摘要
提出了一种利用全局和时序结构的组合特征以及MMD进行情感特征识别的方法.对于从10名话者中 搜集的带有欢快,愤怒,惊奇和悲伤4种情感的1000句语句,利用提出的新的识别方法获得了94％的平均情感识别 率.
This paper presents a new emotion recognition method based on the hybrid feature and MMD.The hybrid feature was composed of global and time sequence. Experiments were conducted on a task of 10 speakers' 1000 sentences including happy, angry, surprising and sorrowful emotions to demonstrate has effectiveness of the new method. The average emotion recognition rate was reached as higher as 94% .
出处
《电子学报》
EI
CAS
CSCD
北大核心
2004年第4期606-609,共4页
Acta Electronica Sinica
基金
教育部<面向21世纪教育振兴行动计划>
教育部科学技术重点项目(No.03082)
关键词
语音信号
情感特征分析
MMD
情感识别
speech signal
emotional feature analysis
MMD
emotion recognition
