- 3-D Convolutional Recurrent Neural Networks with Attention Model for Speech Emotion Recognition Mingyi Chen, Xuanji He, Jing Yang, and Han Zhang [paper]
CRNN
IEMOCAP数据库
- Speech Emotion Recognition Using Deep Neural Network and Extreme Learning Machine Kun Han, Dong Yu, Ivan Tashev [paper]
DNN
Extreme Learning Machine
- Interpretazione affettiva di feature audio Marco Faleri [paper]
Italy
PCA + SVM + KNN
复现了代码，使用PCA对声音特征数据降维之后，再使用SVM，在中文语音数据集上准确率最高为77.8%
使用KNN神经网络调参还没有复现，作者实现之后的准确率最高达到89.62%，数据集为柏林数据集，不过二者相差结果不大
目前代码已经整理完成，具体使用方法参见[README.md]
- Emotion Recognition from Chinese Speech for Smart Affective Services Using a Combination of SVM and DBN Lianzhang Zhu, Leiming Chen, Dehai Zhao, Jiehan Zhou and Weishan Zhang [paper]
SVM + DBN
中文语音情感集，作者的准确度达到95.8%??
- 复现Interpretazione affettiva di feature audio的代码[codes]
PCA + SVM + KNN
- CRNN [codes]
IEMOCAP数据集
- BLSTM [codes]
柏林数据集
作者最好的准确率68.6%