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 2005.08447cs.SDeess.AS
收藏

增强生成对抗网络的语音情感识别

Augmenting Generative Adversarial Networks for Speech Emotion Recognition

University of Southern Queensland(南昆士兰大学);Data61, CSIRO(联邦科学与工业研究组织数据61研究所);Information Technology University(信息技术大学);University of New South Wales(新南威尔士大学);Queensland University of Technology(昆士兰科技大学);Imperial College London(帝国理工学院);University of Augsburg(奥格斯堡大学)

Siddique Latif, Muhammad Asim, Rajib Rana, Sara Khalifa, Raja Jurdak, Björn W. Schuller

摘要:

生成对抗网络(GANs)在学习情感属性和生成新数据样本方面表现出了潜力。然而,它们的性能通常受到较大的语音情感识别(SER)数据的不可用性的阻碍。在这项工作中,我们提出了一个框架,利用混合数据增强计划,以增加GAN的特征学习和生成。为了显示所提出的框架的有效性,我们目前的结果SER(i)合成特征向量,(ii)增强的训练数据与合成功能,(iii)压缩表示的编码功能。我们的研究结果表明,该框架可以有效地学习压缩的情感表示,以及它可以生成合成样本,有助于提高性能的语料库内和跨语料库的评价。

英文摘要:

Generative adversarial networks (GANs) have shown potential in learning emotional attributes and generating new data samples. However, their performance is usually hindered by the unavailability of larger speech emotion recognition (SER) data. In this work, we propose a framework that utilises the mixup data augmentation scheme to augment the GAN in feature learning and generation. To show the effectiveness of the proposed framework, we present results for SER on (i) synthetic feature vectors, (ii) augmentation of the training data with synthetic features, (iii) encoded features in compressed representation. Our results show that the proposed framework can effectively learn compressed emotional representations as well as it can generate synthetic samples that help improve performance in within-corpus and cross-corpus evaluation.

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。

↑