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 2009.08474eess.AScs.LGcs.SD
收藏

表达性语音合成的层次多粒度生成模型

Hierarchical Multi-Grained Generative Model for Expressive Speech Synthesis

Nagoya Institute of Technology(名古屋工业大学);Microsoft Development Co., Ltd.(微软开发有限公司)

Yukiya Hono, Kazuna Tsuboi, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, Keiichi Tokuda

摘要:

本文提出了一种多粒度隐变量的层次生成模型来合成表达性语音。近年来,文本到语音合成中引入了细粒度的隐变量,使得能够对合成语音的韵律和说话风格进行精细控制。然而,当这些潜变量通过从标准高斯先验中采样来获得时,语音的自然度降低。为了解决这个问题,我们提出了一个新的框架,考虑到依赖于输入文本,层次语言结构,和时间结构的潜变量建模的细粒度的潜变量。该框架包括一个多粒度的变分自编码器,一个条件先验,和一个多层次的自回归潜转换器,以获得不同的时间分辨率的潜变量,并考虑到输入文本从粗层次的潜变量采样细层次的潜变量。实验结果表明,适当的方法采样细粒度的潜变量没有参考信号在合成阶段。我们提出的框架还提供了在整个话语的说话风格的可控性。

英文摘要:

This paper proposes a hierarchical generative model with a multi-grained latent variable to synthesize expressive speech. In recent years, fine-grained latent variables are introduced into the text-to-speech synthesis that enable the fine control of the prosody and speaking styles of synthesized speech. However, the naturalness of speech degrades when these latent variables are obtained by sampling from the standard Gaussian prior. To solve this problem, we propose a novel framework for modeling the fine-grained latent variables, considering the dependence on an input text, a hierarchical linguistic structure, and a temporal structure of latent variables. This framework consists of a multi-grained variational autoencoder, a conditional prior, and a multi-level auto-regressive latent converter to obtain the different time-resolution latent variables and sample the finer-level latent variables from the coarser-level ones by taking into account the input text. Experimental results indicate an appropriate method of sampling fine-grained latent variables without the reference signal at the synthesis stage. Our proposed framework also provides the controllability of speaking style in an entire utterance.

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。

↑