专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20090119096A1 PARTIAL SPEECH RECONSTRUCTION 有权
标题翻译：部分语音重建
公开(公告)号：US20090119096A1
公开(公告)日：2009-05-07
申请号：US12254488
申请日：2008-10-20
申请人： Franz Gerl , Tobias Herbig , Mohamed Krini , Gerhard Uwe Schmidt
发明人： Franz Gerl , Tobias Herbig , Mohamed Krini , Gerhard Uwe Schmidt
IPC分类号： G10L11/04 , G10L21/02 , G06F17/00
CPC分类号： H04R3/005 , G10L21/0208 , G10L21/0264 , G10L2021/02165 , H04R2410/05 , H04R2410/07 , H04R2499/11 , H04R2499/13
摘要： A system enhances the quality of a digital speech signal that may include noise. The system identifies vocal expressions that correspond to the digital speech signal. A signal-to-noise ratio of the digital speech signal is measured before a portion of the digital speech signal is synthesized. The selected portion of the digital speech signal may have a signal-to-noise ratio below a predetermined level and the synthesis of the digital speech signal may be based on speaker identification.
摘要翻译：系统提高可能包括噪声的数字语音信号的质量。该系统识别对应于数字语音信号的声乐表达。在数字语音信号的一部分被合成之前测量数字语音信号的信噪比。数字语音信号的所选部分可以具有低于预定电平的信噪比，并且数字语音信号的合成可以基于说话者识别。

2. 发明授权

US08706483B2 Partial speech reconstruction 有权
标题翻译：部分语言重建
公开(公告)号：US08706483B2
公开(公告)日：2014-04-22
申请号：US12254488
申请日：2008-10-20
申请人： Franz Gerl , Tobias Herbig , Mohamed Krini , Gerhard Uwe Schmidt
发明人： Franz Gerl , Tobias Herbig , Mohamed Krini , Gerhard Uwe Schmidt
IPC分类号： G10L21/00
CPC分类号： H04R3/005 , G10L21/0208 , G10L21/0264 , G10L2021/02165 , H04R2410/05 , H04R2410/07 , H04R2499/11 , H04R2499/13
摘要： A system enhances the quality of a digital speech signal that may include noise. The system identifies vocal expressions that correspond to the digital speech signal. A signal-to-noise ratio of the digital speech signal is measured before a portion of the digital speech signal is synthesized. The selected portion of the digital speech signal may have a signal-to-noise ratio below a predetermined level and the synthesis of the digital speech signal may be based on speaker identification.
摘要翻译：系统提高可能包括噪声的数字语音信号的质量。该系统识别对应于数字语音信号的声乐表达。在数字语音信号的一部分被合成之前测量数字语音信号的信噪比。数字语音信号的所选部分可以具有低于预定电平的信噪比，并且数字语音信号的合成可以基于说话者识别。

3. 发明授权

US09026438B2 Detecting barge-in in a speech dialogue system 有权
标题翻译：在语音对话系统中检测插入
公开(公告)号：US09026438B2
公开(公告)日：2015-05-05
申请号：US12415927
申请日：2009-03-31
申请人： Markus Buck , Franz Gerl , Tim Haulick , Tobias Herbig , Gerhard Uwe Schmidt , Matthias Schulz
发明人： Markus Buck , Franz Gerl , Tim Haulick , Tobias Herbig , Gerhard Uwe Schmidt , Matthias Schulz
IPC分类号： G10L15/00 , G10L25/78 , G10L15/22 , G10L21/00 , G10L13/00 , H04M1/64 , H04M3/42 , G06F11/00 , G10L17/00
CPC分类号： G10L25/78 , G10L15/222 , G10L17/00 , G10L2025/786
摘要： A method for detecting barge-in in a speech dialog system comprising determining whether a speech prompt is output by the speech dialog system, and detecting whether speech activity is present in an input signal based on a time-varying sensitivity threshold of a speech activity detector and/or based on speaker information, where the sensitivity threshold is increased if output of a speech prompt is determined and decreased if no output of a speech prompt is determined. If speech activity is detected in the input signal, the speech prompt may be interrupted or faded out. A speech dialog system configured to detect barge-in is also disclosed.
摘要翻译：一种用于检测语音对话系统中的插入的方法，包括：确定语音对话系统是否输出语音提示，以及基于语音活动检测器的时变灵敏度阈值检测语音活动是否存在于输入信号中和/或基于说话者信息，其中如果确定了语音提示的输出，则在不确定语音提示的输出的情况下确定并减小了灵敏度阈值。如果在输入信号中检测到语音活动，语音提示可能被中断或淡出。还公开了一种配置成检测插入的语音对话系统。

4. 发明申请

US20090254342A1 DETECTING BARGE-IN IN A SPEECH DIALOGUE SYSTEM 有权
标题翻译：检测语音对话系统中的边界
公开(公告)号：US20090254342A1
公开(公告)日：2009-10-08
申请号：US12415927
申请日：2009-03-31
申请人： Markus Buck , Franz Gerl , Tim Haulick , Tobias Herbig , Gerhard Uwe Schmidt , Matthias Schulz
发明人： Markus Buck , Franz Gerl , Tim Haulick , Tobias Herbig , Gerhard Uwe Schmidt , Matthias Schulz
IPC分类号： G10L17/00 , G10L15/20
CPC分类号： G10L25/78 , G10L15/222 , G10L17/00 , G10L2025/786
摘要： A method for detecting barge-in in a speech dialogue system comprising determining whether a speech prompt is output by the speech dialogue system, and detecting whether speech activity is present in an input signal based on a time-varying sensitivity threshold of a speech activity detector and/or based on speaker information, where the sensitivity threshold is increased if output of a speech prompt is determined and decreased if no output of a speech prompt is determined. If speech activity is detected in the input signal, the speech prompt may be interrupted or faded out. A speech dialogue system configured to detect barge-in is also disclosed.
摘要翻译：一种用于检测语音对话系统中的插入的方法，包括确定语音对话系统是否输出语音提示，以及基于语音活动检测器的时变灵敏度阈值来检测语音活动是否存在于输入信号中和/或基于说话者信息，其中如果确定了语音提示的输出，则在不确定语音提示的输出的情况下确定并减小了灵敏度阈值。如果在输入信号中检测到语音活动，语音提示可能被中断或淡出。还公开了一种配置成检测插入的语音对话系统。

5. 发明申请

US20100138222A1 Method for Adapting a Codebook for Speech Recognition 有权
标题翻译：适应语音识别码本的方法
公开(公告)号：US20100138222A1
公开(公告)日：2010-06-03
申请号：US12622717
申请日：2009-11-20
申请人： Tobias Herbig , Franz Gerl
发明人： Tobias Herbig , Franz Gerl
IPC分类号： G10L15/06
CPC分类号： G10L15/065
摘要： A method for adapting a codebook for speech recognition, wherein the codebook is from a set of codebooks comprising a speaker-independent codebook and at least one speaker-dependent codebook is disclosed. A speech input is received and a feature vector based on the received speech input is determined. For each of the Gaussian densities, a first mean vector is estimated using an expectation process and taking into account the determined feature vector. For each of the Gaussian densities, a second mean vector using an Eigenvoice adaptation is determined taking into account the determined feature vector. For each of the Gaussian densities, the mean vector is set to a convex combination of the first and the second mean vector. Thus, this process allows for adaptation during operation and does not require a lengthy training phase.
摘要翻译：一种用于适应用于语音识别的码本的方法，其中所述码本来自包括与扬声器无关的码本和至少一个与扬声器相关的码本的码本集合。接收到语音输入，并且确定基于所接收的语音输入的特征向量。对于每个高斯密度，使用期望过程并且考虑确定的特征向量来估计第一平均向量。对于每个高斯密度，使用特征语音适配的第二平均向量被确定，其考虑所确定的特征向量。对于每个高斯密度，将平均矢量设置为第一和第二平均矢量的凸组合。因此，该过程允许在操作期间的适应并且不需要冗长的训练阶段。

6. 发明授权

US09530432B2 Method for determining the presence of a wanted signal component 有权
标题翻译：用于确定有用信号分量的存在的方法
公开(公告)号：US09530432B2
公开(公告)日：2016-12-27
申请号：US12507444
申请日：2009-07-22
申请人： Tobias Herbig , Franz Gerl
发明人： Tobias Herbig , Franz Gerl
IPC分类号： G10L25/78 , G10L15/22
CPC分类号： G10L25/78 , G10L15/222 , G10L2015/223
摘要： This invention provides a method for determining, in a speech dialog system issuing speech prompts, a score value as an indicator for the presence of a wanted signal component in an input signal stemming from a microphone, comprising the steps of: using a first likelihood function to determine a first likelihood value for the presence of the wanted signal component in the input signal, using a second likelihood function to determine a second likelihood value for the presence of a noise signal component in the input signal, and determining a score value based on the first and the second likelihood values, wherein the first likelihood function is based on a predetermined reference wanted signal, and the second likelihood function is based on a predetermined reference noise signal.
摘要翻译：本发明提供了一种在发出语音提示的语音对话系统中确定得分值作为来自麦克风的输入信号中有用信号分量的存在的指标的方法，包括以下步骤：使用第一似然函数使用第二似然函数来确定用于在输入信号中存在噪声信号分量的第二似然值，并且基于该输入信号确定分数值来确定用于输入信号中有用信号分量的存在的第一似然值第一似然值和第二似然值，其中第一似然函数基于预定的参考有用信号，第二似然函数基于预定的参考噪声信号。

7. 发明授权

US08346551B2 Method for adapting a codebook for speech recognition 有权
标题翻译：适用于语音识别的码本的方法
公开(公告)号：US08346551B2
公开(公告)日：2013-01-01
申请号：US12622717
申请日：2009-11-20
申请人： Tobias Herbig , Franz Gerl
发明人： Tobias Herbig , Franz Gerl
IPC分类号： G10L15/06
CPC分类号： G10L15/065
摘要： A method for adapting a codebook for speech recognition, wherein the codebook is from a set of codebooks comprising a speaker-independent codebook and at least one speaker dependent codebook. A speech input is received and a feature vector based on the received speech input is determined. For each of the Gaussian densities, a first mean vector is estimated using an expectation process and taking into account the determined feature vector. For each of the Gaussian densities, a second mean vector using an Eigenvoice adaptation is determined taking into account the determined feature vector. For each of the Gaussian densities, the mean vector is set to a convex combination of the first and the second mean vector. Thus, this process allows for adaptation during operation and does not require a lengthy training phase.
摘要翻译：一种用于调整用于语音识别的码本的方法，其中码本来自包括与扬声器无关的码本和至少一个扬声器相关码本的一组码本。接收到语音输入，并且确定基于所接收的语音输入的特征向量。对于每个高斯密度，使用期望过程并且考虑确定的特征向量来估计第一平均向量。对于每个高斯密度，使用特征语音适配的第二平均向量被确定，其考虑所确定的特征向量。对于每个高斯密度，将平均矢量设置为第一和第二平均矢量的凸组合。因此，该过程允许在操作期间的适应并且不需要冗长的训练阶段。

8. 发明申请

US20100030558A1 Method for Determining the Presence of a Wanted Signal Component 有权
标题翻译：确定需要信号分量的方法
公开(公告)号：US20100030558A1
公开(公告)日：2010-02-04
申请号：US12507444
申请日：2009-07-22
申请人： Tobias Herbig , Franz Gerl
发明人： Tobias Herbig , Franz Gerl
IPC分类号： G10L11/00
CPC分类号： G10L25/78 , G10L15/222 , G10L2015/223
摘要： This invention provides a method for determining, in a speech dialogue system issuing speech prompts, a score value as an indicator for the presence of a wanted signal component in an input signal stemming from a microphone, comprising the steps of: using a first likelihood function to determine a first likelihood value for the presence of the wanted signal component in the input signal, using a second likelihood function to determine a second likelihood value for the presence of a noise signal component in the input signal, and determining a score value based on the first and the second likelihood values, wherein the first likelihood function is based on a predetermined reference wanted signal, and the second likelihood function is based on a predetermined reference noise signal.
摘要翻译：本发明提供了一种在发出语音提示的语音对话系统中确定得分值作为来自麦克风的输入信号中有用信号分量的存在的指标的方法，包括以下步骤：使用第一似然函数使用第二似然函数来确定用于在输入信号中存在噪声信号分量的第二似然值，并且基于该输入信号确定分数值来确定用于输入信号中有用信号分量的存在的第一似然值第一似然值和第二似然值，其中第一似然函数基于预定的参考有用信号，第二似然函数基于预定的参考噪声信号。

9. 发明申请

US20090119103A1 SPEAKER RECOGNITION SYSTEM 审中-公开
标题翻译：扬声器识别系统
公开(公告)号：US20090119103A1
公开(公告)日：2009-05-07
申请号：US12249089
申请日：2008-10-10
申请人： Franz Gerl , Tobias Herbig
发明人： Franz Gerl , Tobias Herbig
IPC分类号： G10L15/06 , G10L17/00
CPC分类号： G10L17/04
摘要： A method automatically recognizes speech received through an input. The method accesses one or more speaker-independent speaker models. The method detects whether the received speech input matches a speaker model according to an adaptable predetermined criterion. The method creates a speaker model assigned to a speaker model set when no match occurs based on the input.
摘要翻译：一种方法自动识别通过输入接收的语音。该方法访问一个或多个与扬声器无关的扬声器模型。该方法根据可适应的预定标准检测接收到的语音输入是否与扬声器模型相匹配。该方法基于输入，在不匹配发生时创建分配给扬声器模型集的扬声器模型。

10. 发明授权

US08131544B2 System for distinguishing desired audio signals from noise 有权
标题翻译：用于区分所需音频信号和噪声的系统
公开(公告)号：US08131544B2
公开(公告)日：2012-03-06
申请号：US12269837
申请日：2008-11-12
申请人： Tobias Herbig , Oliver Gaupp , Franz Gerl
发明人： Tobias Herbig , Oliver Gaupp , Franz Gerl
IPC分类号： G10L15/20
CPC分类号： G10L25/78 , G10L25/24 , G10L2021/02166
摘要： A system distinguishes a primary audio source and background noise to improve the quality of an audio signal. A speech signal from a microphone may be improved by identifying and dampening background noise to enhance speech. Stochastic models may be used to model speech and to model background noise. The models may determine which portions of the signal are speech and which portions are noise. The distinction may be used to improve the signal's quality, and for speaker identification or verification.
摘要翻译：系统区分主音频源和背景噪声，以提高音频信号的质量。来自麦克风的语音信号可以通过识别和衰减背景噪声来增强语音来提高。随机模型可用于对语音进行建模并对背景噪声进行建模。模型可以确定信号的哪些部分是语音，哪些部分是噪声。该区别可用于提高信号的质量，并用于说话者识别或验证。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式