arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.07288eess.AScs.LGcs.SD
收藏

说话人注意语音情感识别

Speaker Attentive Speech Emotion Recognition

STMS Lab IRCAM, CNRS, Sorbonne Université(STMS实验室 IRCAM、法国国家科学研究中心、索邦大学)

Clément Le Moine, Nicolas Obin, Axel Roebel

摘要:

随着深度神经网络(DNN)的出现,语音情感识别(SER)任务在过去几年中有了显着的改进。然而,即使是最成功的方法,在需要适应特定的扬声器和场景时仍然相当失败,不可避免地导致与人类相比表现较差。在本文中,我们提出了新的工作的基础上的思想,教学的情绪识别网络的发言人身份。我们的系统是两个ACRNN分类器分别致力于扬声器和情感识别的组合。第一个通知后者通过一个自我扬声器注意(SSA)的机制,这是相当大的帮助,专注于语音信号的情感信息。在社会态度数据库Att-HACK和IEMOCAP语料库上的实验表明了该方法的有效性,在未加权平均召回率方面达到了最佳的性能。

英文摘要:

Speech Emotion Recognition (SER) task has known significant improvements over the last years with the advent of Deep Neural Networks (DNNs). However, even the most successful methods are still rather failing when adaptation to specific speakers and scenarios is needed, inevitably leading to poorer performances when compared to humans. In this paper, we present novel work based on the idea of teaching the emotion recognition network about speaker identity. Our system is a combination of two ACRNN classifiers respectively dedicated to speaker and emotion recognition. The first informs the latter through a Self Speaker Attention (SSA) mechanism that is shown to considerably help to focus on emotional information of the speech signal. Experiments on social attitudes database Att-HACK and IEMOCAP corpus demonstrate the effectiveness of the proposed method and achieve the state-of-the-art performance in terms of unweighted average recall.

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。

↑