arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2108.06270eess.AScs.AI
收藏

增强表达性神经文本转语音的音频质量

Enhancing audio quality for expressive Neural Text-to-Speech

Amazon Text-to-Speech Research(亚马逊文本转语音研究院)

Abdelhamid Ezzerg, Adam Gabrys, Bartosz Putrycz, Daniel Korzekwa, Daniel Saez-Trigueros, David McHardy, Kamil Pokora, Jakub Lachowicz, Jaime Lorenzo-Trueba, Viacheslav Klimkov

摘要:

人工语音合成在自然度方面取得了巨大的飞跃,因为最近的文本转语音(TTS)系统能够产生与人类录音质量相似的语音。然而,并不是所有的说话风格都很容易建模:即使对最近的TTS架构来说,高度表达的声音仍然具有挑战性,因为在生成的音频的表达能力和信号质量之间似乎存在权衡。在本文中,我们提出了一套技术,可以利用它来提高信号质量的一个高度表达的声音,而不使用额外的数据。建议的技术包括:在训练过程中调整自回归循环的粒度;在声学建模中使用生成对抗网络;以及在声学模型和神经声码器中使用变分自动编码器。我们发现,当结合起来,这些技术大大关闭了39%的MUSHRA分数方面的表现力的名人的声音在基线系统和录音之间的感知自然度的差距。

英文摘要:

Artificial speech synthesis has made a great leap in terms of naturalness as recent Text-to-Speech (TTS) systems are capable of producing speech with similar quality to human recordings. However, not all speaking styles are easy to model: highly expressive voices are still challenging even to recent TTS architectures since there seems to be a trade-off between expressiveness in a generated audio and its signal quality. In this paper, we present a set of techniques that can be leveraged to enhance the signal quality of a highly-expressive voice without the use of additional data. The proposed techniques include: tuning the autoregressive loop's granularity during training; using Generative Adversarial Networks in acoustic modelling; and the use of Variational Auto-Encoders in both the acoustic model and the neural vocoder. We show that, when combined, these techniques greatly closed the gap in perceived naturalness between the baseline system and recordings by 39% in terms of MUSHRA scores for an expressive celebrity voice.

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。

↑