| TTS系统中基于双音素的基元选择方法 |
| |
| 引用本文： | 杨玉红,胡瑞敏,徐正全,艾浩军.TTS系统中基于双音素的基元选择方法[J].武汉大学学报(信息科学版),2002,27(1):94-97. |
| |
| 作者姓名： | 杨玉红 胡瑞敏 徐正全 艾浩军 |
| |
| 作者单位： | 武汉大学多媒体网络通信工程湖北省重点实验室,武汉市珞喻路129号,430079 |
| |
| 摘 要： | 为寻示能较好解决音节内和音节间的协同发音单元方案，提出了采用类似英文语转换系统中使用的双音素作为合成单元方案，并根据普通话音中只包含410个音节特点，进一步完善了双音素在汉语中的应用。试验结果表明，该方案包含了连续语流中的所有过渡音征，使合成语音转接流畅、自然。
|
| 关 键 词： | TTS系统 双音素 基元选择 汉语语音合成 拼接单元 文语转换 |
| 文章编号： | 1000-050X(2002)01-0094-04 |
| 修稿时间： | 2001年12月10 |
Diphone-based Unit Selection in Text-to-Speech Conversion for Mandarin |
| |
| YANG Yuhong,HU Ruimin,XU Zhengquan,AI Haojun.Diphone-based Unit Selection in Text-to-Speech Conversion for Mandarin[J].Geomatics and Information Science of Wuhan University,2002,27(1):94-97. |
| |
| Authors: | YANG Yuhong HU Ruimin XU Zhengquan AI Haojun |
| |
| Institution: | YANG Yuhong 1 HU Ruimin 1 XU Zhengquan 1 AI Haojun 1 |
| |
| Abstract: | Most of the mandarin text_to_speech systems are syllable_based which only include syllable_internal coarticulation while cross out any cross_syllable coarticulation. One solution to this problem is to abandon syllable_based models in favor of units which can model both syllable_internal and cross_syllable coarticulation. One such unit is the diphone which has been quite used in English TTS sysem. The concept of diphone can be improved in Chinese speech synthesis for there are only 410 syllables in the mandarin.It is shown that diphone_based models can model cross_syllable coarticulation and produce natural_sounding speech rather than syllable_based systems which just insert silence between syllables. |
| |
| Keywords: | Chinese speech synthesis concatenative units diphone text to speech |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
