首页> 外国专利> Acoustic model parameter update treatment manner, acoustic model parameter update central processing unit, program and record media

Acoustic model parameter update treatment manner, acoustic model parameter update central processing unit, program and record media

机译：声学模型参数更新处理方式，声学模型参数更新中央处理器，程序和记录介质

页面导航

摘要
著录项
相似文献

摘要

P To reduction a computational complexity for optimization of recognition parameters, such as a language weight, a word insertion penalty, and a beam width and accompanying conversion of a feature quantity of a feature parameter extracted from an input speech. PSOLUTION: In a sound model update processing method of outputting, as a recognition result, a speech unit that a model having maximum collation likelihood represents by extracting a feature quantity of an input speech signal, subjecting the extracted feature quantity to feature quantity conversion processing, and collating the feature-quantity converted feature quantity with a sound model parameter, a language parameter, and a recognition parameter, and adaptively processing sound model parameters to be used for speech recognition and respective dimensions of the feature quantity are multiplied by a parameter range conversion coefficient (k) after the feature quantity conversion processing of the feature parameter to suppress range variation in sound score during the feature quantity conversion. PCOPYRIGHT: (C)2008 and JPO& INPIT

机译：

为了降低用于识别参数的优化的计算复杂度，所述识别参数诸如语言权重，单词插入罚分，波束宽度以及从输入语音中提取的特征参数的特征量的伴随转换。

解决方案：在声音模型更新处理方法中，通过提取输入的语音信号的特征量，并对提取的特征量进行特征量处理，输出具有最大核对似然性的模型所代表的语音单元作为识别结果。转换处理，将特征量转换后的特征量与声音模型参数，语言参数和识别参数进行核对，并自适应处理用于语音识别的声音模型参数和特征量的各个维度特征量的特征量转换处理后的参数范围转换系数（k），以抑制特征量转换期间声音分数的范围变化。

版权：（C）2008和JPO＆INPIT 展开▼

著录项

公开/公告号JP4856526B2

专利类型
公开/公告日2012-01-18

原文格式PDF
申请/专利权人日本電信電話株式会社;
展开▼

申请/专利号JP20060328029
发明设计人小橋川哲;小川厚徳;
展开▼

申请日2006-12-05
分类号G10L15/06;
国家 JP
入库时间 2022-08-21 17:37:41

相似文献

专利
外文文献
中文文献