人机交互中的语音情感识别研究进展

摘要

Speech emotion recognition is a current active research topic in the fields of signal processing,pattern recognition,artificial intelligence,human computer interaction,etc.The ultimate purpose of such research is to endow computers with emotion ability and make human computer interaction be genuinely harmonic and natural.This paper reviews the recent advance of several key problems involved in speech emotion recognition,including emotional description theory,emotional speech databases,emotional acoustic analysis as well as emotion recognition methods.In addition,the existing research problems and the future direction are presented.%语音情感识别是当前信号处理、模式识别、人工智能、人机交互等领域的热点研究课题,其研究的最终目的是赋予计算机情感能力,使得人机交互做到真正的和谐和自然.本文综述了语音情感识别所涉及到的几个关键问题,包括情感表示理论、情感语音数据库、情感声学特征分析以及情感识别方法四个方面的最新进展,并指出了研究中存在的问题及下一步发展的方向.