| 用于统计语音合成的大尺度压缩HMM的方法 |
| |
| 引用本文： | 那兴宇,谢湘,匡镜明,何娅玲. 用于统计语音合成的大尺度压缩HMM的方法[J]. 清华大学学报(自然科学版), 2011, 0(9) |
| |
| 作者姓名： | 那兴宇 谢湘 匡镜明 何娅玲 |
| |
| 作者单位： | 北京理工大学信息与电子学院;北京益世通利公司; |
| |
| 基金项目： | 国家自然科学基金青年科学基金项目(61001188); 教育部博士点基金(20101101110020) |
| |
| 摘 要： | 统计语音合成使用隐Markov模型(HMM)作为声学特征的统计模型。提出了一种利用声学模型空间距离进行HMM的大尺度压缩的量化方法,通过对矢量量化码本进行的优选迭代步骤,减小压缩后的声道谱模型与原模型之间的声学距离,使通过量化模型合成的语音更加接近未量化模型。主观和客观测试结果显示:使用该方法进行声道谱模型的压缩,在压缩至原模型大小的0.06左右时,仍有约90%的评价得分认为合成语音的质量没有明显下降。
|
| 关 键 词： | 隐Markov模型(HMM)语音合成 HMM量化 矢量量化 |
Large scale compression of HMM for statistical speech synthesis |
| |
| NA Xingyu,XIE Xiang,KUANG Jingming,HE Yaling. Large scale compression of HMM for statistical speech synthesis[J]. Journal of Tsinghua University(Science and Technology), 2011, 0(9) |
| |
| Authors: | NA Xingyu XIE Xiang KUANG Jingming HE Yaling |
| |
| Affiliation: | NA Xingyu1,XIE Xiang1,KUANG Jingming1,HE Yaling2(1.School of Information and Electronics,Beijing Institute of Technology,100081,China,2.Eastel Corporation,Beijing 100085,China) |
| |
| Abstract: | A hidden Markov model(HMM) is used as the statistical model for the acoustic parameters in statistical speech synthesis.This paper presents a quantization method for large scale compression of the HMM based on the acoustic space distance in the speech models.The quality loss caused by the quantization is reduced by an optimal iteration procedure that optimizes the vector quantization codebook.Objective and subjective evaluations show that 90% of the scores indicate no significant reduction of the speech qua... |
| |
| Keywords: | hidden Markov model(HMM) based speech synthesis quantized HMM vector quantization |
| 本文献已被 CNKI 等数据库收录！ |
|
