Embedding a Differentiable Mel-Cepstral Synthesis Filter to a Neural Speech Synthesis System
Embedding a Differentiable Mel-Cepstral Synthesis Filter to a Neural Speech Synthesis System
复制标题
DOI：
10.1109/icassp49357.2023.10094872
复制
发表时间：
2022-11
期刊：
影响因子：
--
通讯作者：
Takenori Yoshimura;Shinji Takaki;Kazuhiro Nakamura;Keiichiro Oura;Yukiya Hono;Kei Hashimoto;Yoshihiko Nankaku;K. Tokuda
中科院分区：
文献类型：
--
作者：
Takenori Yoshimura;Shinji Takaki;Kazuhiro Nakamura;Keiichiro Oura;Yukiya Hono;Kei Hashimoto;Yoshihiko Nankaku;K. Tokuda
本文将经典的Mel-Cepstral合成滤波器集成到一个面向端到端可控语音合成的现代神经语音合成系统中。在该系统中，由于Mel-Cepstral合成滤波器被显式嵌入到神经波形模型中，因此合成语音的语音特征和基音高分别通过频率扭曲参数和基频来控制。我们将Mel-Cepstral合成滤波器实现为一个可区分的、对GPU友好的模块，以使所提出的系统中的声学模型和波形模型能够以端到端的方式同时优化。实验表明，该系统在保持可控性的基础上，提高了语音质量。实验中使用的核心PyTorch模块在GitHub1上公开提供。
This paper integrates a classic mel-cepstral synthesis filter into a modern neural speech synthesis system towards end-to-end controllable speech synthesis. Since the mel-cepstral synthesis filter is explicitly embedded in neural waveform models in the proposed system, both voice characteristics and the pitch of synthesized speech are highly controlled via a frequency warping parameter and fundamental frequency, respectively. We implement the mel-cepstral synthesis filter as a differentiable and GPU-friendly module to enable the acoustic and waveform models in the proposed system to be simultaneously optimized in an end-to-end manner. Experiments show that the proposed system improves speech quality from a baseline system maintaining controllability. The core PyTorch modules used in the experiments are publicly available on GitHub1.