JaiTTS: A Thai Voice Cloning Model

TL;DR

JaiTTS-v1.0 is a Thai voice cloning TTS model that handles numerals and code-switching without normalization, achieving state-of-the-art accuracy and outperforming commercial systems in human evaluations.

Contribution

The paper introduces JaiTTS-v1.0, a novel Thai TTS model that processes numerals and code-switching directly, with superior performance and publicly available code.

Findings

Achieves CER of 1.94%, surpassing human ground truth of 1.98%.

Outperforms commercial TTS systems in human preference tests.

Handles numerals and code-switching without explicit normalization.

Abstract

We present JaiTTS-v1.0, a state-of-the-art Thai voice cloning text-to-speech model built through continual training on a large Thai-centric speech corpus. The model architecture is adapted from VoxCPM, a tokenizer-free autoregressive TTS model. JaiTTS-v1.0 directly processes numerals and Thai-English code-switching, which is very common in realistic settings, without explicit text normalization. We test the models on short- and long-duration speech generation, which reflects many real-world use cases. JaiTTS-v1.0 achieves a state-of-the-art CER of 1.94%, surpassing the human ground truth of 1.98% for short-duration tasks while performing on par with human ground truth for long-duration tasks. In human judgment evaluations, our model wins 283 of 400 pairwise comparisons against commercial flagships, with only 58 losses. Our code and demo are available at https://github.com/JTS-AI-Team/JaiTTS .