公开(公告)号：EP2867890B1

公开(公告)日：2018-04-25

申请号：EP12735984.2

申请日：2012-06-28

申请人： Nuance Communications, Inc.

发明人： WILLETT, Daniel ,  LÖÖF, Karl, Jonas ,  PAN, Yue ,  PINTO, Joel ,  GOLLAN, Christian

IPC分类号： G10L25/30 ,  G10L15/20 ,  G10L15/02 ,  G10L15/14

CPC分类号： G10L15/14 ,  G10L15/02 ,  G10L15/20 ,  G10L25/30

摘要： A computer-implemented method is described for front end speech processing for automatic speech recognition. A sequence of speech features which characterize an unknown speech input provided on an audio input channel and associated meta-data which characterize the audio input channel are received. The speech features are transformed with a computer process that uses a trained mapping function controlled by the meta-data, and automatic speech recognition is performed of the transformed speech features.

公开(公告)号：EP2867890A1

公开(公告)日：2015-05-06

申请号：EP12735984.2

申请日：2012-06-28

申请人： Nuance Communications, Inc.

发明人： WILLETT, Daniel ,  LÖÖF, Karl, Jonas ,  PAN, Yue ,  PINTO, Joel ,  GOLLAN, Christian

IPC分类号： G10L15/20 ,  G10L15/24 ,  G06F17/30

CPC分类号： G10L15/14 ,  G10L15/02 ,  G10L15/20 ,  G10L25/30

摘要： A computer-implemented method is described for front end speech processing for automatic speech recognition. A sequence of speech features which characterize an unknown speech input provided on an audio input channel and associated meta-data which characterize the audio input channel are received. The speech features are transformed with a computer process that uses a trained mapping function controlled by the meta-data, and automatic speech recognition is performed of the transformed speech features.

摘要翻译： 描述了用于自动语音识别的前端语音处理的计算机实现的方法。 接收表征提供在音频输入通道上的未知语音输入和表征音频输入通道的相关联的元数据数据的语音特征序列。 通过使用由元数据控制的经过训练的映射函数的计算机处理来转换语音特征，并且对转换后的语音特征进行自动语音识别。