公开(公告)号：US07860716B2

公开(公告)日：2010-12-28

申请号：US11789132

申请日：2007-04-24

申请人： Ye Tian ,  Yifan Gong ,  Frank K. Soong

发明人： Ye Tian ,  Yifan Gong ,  Frank K. Soong

IPC分类号： G10L15/10

CPC分类号： G10L15/065 ,  G10L15/01 ,  G10L15/08 ,  G10L2015/0631

摘要： Reliable transcription error-checking algorithm that uses a word confidence score and a word duration probability to detect transcription errors for improved results through the automatic detection of transcription errors in a corpus. The transcription error-checking algorithm is combined model training so as to use a current model to detect transcription errors, remove utterances which contain incorrect transcription (or manually fix the found errors), and retrain the model. This process can be repeated for several iterations to obtain an improved speech recognition model. The speech model is employed to achieve speech-transcription alignment to obtain a word boundary. Speech recognizer is then utilized to generate a word-lattice. Using the word boundary and word lattice, error detection is computed using a word confidence score and a word duration probability.

摘要翻译： 可靠的转录错误检查算法，通过自动检测语料库中的转录错误，使用单词置信度分数和单词持续时间概率来检测转录错误以改善结果。 转录错误检查算法是组合模型训练，以便使用当前模型来检测转录错误，删除包含不正确转录（或手动修复发现的错误）的话语，并重新训练模型。 该过程可以重复几次迭代以获得改进的语音识别模型。 语音模型用于实现语音转录对齐以获得字边界。 然后利用语音识别器来产生一个单词格。 使用单词边界和单词格，使用单词置信分数和单词持续时间概率来计算错误检测。

公开(公告)号：US20080270133A1

公开(公告)日：2008-10-30

申请号：US11789132

申请日：2007-04-24

申请人： Ye Tian ,  Yifan Gong ,  Frank K. Soong

发明人： Ye Tian ,  Yifan Gong ,  Frank K. Soong

IPC分类号： G10L15/00

CPC分类号： G10L15/065 ,  G10L15/01 ,  G10L15/08 ,  G10L2015/0631

摘要： Reliable transcription error-checking algorithm that uses a word confidence score and a word duration probability to detect transcription errors for improved results through the automatic detection of transcription errors in a corpus. The transcription error-checking algorithm is combined model training so as to use a current model to detect transcription errors, remove utterances which contain incorrect transcription (or manually fix the found errors), and retrain the model. This process can be repeated for several iterations to obtain an improved speech recognition model. The speech model is employed to achieve speech-transcription alignment to obtain a word boundary. Speech recognizer is then utilized to generate a word-lattice. Using the word boundary and word lattice, error detection is computed using a word confidence score and a word duration probability.

摘要翻译： 可靠的转录错误检查算法，通过自动检测语料库中的转录错误，使用单词置信度分数和单词持续时间概率来检测转录错误以改善结果。 转录错误检查算法是组合模型训练，以便使用当前模型来检测转录错误，删除包含不正确转录（或手动修复发现的错误）的话语，并重新训练模型。 该过程可以重复几次迭代以获得改进的语音识别模型。 语音模型用于实现语音转录对齐以获得字边界。 然后利用语音识别器来产生一个单词格。 使用单词边界和单词格，使用单词置信分数和单词持续时间概率来计算错误检测。

公开(公告)号：US6009390A

公开(公告)日：1999-12-28

申请号：US927883

申请日：1997-09-11

申请人： Sunil K. Gupta ,  Raziel Haimi-Cohen ,  Frank K. Soong

发明人： Sunil K. Gupta ,  Raziel Haimi-Cohen ,  Frank K. Soong

IPC分类号： G10L15/14 ,  G10L7/08

CPC分类号： G10L15/144

摘要： In a speech recognition system, tied-mixture hidden Markov models (HMMs) are used to match, in the maximum likelihood sense, the phonemes of spoken words given the acoustic input thereof. In a well known manner, such speech recognition requires computation of state observation likelihoods (SOLs). Because of the use of HMMs, each SOL computation involves a substantial number of Gaussian kernels and mixture component weights. In accordance with the invention, the number of Gaussian kernels is cut down to reduce the computational complexity and increase the efficiency of memory access to the kernels. For example, only the non-zero mixture component weights and the Gaussian kernels associated therewith are considered in the SOL computation. In accordance with an aspect of the invention, only a subset of the Gaussian kernels of significant values, regardless of the values of the associated mixture component weights, are considered in the SOL computation. In accordance with another aspect of the invention, at least some of the mixture component weights are quantized to reduce memory space needed to store them. As such, the computational complexity and memory access efficiency are further improved.

摘要翻译： 在语音识别系统中，绑定混合隐马尔可夫模型（HMM）用于在最大似然意义上匹配给定其声输入的口语字的音素。 以众所周知的方式，这种语音识别需要计算状态观察可能性（SOLs）。 由于使用HMM，每个SOL计算涉及大量高斯核和混合分量权重。 根据本发明，削减高斯内核的数量以减少计算复杂度并提高对内核的存储器访问的效率。 例如，在SOL计算中仅考虑非零混合分量权重和与其相关联的高斯内核。 根据本发明的一个方面，在SOL计算中仅考虑与有关混合分量权重值相关的有效值的高斯核的子集。 根据本发明的另一方面，至少部分混合组分权重被量化以减少存储它们所需的存储空间。 因此，计算复杂度和存储器访问效率进一步提高。

公开(公告)号：US5995926A

公开(公告)日：1999-11-30

申请号：US897806

申请日：1997-07-21

申请人： Sunil K. Gupta ,  Frank K. Soong

发明人： Sunil K. Gupta ,  Frank K. Soong

IPC分类号： G10L15/18 ,  H04M3/42 ,  G10L5/00

CPC分类号： G10L15/197 ,  H04M3/42204

摘要： In a speech recognition system for performing voice dialing, an inventive connected digit recognizer is employed to recognize a sequence of spoken digits. The inventive recognizer generates the maximum-likelihood digit sequence corresponding to the spoken sequence in accordance with the Viterbi algorithm. However, unlike a prior art connected digit recognizer, the inventive recognizer does not assume that a digit model in a sequence can be followed by any digit model with equal probability. Rather, the inventive recognizer takes into account, for each digit model being decided on, a conditional probability that that digit model would follow a given digit model preceding thereto.

摘要翻译： 在用于执行语音拨号的语音识别系统中，本发明的连接数字识别器用于识别口语数字序列。 本发明的识别器根据维特比算法生成对应于口语序列的最大似然数字序列。 然而，与现有技术的连接的数字识别器不同，本发明的识别器不认为序列中的数字模型可以跟随具有相等概率的任何数字模型。 相反，本发明的识别器考虑到所确定的每个数字模型，该数字模型将遵循之前的给定数字模型的条件概率。