ISCA Archive SSW 2021
ISCA Archive SSW 2021

Enhancing audio quality for expressive Neural Text-to-Speech

Abdelhamid Ezzerg, Adam Gabrys, Bartosz Putrycz, Daniel Korzekwa, Daniel Saez-Trigueros, David McHardy, Kamil Pokora, Jakub Lachowicz, Jaime Lorenzo-Trueba, Viacheslav Klimkov

Artificial speech synthesis has made a great leap in terms of naturalness as recent Text-to-Speech (TTS) systems are capable of producing speech with similar quality to human recordings. However, not all speaking styles are easy to model: highly expressive voices are still challenging even to recent TTS architectures since there seems to be a trade-off between expressiveness in a generated audio and its signal quality. In this paper, we present a set of techniques that can be leveraged to enhance the signal quality of a highly-expressive voice without the use of additional data. The proposed techniques include: tuning the autoregressive loop’s granularity during training; using Generative Adversarial Networks in acoustic modeling; and the use of Variational Auto-Encoders in both the acoustic model and the neural vocoder. We show that, when combined, these techniques greatly closed the gap in perceived naturalness between the baseline system and recordings by 39% in terms of MUSHRA scores for an expressive celebrity voice.

doi: 10.21437/SSW.2021-14

Cite as: Ezzerg, A., Gabrys, A., Putrycz, B., Korzekwa, D., Saez-Trigueros, D., McHardy, D., Pokora, K., Lachowicz, J., Lorenzo-Trueba, J., Klimkov, V. (2021) Enhancing audio quality for expressive Neural Text-to-Speech. Proc. 11th ISCA Speech Synthesis Workshop (SSW 11), 78-83, doi: 10.21437/SSW.2021-14

  author={Abdelhamid Ezzerg and Adam Gabrys and Bartosz Putrycz and Daniel Korzekwa and Daniel Saez-Trigueros and David McHardy and Kamil Pokora and Jakub Lachowicz and Jaime Lorenzo-Trueba and Viacheslav Klimkov},
  title={{Enhancing audio quality for expressive Neural Text-to-Speech}},
  booktitle={Proc. 11th ISCA Speech Synthesis Workshop (SSW 11)},