English

Embedding a Differentiable Mel-cepstral Synthesis Filter to a Neural Speech Synthesis System

Audio and Speech Processing 2022-11-22 v1 Computation and Language Sound

Abstract

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 will be publicly available on GitHub.

Keywords

Cite

@article{arxiv.2211.11222,
  title  = {Embedding a Differentiable Mel-cepstral Synthesis Filter to a Neural Speech Synthesis System},
  author = {Takenori Yoshimura and Shinji Takaki and Kazuhiro Nakamura and Keiichiro Oura and Yukiya Hono and Kei Hashimoto and Yoshihiko Nankaku and Keiichi Tokuda},
  journal= {arXiv preprint arXiv:2211.11222},
  year   = {2022}
}

Comments

Submitted to ICASSP 2023

R2 v1 2026-06-28T06:20:23.484Z