English

Diffusion-Based Mel-Spectrogram Enhancement for Personalized Speech Synthesis with Found Data

Audio and Speech Processing 2023-10-03 v3

Abstract

Creating synthetic voices with found data is challenging, as real-world recordings often contain various types of audio degradation. One way to address this problem is to pre-enhance the speech with an enhancement model and then use the enhanced data for text-to-speech (TTS) model training. This paper investigates the use of conditional diffusion models for generalized speech enhancement, which aims at addressing multiple types of audio degradation simultaneously. The enhancement is performed on the log Mel-spectrogram domain to align with the TTS training objective. Text information is introduced as an additional condition to improve the model robustness. Experiments on real-world recordings demonstrate that the synthetic voice built on data enhanced by the proposed model produces higher-quality synthetic speech, compared to those trained on data enhanced by strong baselines. Code and pre-trained parameters of the proposed enhancement model are available at \url{https://github.com/dmse4tts/DMSE4TTS}

Keywords

Cite

@article{arxiv.2305.10891,
  title  = {Diffusion-Based Mel-Spectrogram Enhancement for Personalized Speech Synthesis with Found Data},
  author = {Yusheng Tian and Wei Liu and Tan Lee},
  journal= {arXiv preprint arXiv:2305.10891},
  year   = {2023}
}

Comments

Accepted to ASRU 2023

R2 v1 2026-06-28T10:38:07.377Z