| 采用重音调整模型的HMM语音合成系统 |
| |
| 引用本文： | 李雅,潘诗锋,陶建华.采用重音调整模型的HMM语音合成系统[J].清华大学学报(自然科学版),2011(9):1171-1175. |
| |
| 作者姓名： | 李雅 潘诗锋 陶建华 |
| |
| 作者单位： | 中国科学院自动化研究所模式识别国家重点实验室; |
| |
| 基金项目： | 国家自然科学基金面上项目(60873160,61011140075);国家自然科学基金重大研究计划项目重点课题(90820303) |
| |
| 摘 要： | 韵律模型一直是语音合成中的研究重点,而重音则是目前韵律研究中的主要难点。在已有的研究工作中重音的定性分析较多,但重音生成则相对较少。该文采用基于重音调整的方法,构建了一个支持重音的隐Markov模型(hid-den Markov model,HMM)语音合成系统。在文本分析模块引入最大熵模型完成了基于文本特征的重音预测,然后根据重音调整韵律参数得到调整后的HMM模型,最后采用基于隐Markov模型的语音合成技术(hidden Markov modelbased speech synthesis,HTS)系统合成语音。实验结果表明:采用该方法能够合成出抑扬顿挫的语音。该方法的优势在于能够灵活地扩展到对其他语音表现力的合成。
|
| 关 键 词： | 重音 语音合成 韵律 |
HMM-based expressive speech synthesis with a flexible Mandarin stress adaptation model |
| |
| LI Ya,PAN Shifeng,TAO Jianhua.HMM-based expressive speech synthesis with a flexible Mandarin stress adaptation model[J].Journal of Tsinghua University(Science and Technology),2011(9):1171-1175. |
| |
| Authors: | LI Ya PAN Shifeng TAO Jianhua |
| |
| Institution: | LI Ya,PAN Shifeng,TAO Jianhua(National Laboratory of Pattern Recognition,Institute of Automation,Chinese Academy of Sciences,Beijing 100190,China) |
| |
| Abstract: | Expressive speech synthesis has recently received much attention.Stress is one key issue which will improve the expressiveness of the synthetic speech.However,little work has been done on Mandarin stress prediction and expression.This paper presents a hidden Markov model(HMM) based expressive speech synthesis system which supports Mandarin stress synthesis.Mandarin stresses are automatically predicted with textual features utilizing a maximum entropy(ME) model.The linear stress adaptation model was extracte... |
| |
| Keywords: | stress text-to-speech prosody |
| 本文献已被 CNKI 等数据库收录！ |
|
