情感语音合成综述

情感语音合成作为一个新兴的语音合成方向，糅合生理学、心理学、语言学和信息科学等各学科知识，可以应用于文本阅读、信息查询发布和计算机辅助教学等领域，能够很好地将语音的口语分析、情感分析与计算机技术有机融合，为实现以人为本，具有个性化特征的语音合成系统奠定基础。目前的情感语音合成工作可分为基于规则合成和基于波形拼接合成两类。情感语音合成研究分为情感分析和语音合成两个部分。其中．情感分析的主要工作是收集不同情感的语音数据、提取声学特征，分析声学特征与情感联系；语音合成的主要工作是建立情感转换模型，利用情感转换模型实现合成。

Summary of Emotional Speech Synthesis

Emotional speech synthesis as a new direction of a speech synthesis, which is involved in physiology, psychology, linguistics, information science and other disciplines of knowledge, can be applied fo text reading, publish information query system, computer aided instruction and so on, the organic integration of speaking convergence of voice analysis, sentiment analysis and computer technology to realize peo- ple-oriented and establish the foundation for the speech synthesis system with personalized features. Currently emotional speech synthesis categories includes rule-based synthesis and splicing-based waveform synthesis. Study on emotional speech synthesis is divided into e- motional analysis and speech synthesis of two parts. The main work of emotional analysis, collect speech data from different emotional a- coustic, features extraction, analysis of acoustic characteristics and emotional connection; the main work of speech synthesis is to establish emotional conversion model, use the model for synthesis.