Automatic Recognition of Emotional State in Polish Speech

被引:0
|
作者
Staroniewicz, Piotr [1 ]
机构
[1] Wroclaw Univ Technol, Inst Telecommun Teleinformat & Acoust, PL-50370 Wroclaw, Poland
关键词
emotional speech; emotional state recognition;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The paper presents the comparison of scores for emotional state automatic recognition tests. The database of Polish emotional speech used during tests includes recordings of six acted emotional states (anger, sadness, happiness, fear, disgust, surprise) and the neutral state of 13 amateur speakers (2118 utterances). The features based on F0, intensity, formants and LPC coefficients were applied in seven chosen classifiers. The highest scores were reached for SVM, ANN and DTC classifiers.
引用
收藏
页码:347 / 353
页数:7
相关论文
共 50 条