral shape of a communication signal is preserved by filtering it into a selected number of frequency band signals representing a selected number of the frequency bands. A calculator generates a plurality of initial gain signals having initial gain values for altering the gain of the frequency band signals. Each initial gain signal corresponds to one of the frequency band signals. Each initial gain value is derived from a measurement of the power of at least a portion of one of the frequency band signals. The calculator also generates a plurality of modified gain signals having modified gain values. Each modified gain signal corresponds to at least one of the frequency band signals and each modified gain value is derived from one or more functions of at least two of the initial gain values. The frequency band signals are altered in response to the modified gain signals to generate weighted frequency band signals which are combined to generate an improved communication signal.

摘要翻译： 通过将通信信号的频谱形状滤波成表示选定数量的频带的选定数量的频带信号来保存通信信号的频谱形状。 计算器产生具有用于改变频带信号的增益的初始增益值的多个初始增益信号。 每个初始增益信号对应于一个频带信号。 每个初始增益值是从一个频带信号的至少一部分的功率的测量得出的。 计算器还产生具有修改的增益值的多个经修改的增益信号。 每个修改的增益信号对应于至少一个频带信号，并且每个修改的增益值从至少两个初始增益值的一个或多个函数导出。 响应于修改的增益信号改变频带信号以产生加权的频带信号，其被组合以产生改进的通信信号。

公开(公告)号：US06529868B1

公开(公告)日：2003-03-04

申请号：US09536941

申请日：2000-03-28

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L2102

CPC分类号： G10L21/0208 ,  G10L19/0204 ,  G10L21/02 ,  G10L21/0232

摘要： In order to enhance the quality of a communication signal derived from speech and noise, a filter divides the communication signal into a plurality of frequency band signals. A calculator generates a plurality of power band signals each having a power band value and corresponding to one of the frequency band signals. The power band values are based on estimating, over a time period, the power of one of the frequency band signals. The time period is different for different ones of the frequency band signals. The power band values are used to calculate weighting factors which are used to alter the frequency band signals that are combined to generate an improved communication signal.

摘要翻译： 为了提高从语音和噪声导出的通信信号的质量，滤波器将通信信号分成多个频带信号。 计算器产生多个功率带信号，每个功率带信号具有功率带值并对应于一个频带信号。 功率带值基于在一段时间内估计一个频带信号的功率。 对于不同的频带信号，时间周期是不同的。 功率带值用于计算用于改变被组合以产生改进的通信信号的频带信号的加权因子。

公开(公告)号：US07424424B2

公开(公告)日：2008-09-09

申请号：US11476309

申请日：2006-06-28

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L21/02

CPC分类号： G10L21/0208 ,  G10L19/0204 ,  G10L21/02 ,  G10L21/0232

摘要： In order to enhance the quality of a communication signal derived from speech and noise, a filter divides the communication signal into a plurality of frequency band signals. A calculator generates a plurality of power band signals each having a power band value and corresponding to one of the frequency band signals. The power band values are based on estimating, over a time period, the power of one of the frequency band signals. The time period is different for different ones of the frequency band signals. The power band values are used to calculate weighting factors which are used to alter the frequency band signals that are combined to generate an improved communication signal.

摘要翻译： 为了提高从语音和噪声导出的通信信号的质量，滤波器将通信信号分成多个频带信号。 计算器产生多个功率带信号，每个功率带信号具有功率带值并对应于一个频带信号。 功率带值基于在一段时间内估计一个频带信号的功率。 对于不同的频带信号，时间周期是不同的。 功率带值用于计算用于改变被组合以产生改进的通信信号的频带信号的加权因子。

公开(公告)号：US06591234B1

公开(公告)日：2003-07-08

申请号：US09479120

申请日：2000-01-07

申请人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

发明人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

IPC分类号： G10L2102

CPC分类号： G10L21/0208 ,  G10L21/0232 ,  G10L25/18

摘要： A processor (300) is arranged to divide a communication signal into a plurality of frequency band signals including speech and noise components due to speech and noise. The processor generates first and second power signals for the frequency band signals. Each first power signal is based on estimating over a first time period the power of one of the frequency band signals. Each second power signal is based on estimating over a second time period less than the first time period the power of one of the frequency band signals. The processor generates condition signals representing conditions of the frequency band signals, and adjusts the gain of the frequency band signals in response to the condition signals to generate adjusted frequency band signals. The processor then combines the adjusted frequency band signals to generate an adjusted communication signal.

摘要翻译： 处理器（300）被布置为由于语音和噪声将通信信号划分成包括语音和噪声分量的多个频带信号。 处理器产生用于频带信号的第一和第二功率信号。 每个第一功率信号基于在第一时间段内估计一个频带信号的功率。 每个第二功率信号基于在第一时间周期内估计超过一个频带信号的功率的第二时间段。 处理器产生表示频带信号的条件的条件信号，并且响应于条件信号调整频带信号的增益以产生经调整的频带信号。 然后处理器组合经调整的频带信号以产生经调整的通信信号。

公开(公告)号：US07366294B2

公开(公告)日：2008-04-29

申请号：US11046161

申请日：2005-01-28

申请人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

发明人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

IPC分类号： H04M1/00 ,  H04M3/00 ,  H04M7/00

CPC分类号： G10L21/0208 ,  G10L21/0232 ,  G10L25/18

摘要： An apparatus and method for suppressing noise is presented. The apparatus may utilize a filter bank of bandpass filters to split the input noisy speech-containing signal into separate frequency bands. To determine whether the input signal contains speech, DTMF tones or silence, a joint voice activity & DTMF activity detector (JVADAD) may be used. The overall average noise-to-signal ratio (NSR) of the input signal is estimated in the overall NSR estimator, which estimates the average noisy signal power in the input signal during speech activity and the average noise power during silence. Two indirect power measures are performed for each band, measuring a short-term power and a long-term power. The power estimation processes are adapted based on the signal activity indicated by the JVADAD. A NSR adapter adapts the NSR for each frequency band based on the long-term and short-term power measures, the overall NSR and the signal activity indicated by the JVADAD. The NSR adaptation may then be performed. The gain computer utilizes these NSR values to determine the gain factors for each frequency band. The gain multiplier may then perform the attenuation of each frequency band. Finally, the processed signals in the separate frequency bands are summed up in the combiner to produce the clean output signal. In another embodiment of the present invention, a method for suppressing noise is presented. An alternative embodiment of the present invention includes a method and apparatus for extending DTMF tones. Yet another embodiment of the present invention includes regenerating DTMF tones.

摘要翻译： 提出了一种抑制噪声的装置和方法。 该装置可以利用滤波器​​组的带通滤波器将输入的含噪声语音的信号分离成单独的频带。 为了确定输入信号是否包含语音，DTMF音或静音，可以使用联合语音活动和DTMF活动检测器（JVADAD）。 在整个NSR估计器中估计输入信号的总体平均噪声信噪比（NSR），NSR估计器估计语音活动期间输入信号中的平均噪声信号功率和静音期间的平均噪声功率。 对每个频带执行两个间接功率测量，测量短期功率和长期功率。 功率估计过程基于JVADAD指示的信号活动进行调整。 NSR适配器根据长期和短期功率测量，整个NSR和JVADAD指示的信号活动，为每个频段调整NSR。 然后可以执行NSR适配。 增益计算机利用这些NSR值来确定每个频带的增益因子。 然后，增益乘法器可以执行每个频带的衰减。 最后，在组合器中将分离的频带中的经处理的信号相加，以产生干净的输出信号。 在本发明的另一实施例中，提出了一种抑制噪声的方法。 本发明的替代实施例包括用于扩展DTMF音调的方法和装置。 本发明的另一个实施例包括再生DTMF音调。

公开(公告)号：US07158572B2

公开(公告)日：2007-01-02

申请号：US10075145

申请日：2002-02-14

申请人： Bruce E. Dunne ,  Ravi Chandran ,  Daniel J. Marchok

发明人： Bruce E. Dunne ,  Ravi Chandran ,  Daniel J. Marchok

IPC分类号： H04B1/66

CPC分类号： G10L19/18 ,  G10L21/0364 ,  H04B1/66

摘要： A communication system (10) receives a communication signal comprising first and second data with different compression levels, such as highly compressed and weakly compressed levels. A mode detector (15) detects the level of compression. One or more signal decoders (20, 22) decode the highly compressed data. An analyzer (30) determines the type of enhancement required. One or more processors (48, 50, 80) enhance the data as required. An encoder (60) reencodes the enhanced decoded data. Metrics (90) may aid the operation of the analyzer (30).The communication system may include telephones (120, 122, 124, 126). Processors (103, 104) enhance signals in opposite first and second directions between pairs of the telephones. A path (106) connects the processors in tandem. One or more switches (101, 102) disable signal enhancement for one of the processors depending on the compression level of the signals to avoid degrading call quality.

公开(公告)号：US07096182B2

公开(公告)日：2006-08-22

申请号：US10376849

申请日：2003-02-28

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L21/02

CPC分类号： G10L21/0208 ,  G10L19/0204 ,  G10L21/02 ,  G10L21/0232

摘要： In order to enhance the quality of a communication signal derived from speech and noise, a filter divides the communication signal into a plurality of frequency band signals. A calculator generates a plurality of power band signals each having a power band value and corresponding to one of the frequency band signals. The power band values are based on estimating, over a time period, the power of one of the frequency band signals. The time period is different for different ones of the frequency band signals. The power band values are used to calculate weighting factors which are used to alter the frequency band signals that are combined to generate an improved communication signal.

公开(公告)号：US06766292B1

公开(公告)日：2004-07-20

申请号：US09535844

申请日：2000-03-28

申请人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

发明人： Ravi Chandran ,  Bruce E. Dunne ,  Daniel J. Marchok

IPC分类号： G10L2102

CPC分类号： G10L21/0208

摘要： In order to enhance the quality of a communication signal comprising speech signal components due to speech and noise signal components due to noise, a filter divides the communication signal into a plurality of frequency band signals representing the speech signal components and the noise signal components in a plurality of frequency bands. A calculator generates a plurality of weighting signals having weighting values corresponding to the frequency band signals. The weighting values represent at least approximations of the normalized powers of the noise signal components in the frequency band signals. The frequency band signals are altered in response to the weighting signals to generate weighted frequency band signals which are combined to generate a communication signal with enhanced quality.

摘要翻译： 为了提高包含由噪声引起的语音和噪声信号分量的语音信号分量的通信信号的质量，滤波器将通信信号分成多个频带信号，该信号表示语音信号分量和噪声信号分量 多个频带。 计算器产生具有对应于频带信号的加权值的多个加权信号。 加权值表示频带信号中的噪声信号分量的归一化功率的近似值。 响应于加权信号改变频带信号以产生加权的频带信号，其被组合以产生具有增强质量的通信信号。

公开(公告)号：US08031861B2

公开(公告)日：2011-10-04

申请号：US12072500

申请日：2008-02-26

申请人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

发明人： Ravi Chandran ,  Daniel J. Marchok ,  Bruce E. Dunne

IPC分类号： H04M7/00 ,  H04M1/00 ,  H04M5/00

CPC分类号： G10L21/0208 ,  G10L21/0232 ,  G10L25/18

摘要： An apparatus and method for suppressing noise is presented. The apparatus may utilize a filter bank of bandpass filters to split the input noisy speech-containing signal into separate frequency bands. To determine whether the input signal contains speech, DTMF tones or silence, a joint voice activity & DTMF activity detector (JVADAD) may be used. The overall average noise-to-signal ratio (NSR) of the input signal is estimated in the overall NSR estimator, which estimates the average noisy signal power in the input signal during speech activity and the average noise power during silence. Two indirect power measures are performed for each band, measuring a short-term power and a long-term power. The power estimation processes are adapted based on the signal activity indicated by the JVADAD. A NSR adapter adapts the NSR for each frequency band based on the long-term and short-term power measures, the overall NSR and the signal activity indicated by the JVADAD. The NSR adaptation may then be performed. The gain computer utilizes these NSR values to determine the gain factors for each frequency band. The gain multiplier may then perform the attenuation of each frequency band. Finally, the processed signals in the separate frequency bands are summed up in the combiner to produce the clean output signal. In another embodiment of the present invention, a method for suppressing noise is presented. An alternative embodiment of the present invention includes a method and apparatus for extending DTMF tones. Yet another embodiment of the present invention includes regenerating DTMF tones.

摘要翻译： 提出了一种抑制噪声的装置和方法。 该装置可以利用滤波器​​组的带通滤波器将输入的含噪声语音的信号分离成单独的频带。 为了确定输入信号是否包含语音，DTMF音或静音，可以使用联合语音活动和DTMF活动检测器（JVADAD）。 在整个NSR估计器中估计输入信号的总体平均噪声信噪比（NSR），NSR估计器估计语音活动期间输入信号中的平均噪声信号功率和静音期间的平均噪声功率。 对每个频带执行两个间接功率测量，测量短期功率和长期功率。 功率估计过程基于JVADAD指示的信号活动进行调整。 NSR适配器根据长期和短期功率测量，整个NSR和JVADAD指示的信号活动，为每个频段调整NSR。 然后可以执行NSR适配。 增益计算机利用这些NSR值来确定每个频带的增益因子。 然后，增益乘法器可以执行每个频带的衰减。 最后，在组合器中将分离的频带中的经处理的信号相加，以产生干净的输出信号。 在本发明的另一实施例中，提出了一种抑制噪声的方法。 本发明的替代实施例包括用于扩展DTMF音调的方法和装置。 本发明的另一个实施例包括再生DTMF音调。