Compact Thai TTS distilled from voice-cloning teacher runs on-device
wayu-ai · hf · 2026-09-14
wayu-ai released a compact Thai text-to-speech model distilled from a large voice-cloning teacher, using synthetic speech generated from brief voice references. It achieves strong on-device accuracy and prosody with minimal errors in a fixed-voice setup.
More from Multimodal
- Floor-plan-to-whitebox demo shows Astra hallucinating camera positions — letandrewcook · 2026-09-14
- 10 ChatGPT photo-editing prompt templates that beat vague one-line requests — TawohAwa · 2026-09-14
- Seedance 2.5 demo shows realistic skin pores and flyaway hairs in motion — nikola_mr64990 · 2026-09-14
- Dev open-sources Claude skills repo that turns LLM into a creative partner with a web studio UI — jh3yy · 2026-09-14
- AI-generated seaside coming-of-age scene looks straight out of a real film — SimplyAnnisa · 2026-09-14
- Midjourney V8.2 style code 612407510 earns community praise — azed_ai · 2026-09-14