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 2008.09483eess.AScs.CLcs.LGcs.SD
收藏

笑声合成:结合Seq2seq建模和迁移学习

Laughter Synthesis: Combining Seq2seq modeling with Transfer Learning

Numediart Institute, University of Mons(蒙斯大学努梅迪亚尔研究所)

Noé Tits, Kevin El Haddad, Thierry Dutoit

摘要:

尽管人们对表达性语音合成越来越感兴趣,但非语言表达的合成是一个未被开发的领域。在本文中,我们提出了一个音频笑声合成系统的基础上,序列到序列的TTS合成系统。我们通过训练深度学习模型来利用迁移学习,学习从注释中生成语音和笑声。我们用听力测试来评估我们的模型,将其性能与基于HMM的笑声合成进行比较,并评估它达到了更高的感知自然度。我们的解决方案是迈向TTS系统的第一步,该系统将能够合成语音,并在娱乐水平上控制笑声集成。

英文摘要:

Despite the growing interest for expressive speech synthesis, synthesis of nonverbal expressions is an under-explored area. In this paper we propose an audio laughter synthesis system based on a sequence-to-sequence TTS synthesis system. We leverage transfer learning by training a deep learning model to learn to generate both speech and laughs from annotations. We evaluate our model with a listening test, comparing its performance to an HMM-based laughter synthesis one and assess that it reaches higher perceived naturalness. Our solution is a first step towards a TTS system that would be able to synthesize speech with a control on amusement level with laughter integration.

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。

↑