Metraipolis › archive › dc_ee594a16bf8b

Building and Evaluating Fixed-Voice Thai TTS from Synthetic Speech — arXiv:2609.03502

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.03502
License (as archived)
CC0-1.0
Archived
2026-10-17T01:11:54.132Z
SHA-256 of the archived text
5e2e73ea6537fa27d3ba4a93890944299c0bbb5c8917590d9a29a115f9225cd9
Stable URL of this page
https://metraipolis.com/documento/dc_ee594a16bf8b

How to cite

Building and Evaluating Fixed-Voice Thai TTS from Synthetic Speech — arXiv:2609.03502 — arXiv (en), https://arxiv.org/abs/2609.03502 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ee594a16bf8b (Metraipolis archive dc_ee594a16bf8b, sha256 5e2e73ea6537fa27d3ba4a93890944299c0bbb5c8917590d9a29a115f9225cd9)
BibTeX
@misc{dcee594a16bf8b,
  title = {Building and Evaluating Fixed-Voice Thai TTS from Synthetic Speech — arXiv:2609.03502},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.03502},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ee594a16bf8b; sha256 5e2e73ea6537fa27d3ba4a93890944299c0bbb5c8917590d9a29a115f9225cd9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.03502. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In low-resource settings, deploying TTS typically requires choosing between a large voice-cloning model with costly inference or a compact fixed-voice system that requires a speaker-specific corpus. We study a third route: using a large voice-cloning model as a programmable data source to turn a short voice reference (e.g., 15 seconds) into a compact fixed-voice student trained entirely on synthetic speech. This setting makes pipeline design consequential: teacher errors become training targets, while filtering failed generations can reduce coverage of difficult texts. Thai further introduces challenges from ambiguous word boundaries, lexical tone, names and loanwords, numeric verbalization, and Thai-English code-switching. We study how text preparation, synthetic generation, quality filtering, rejection sampling, and frontend choices affect the resulting student, and where teacher limitations remain. We evaluate CER, Challenge-Set Keyword Accuracy, Prosody Pause Accuracy, speaker similarity, and speaking rate. The resulting 82M-parameter model, Wayu-Paxa-TTS-Edge, enables on-device Thai TTS without reference audio. It achieves 68.2% Challenge-Set Keyword Accuracy (85.5% of Gemini 3.1) and 91.4% pause precision, outperforming its OmniVoice teacher (89.9%) and reaching 94.8% of Gemini 3.1. It also achieves the lowest pause-placement error and intra-word pause rates among the three systems, and 3.7% and 1.1% CER on Thai and English, respectively. We open-source the model and evaluation framework for Thai TTS development.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.03502 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T00:28:02.969Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).