Synthesis of Tibetan Amdo based on VITS

DOI

10.2991/978-94-6463-490-7_43How to use a DOI?

Keywords

Tibetan Amdo; VITS; Latin alphabet; speech synthesis

Abstract

Tibetan Amdo, a significant dialect of the Tibetan language, currently lacks large-scale, high-quality speech databases. It faces challenges such as a limited number of researchers, incomplete and inaccurate coverage of the Tibetan phoneme lexicon, and subpar quality of synthesized speech. This paper employs the VITS framework for Tibetan Amdo speech synthesis, exploring the conversion of Tibetan characters into Latin letters for speech synthesis. The experimental results indicate that synthesizing natural and fluent Tibetan Amdo speech based on Latin alphabet conversion yields better outcomes, with a Mean Opinion Score (MOS) of 4.13, providing an effective approach for Tibetan Amdo speech synthesis.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Chao Wei
AU  - Guanyu Li
AU  - Dongliang Chen
PY  - 2024
DA  - 2024/08/31
TI  - Synthesis of Tibetan Amdo based on VITS
BT  - Proceedings of the 2024 3rd International Conference on Artificial Intelligence, Internet and Digital Economy (ICAID 2024)
PB  - Atlantis Press
SP  - 391
EP  - 397
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-490-7_43
DO  - 10.2991/978-94-6463-490-7_43
ID  - Wei2024
ER  -