公开(公告)号：US20050010406A1

公开(公告)日：2005-01-13

申请号：US10850105

申请日：2004-05-21

申请人： Ryosuke Koshiba ,  Akinori Kawamura

发明人： Ryosuke Koshiba ,  Akinori Kawamura

IPC分类号： G10L15/02 ,  G10L15/20 ,  G10L21/02 ,  G10L15/00

CPC分类号： G10L15/20 ,  G10L15/02 ,  G10L21/0208

摘要： A speech recognition apparatus, method and computer program product whereby noise is subtracted from an input speech signal by a plurality of spectral subtractions having differing rates of noise subtraction to produce plural noise-subtracted signals, at least one speech features is extracted from the noise-subtracted signals, and the extracted feature is compared with a standard speech pattern obtained beforehand to recognize the speech signal based on a result of the comparison. In addition, features can be extracted from at least one of the noise-subtracted signals and also the input speech signal for comparison with the standard speech pattern. Plural features can be combined into a single feature for the comparison.

摘要翻译： 一种语音识别装置，方法和计算机程序产品，其中通过具有不同噪声减法率的多个频谱减法从输入语音信号中减去噪声，以产生多个降噪信号，从噪声抑制中提取至少一个语音特征， 将所提取的特征与预先获得的标准语音模式进行比较，以基于比较结果识别语音信号。 此外，可以从噪声相减信号中的至少一个以及输入语音信号中提取特征，以与标准语音模式进行比较。 多个功能可以组合成一个单独的功能进行比较。

公开(公告)号：US07415408B2

公开(公告)日：2008-08-19

申请号：US11760912

申请日：2007-06-11

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L21/02 ,  G10L15/04

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of recognizing target vocabulary, i.e., a compared likelihood of registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit compares the time series of the amount of characteristics with one recognizing standard pattern or with two or more combined recognizing standard patterns one-by-one to obtain a likelihood that respective environment adaptive noise models coincide with the time series of the amount of characteristics. A rejection determining unit determining unit determines whether or not the input signal is a noise by comparing the likelihood obtained by the recognizing target vocabulary comparing step with the likelihood obtained by the environment adaptive noise model comparing step. A noise model adapting unit calculates a first likelihood by comparing the time series of the amount of characteristics with recognizing standard patterns and orders the recognizing standard patterns in accordance with the size of the first likelihood. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

摘要翻译： 识别目标词汇比较单元通过使用输入语音的特征量的时间序列来计算识别目标词汇的比较似然率，即登记词汇的比较似然率。 环境自适应噪声模型比较单元将特征量的时间序列与一个识别标准模式或两个或更多个组合的识别标准模式逐一比较，以获得各自的环境自适应噪声模型与时间序列一致的可能性 的特点。 拒绝判定单元确定单元通过将通过识别目标词汇比较步骤获得的可能性与通过环境自适应噪声模型比较步骤获得的可能性进行比较来确定输入信号是否为噪声。 噪声模型适应单元通过将特征量的时间序列与识别标准模式进行比较来计算第一可能性，并根据第一似然度的大小对识别标准模式进行排序。 因此，环境自适应噪声模型与实际环境匹配，并且可以以高精度对噪声输入执行拒绝确定。

公开(公告)号：US20070233476A1

公开(公告)日：2007-10-04

申请号：US11760912

申请日：2007-06-11

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L15/00

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of recognizing target vocabulary, i.e., a compared likelihood of registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit calculates a compared likelihood of a noise model adaptive to a noise environment, i.e., a compared likelihood of environmental noise. A rejection determining unit compares the likelihood of the registered vocabulary with the likelihood of the environmental noise, and determines whether or not the input speech is the noise. When it is determined that the input speech is the noise, a noise model adapting unit adaptively updates an environment adaptive noise model by using the input speech. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

摘要翻译： 识别目标词汇比较单元通过使用输入语音的特征量的时间序列来计算识别目标词汇的比较似然率，即登记词汇的比较似然率。 环境自适应噪声模型比较单元计算自适应噪声环境的噪声模型的比较似然率，即环境噪声的比较可能性。 拒绝判定单元将注册词汇的可能性与环境噪声的可能性进行比较，并且确定输入语音是否为噪声。 当确定输入语音是噪声时，噪声模型适应单元通过使用输入语音自适应地更新环境自适应噪声模型。 因此，环境自适应噪声模型与实际环境匹配，并且可以以高精度对噪声输入执行拒绝确定。

公开(公告)号：US07706550B2

公开(公告)日：2010-04-27

申请号：US11028317

申请日：2005-01-04

申请人： Tadashi Amada ,  Akinori Kawamura ,  Ryosuke Koshiba

发明人： Tadashi Amada ,  Akinori Kawamura ,  Ryosuke Koshiba

IPC分类号： H04B15/00

CPC分类号： G10L21/0208

摘要： A noise estimation unit estimates a noise signal in an input signal. A section decision unit distinguishes a target signal section from a noise signal section in the input signal. A noise suppression unit suppresses the noise signal based on a first suppression coefficient from the input signal. A noise excess suppression unit suppresses the noise signal based on a second suppression coefficient from the input signal. The second suppression coefficient is larger than the first suppression coefficient. A switching unit switches between an output signal from the noise suppression unit and an output signal from the noise excess suppression unit based on a decision result of the section decision unit.

摘要翻译： 噪声估计单元估计输入信号中的噪声信号。 区段决定单元将输入信号中的目标信号区间与噪声信号区间区分开。 噪声抑制单元根据来自输入信号的第一抑制系数来抑制噪声信号。 噪声过剩抑制单元根据来自输入信号的第二抑制系数来抑制噪声信号。 第二抑制系数大于第一抑制系数。 开关单元基于区段判定单元的判定结果，在来自噪声抑制单元的输出信号和来自噪声过剩抑制单元的输出信号之间切换。

公开(公告)号：US07409341B2

公开(公告)日：2008-08-05

申请号：US11760909

申请日：2007-06-11

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L21/02 ,  G10L15/04

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of recognizing target vocabulary, i.e., a compared likelihood of registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adapted noise model comparing unit compares the time series of the amount of characteristics with one recognizing standard pattern or with two or more combined recognizing standard patterns one-by-one to obtain a likelihood that respective environment adaptive noise models coincide with the time series of the amount of characteristics. A rejection determining unit determines whether or not the input signal is noise by comparing the likelihood obtained by the recognizing target vocabulary comparing step with the likelihood obtained by the environment adaptive noise model comparing step. A noise model adapting unit calculates a first likelihood by comparing the time series of the amount of characteristics with recognizing standard patterns and orders the recognizing standard patterns in accordance with the size of the first likelihood. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

摘要翻译： 识别目标词汇比较单元通过使用输入语音的特征量的时间序列来计算识别目标词汇的比较似然率，即登记词汇的比较似然率。 环境适应噪声模型比较单元将特征量的时间序列与一个识别标准模式或两个或更多个组合识别标准模式逐一比较，以获得各个环境自适应噪声模型与时间序列一致的可能性 的特点。 拒绝判定单元通过将通过识别目标词汇比较步骤获得的似然率与通过环境自适应噪声模型比较步骤获得的似然率进行比较来确定输入信号是否为噪声。 噪声模型适应单元通过将特征量的时间序列与识别标准模式进行比较来计算第一可能性，并根据第一似然度的大小对识别标准模式进行排序。 因此，环境自适应噪声模型与实际环境匹配，并且可以以高精度对噪声输入执行拒绝确定。

公开(公告)号：US08423360B2

公开(公告)日：2013-04-16

申请号：US10850105

申请日：2004-05-21

申请人： Ryosuke Koshiba ,  Akinori Kawamura

发明人： Ryosuke Koshiba ,  Akinori Kawamura

IPC分类号： G10L15/20

CPC分类号： G10L15/20 ,  G10L15/02 ,  G10L21/0208

摘要： A speech recognition apparatus, method and computer program product whereby noise is subtracted from an input speech signal by a plurality of spectral subtractions having differing rates of noise subtraction to produce plural noise-subtracted signals, at least one speech features is extracted from the noise-subtracted signals, and the extracted feature is compared with a standard speech pattern obtained beforehand to recognize the speech signal based on a result of the comparison. In addition, features can be extracted from at least one of the noise-subtracted signals and also the input speech signal for comparison with the standard speech pattern. Plural features can be combined into a single feature for the comparison.

摘要翻译： 一种语音识别装置，方法和计算机程序产品，其中通过具有不同噪声减法率的多个频谱减法从输入语音信号中减去噪声，以产生多个降噪信号，从噪声抑制中提取至少一个语音特征， 将所提取的特征与预先获得的标准语音模式进行比较，以基于比较结果识别语音信号。 此外，可以从噪声相减信号中的至少一个以及输入语音信号中提取特征，以与标准语音模式进行比较。 多个功能可以组合成一个单独的功能进行比较。

公开(公告)号：US07447634B2

公开(公告)日：2008-11-04

申请号：US11760902

申请日：2007-06-11

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L15/04 ,  G10L21/02

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of recognizing target vocabulary, i.e., a compared likelihood of registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit obtains a likelihood that respective recognizing-unit standard patterns coincide with a time series of the amount of characteristics representing the characteristics of the input speed. A first rejection unit determines whether or not the input signal is noise, on the basis of a likelihood of coincidence obtained by the recognizing target vocabulary comparing unit, and a second rejection determining unit determines whether or not the input signal determined to be noise by the first rejection determining unit is noise, on the basis of the likelihood of coincidence obtained by the recognizing target vocabulary comparing unit and the likelihood of coincidence obtained by the environment adaptive noise model comparing unit. An optimal phoneme series comparing unit determines a likelihood that respective recognizing-unit standard patterns coincide with the time series of the amount of characteristics. When the first or second rejection determining unit determines the input signal to be noise, the environment adaptive recognizing unit selecting unit adaptively updates an order of selection of the recognizing-unit standard pattern on the basis of the likelihood of coincidence obtained by the optimal phoneme series comparing unit. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

摘要翻译： 识别目标词汇比较单元通过使用输入语音的特征量的时间序列来计算识别目标词汇的比较似然率，即登记词汇的比较似然率。 环境自适应噪声模型比较单元获得各自的识别单位标准模式与表示输入速度的特性的特征量的时间序列一致的似然度。 第一拒绝单元基于由识别目标词汇比较单元获得的符合的可能性来确定输入信号是否为噪声，并且第二拒绝判定单元确定由所述识别目标词汇比较单元确定为噪声的输入信号是否为 基于由识别目标词汇比较单元获得的符合的可能性和由环境自适应噪声模型比较单元获得的符合的可能性，第一拒绝判定单元是噪声。 最佳音素序列比较单元确定各识别单位标准模式与特征量的时间序列一致的可能性。 当第一或第二拒绝判定单元确定输入信号为噪声时，环境自适应识别单元选择单元根据由最佳音素系列获得的符合概率自适应地更新识别单元标准模式的选择顺序 比较单位 因此，环境自适应噪声模型与实际环境匹配，并且可以以高精度对噪声输入执行拒绝确定。

公开(公告)号：US20070233475A1

公开(公告)日：2007-10-04

申请号：US11760902

申请日：2007-06-11

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L15/00

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of recognizing target vocabulary, i.e., a compared likelihood of registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit calculates a compared likelihood of a noise model adaptive to a noise environment, i.e., a compared likelihood of environmental noise. A rejection determining unit compares the likelihood of the registered vocabulary with the likelihood of the environmental noise, and determines whether or not the input speech is the noise. When it is determined that the input speech is the noise, a noise model adapting unit adaptively updates an environment adaptive noise model by using the input speech. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

公开(公告)号：US07260527B2

公开(公告)日：2007-08-21

申请号：US10329553

申请日：2002-12-27

申请人： Ryosuke Koshiba

发明人： Ryosuke Koshiba

IPC分类号： G10L15/20

CPC分类号： G10L15/20

摘要： A recognizing target vocabulary comparing unit calculates a compared likelihood of a recognizing target vocabulary, i.e., a compared likelihood of a registered vocabulary, by using the time series of the amount of characteristics of an input speech. An environment adaptive noise model comparing unit calculates a compared likelihood of a noise model adaptive to a noise environment, i.e., a compared likelihood of environmental noise. A rejection determining unit compares the likelihood of the registered vocabulary with the likelihood of the environmental noise, and determines whether or not the input speech is the noise. When it is determined that the input speech is the noise, a noise model adapting unit adaptively updates an environment adaptive noise model by using the input speech. Thus, the environment adaptive noise model matches to a real environment and the rejection determination can be performed for a noise input with high accuracy.

摘要翻译： 识别目标词汇比较单元通过使用输入语音的特征量的时间序列来计算识别目标词汇的比较似然度，即登记词汇的比较似然度。 环境自适应噪声模型比较单元计算自适应噪声环境的噪声模型的比较似然率，即环境噪声的比较可能性。 拒绝判定单元将注册词汇的可能性与环境噪声的可能性进行比较，并且确定输入语音是否为噪声。 当确定输入语音是噪声时，噪声模型适应单元通过使用输入语音自适应地更新环境自适应噪声模型。 因此，环境自适应噪声模型与实际环境匹配，并且可以以高精度对噪声输入执行拒绝确定。

公开(公告)号：US20050152563A1

公开(公告)日：2005-07-14

申请号：US11028317

申请日：2005-01-04

申请人： Tadashi Amada ,  Akinori Kawamura ,  Ryosuke Koshiba

发明人： Tadashi Amada ,  Akinori Kawamura ,  Ryosuke Koshiba

IPC分类号： G10L15/20 ,  G10L11/02 ,  G10L15/04 ,  G10L15/28 ,  G10L19/00 ,  G10L21/02 ,  H04B15/00 ,  H04M1/60 ,  H04R3/00

CPC分类号： G10L21/0208

摘要： A noise estimation unit estimates a noise signal in an input signal. A section decision unit distinguishes a target signal section from a noise signal section in the input signal. A noise suppression unit suppresses the noise signal based on a first suppression coefficient from the input signal. A noise excess suppression unit suppresses the noise signal based on a second suppression coefficient from the input signal. The second suppression coefficient is larger than the first suppression coefficient. A switching unit switches between an output signal from the noise suppression unit and an output signal from the noise excess suppression unit based on a decision result of the section decision unit.

摘要翻译： 噪声估计单元估计输入信号中的噪声信号。 区段决定单元将输入信号中的目标信号区间与噪声信号区间区分开。 噪声抑制单元根据来自输入信号的第一抑制系数来抑制噪声信号。 噪声过剩抑制单元根据来自输入信号的第二抑制系数抑制噪声信号。 第二抑制系数大于第一抑制系数。 开关单元基于区段判定单元的判定结果，在来自噪声抑制单元的输出信号和来自噪声过剩抑制单元的输出信号之间切换。