arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2203.16928cs.SDcs.CLcs.LGeess.AS
收藏

用于语音情感识别的神经结构搜索

Neural Architecture Search for Speech Emotion Recognition

CUHK Stanley Ho Big Data Decision Analytics Research Centre(香港中文大学何鸿燊大数据决策分析研究中心);Centre for Perceptual and Interactive Intelligence(感知与交互智能研究中心)

Xixin Wu, Shoukang Hu, Zhiyong Wu, Xunying Liu, Helen Meng

摘要:

深度神经网络为语音情感识别(SER)带来了重大进步。然而,SER中的体系结构设计主要基于专家知识和经验(试错)评估,这是耗时和资源密集型的。在本文中,我们建议应用神经架构搜索(NAS)技术来自动配置SER模型。为了加快候选架构的优化,我们提出了一个统一的路径丢弃策略,鼓励所有候选架构操作,以平等地优化。在IEMOCAP上对两种不同神经网络结构的实验结果表明,NAS在保持模型参数大小不变的情况下,可以提高SER性能(54.89%~ 56.28%).所提出的辍学策略也显示出优于以前的方法。

英文摘要:

Deep neural networks have brought significant advancements to speech emotion recognition (SER). However, the architecture design in SER is mainly based on expert knowledge and empirical (trial-and-error) evaluations, which is time-consuming and resource intensive. In this paper, we propose to apply neural architecture search (NAS) techniques to automatically configure the SER models. To accelerate the candidate architecture optimization, we propose a uniform path dropout strategy to encourage all candidate architecture operations to be equally optimized. Experimental results of two different neural structures on IEMOCAP show that NAS can improve SER performance (54.89\% to 56.28\%) while maintaining model parameter sizes. The proposed dropout strategy also shows superiority over the previous approaches.

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。

↑