Research

Selected Publications

2019

Research Projects

Parallel Deep Neural Network for Speech Emotion Recognition

Previous work for emotion recognition have mostly focused on the extraction of carefully hand-crafted and highly-engineered features. Results from these works have demonstrated the importance of discriminative spatio-temporal features to model the continual evolutions of different emotions. How to learn the effective compositional spatio-temporal dynamics for speech emotion recognition has been a fundamental problem of deep representations, herein denoted as deep spectrum representations. In this work, we expolored the possibility of combining speech sptial and temporal features by leveraging CNN and LSTM with a parallel structure.

文章目录
  1. 1. Selected Publications
    1. 1.1. 2019
  2. 2. Research Projects
    1. 2.1. Parallel Deep Neural Network for Speech Emotion Recognition