基于时域的男女声语音转换新途径的研究
文献类型：学位论文
| 作者 | 刘立 |
| 学位类别 | 博士 |
| 答辩日期 | 2000 |
| 授予单位 | 中国科学院声学研究所 |
| 授予地点 | 中国科学院声学研究所 |
| 关键词 | 语音转换 基频 共振峰 |
| 中文摘要 | 本文提出了一种基于时域的男女声语音转换的方法。该方法主要考虑了影响语音个人特性两个主要参数：基频和共振峰。通过把一个基音周期内部的语音信号幅度最小的一部分截去或添加来实现语音基频的改变。在共振峰方面，根据语音半波宽度与共振峰的关系，利用以前的同学在进行语音半波编码算法时所形成的语音半波波形矢量库，依据一定的比例通过DTW技术从矢量库里寻找一定宽度的半波波形替换原始语音信号中的半波波形，从而实现语音共振峰的改变。实验结果表明本算法是可行的。女声、男声平均基频的比例关系为1.5， 共振峰平均的比例关系为1.2。转换后的男声比转换后的女声效果要好一些。本算法的优点是完全从时域上进行语音特性的改变，方法简单易行，物理概念清晰、直观。该技术可应用于语音合成系统，使输出的语音更丰富多样。 |
| 英文摘要 | In this paper, we put forward a time -domain female-male voice conversion algorithm. This method mainly focuses on two acoustic features which are thought to be the most important to speech individuality: pitch frequency and formant frequencies. To change pitch frequency, we cut off or add the low amplitude parts of a pitch period speech signals. And to change formants, according to the relationship between zero-cross rate and formants, and basing on the semi-waveform vector database which the former students formed during carrying out the speech encoding algorithm, we use DTW technology to find a semi-waveform in the database to substitute the original semi-waveform. Experiments show that it is feasible. And the average ratio of pitch frequency between the female and the male is about 1.5 and the average ratio of formant frequencies between them is about 1.2. And the converted male voice is better than the converted female conversions. The advantage of this method is that the conversion of speech individuality is totally based on time domain, the algorithm is easy going, and the physical concept is clear and direct. This technology can be used in speech synthesis system and make the output speech more flexible. |
| 语种 | 中文 |
| 公开日期 | 2011-05-07 |
| 页码 | 41 |
| 源URL | [http://159.226.59.140/handle/311008/698]  |
| 专题 | 声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文 |
| 推荐引用方式 GB/T 7714 | 刘立. 基于时域的男女声语音转换新途径的研究[D]. 中国科学院声学研究所. 中国科学院声学研究所. 2000. |
入库方式： OAI收割
来源：声学研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。