arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2006.08129eess.AScs.CVcs.LGcs.SD
收藏

使用深度神经网络进行音频和视频中的情感识别

Emotion Recognition in Audio and Video Using Deep Neural Networks

Stanford University(斯坦福大学)

Mandeep Singh, Yuan Fang

摘要:

人类能够理解来自多个领域的信息,例如语音,文本和视觉。随着深度学习技术的进步,语音识别有了显著的改进。从语音中识别情感是一个重要的方面,随着深度学习技术的发展,情感识别的准确性和延迟都有所改善。在提高准确性方面仍然存在许多挑战。在这项工作中,我们试图探索不同的神经网络,以提高情感识别的准确性。通过探索不同的架构,我们发现(CNN+RNN)+3D CNN多模型架构,该架构处理音频频谱图和相应的视频帧,使用IEMOCAP[2]数据集,在4种情绪中的情感预测准确率为54.0%,在3种情绪中的情感预测准确率为71.75%。

英文摘要:

Humans are able to comprehend information from multiple domains for e.g. speech, text and visual. With advancement of deep learning technology there has been significant improvement of speech recognition. Recognizing emotion from speech is important aspect and with deep learning technology emotion recognition has improved in accuracy and latency. There are still many challenges to improve accuracy. In this work, we attempt to explore different neural networks to improve accuracy of emotion recognition. With different architectures explored, we find (CNN+RNN) + 3DCNN multi-model architecture which processes audio spectrograms and corresponding video frames giving emotion prediction accuracy of 54.0% among 4 emotions and 71.75% among 3 emotions using IEMOCAP[2] dataset.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑