Generation of Artificial FO-contours of Emotional Speech with Generative Adversarial Networks
Generation of Artificial FO-contours of Emotional Speech with Generative Adversarial Networks
复制标题
利用生成对抗网络生成情感语音的人工 FO 轮廓
DOI：
10.1109/ssci44817.2019.9002917
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
佐宗晃
中科院分区：
文献类型：
--
作者：
松岡駿平;Yao Jiang;佐宗晃
基频（F0）轮廓对于反映语音样本中说话者的情感、身份、意图和态度起着非常重要的作用。在本文中，我们采用生成对抗网络（GAN）来生成情感语音的人工 F0 轮廓。然而，GAN 也面临着一些局限性，因为训练不稳定，它经常生成不需要的数据，并且它可以重复生成非常相似或相同的数据，这就是所谓的模式崩溃。本研究构建了一种基于GAN的F0轮廓生成模型，可以稳定地生成更加多样化、符合训练数据统计特征的F0轮廓。我们测试了五种生成模型生成的 F0 轮廓中四种情绪的分类率。我们还评估了生成的 F0 轮廓的平均局部密度，以表示生成的 F0 轮廓的多样性。初步实验证实了所提出的生成模型的有效性和有效性。
Fundamental frequency (F0) contours play a very important role in reflecting the emotion, identity, intension, and attitude of a speaker in samples of speech. In this paper, we adopted a generative adversarial network (GAN) to generate artificial F0 contours of emotional speech. The GAN faces some limitations, however, in that it frequently generates undesired data because of unstable training, and it can repeatedly generate very similar or the same data, which is known as mode collapse. This study constructed a GAN-based generative model for F0 contours that can stably generate more-various F0 contours that fit the statistical characteristics of the training data. We tested the classification rate of four kinds of emotions in the F0 contours generated from five kinds of generative models. We also evaluated the averaged local density of the generated F0 contours to represent the variety of the generated F0 contours. Preliminary experiments confirmed the validity and effectiveness of the proposed generative model.