ISCA Archive Interspeech 2021
ISCA Archive Interspeech 2021

Improving the Expressiveness of Neural Vocoding with Non-Affine Normalizing Flows

Adam Gabryś, Yunlong Jiao, Viacheslav Klimkov, Daniel Korzekwa, Roberto Barra-Chicote

This paper proposes a general enhancement to the Normalizing Flows (NF) used in neural vocoding. As a case study, we improve expressive speech vocoding with a revamped Parallel Wavenet (PW). Specifically, we propose to extend the affine transformation of PW to the more expressive invertible non-affine function. The greater expressiveness of the improved PW leads to better-perceived signal quality and naturalness in the waveform reconstruction and text-to-speech (TTS) tasks. We evaluate the model across different speaking styles on a multi-speaker, multi-lingual dataset. In the waveform reconstruction task, the proposed model closes the naturalness and signal quality gap from the original PW to recordings by 10%, and from other state-of-the-art neural vocoding systems by more than 60%. We also demonstrate improvements in objective metrics on the evaluation test set with L2 Spectral Distance and Cross-Entropy reduced by 3% and 6‰ comparing to the affine PW. Furthermore, we extend the probability density distillation procedure proposed by the original PW paper, so that it works with any non-affine invertible and differentiable function.

doi: 10.21437/Interspeech.2021-1555

Cite as: Gabryś, A., Jiao, Y., Klimkov, V., Korzekwa, D., Barra-Chicote, R. (2021) Improving the Expressiveness of Neural Vocoding with Non-Affine Normalizing Flows. Proc. Interspeech 2021, 1679-1683, doi: 10.21437/Interspeech.2021-1555

  author={Adam Gabryś and Yunlong Jiao and Viacheslav Klimkov and Daniel Korzekwa and Roberto Barra-Chicote},
  title={{Improving the Expressiveness of Neural Vocoding with Non-Affine Normalizing Flows}},
  booktitle={Proc. Interspeech 2021},