KhanomTan TTS

KhanomTan TTS Ikiganiro

An open, commercially-licensed Thai-first text-to-speech model with multiple speaker voices.

Kwiyandikisha kugirango Inyuguti

Umwandiko in Itagi: ya: Igenzura:

<speak><prosody rate="slow">Slow speech</prosody></speak>

i Byahiswemo Urugero - Kanda Kuri Gukuraho Rimwe Umwandiko:

Urugero: Bisanzwe Umwandiko, Umurongo: Itagi:. Itagi: -, Hindura Kuri Urugero: Nka Cyangwa.

Kugena (Ijambo =):

-12 +12
0.5x 2.0x
Na:,,
Audio Kugaragara. A Urugero:, Injiza Umwandiko, na Kanda.
Byaremwe
0:00
Iyimura Iyimura Ihuza in
Bwite Koresha Kuva: 5 /
TTS.ai? Abayobozi!

Ikiganiro KhanomTan TTS

KhanomTan TTS was built by Thai NLP contributor Wannaphong Phatthiyaphaibun on top of the YourTTS multilingual VITS architecture, and trained on CC0 and other permissively-licensed Thai corpora including TSync. Where most open TTS models only cover Thai under research-only or non-commercial terms, KhanomTan ships under Apache 2.0, making it a rare commercially-safe choice for the language. It offers a small roster of speaker voices and runs fast at roughly five seconds per generation in around 2GB of VRAM. It fits Thai voiceovers, narration for Thai-language apps, and accessibility tooling where licensing clarity matters.

kugirango: Thai voiceovers, Thai-language content and apps

Gushakisha byose KhanomTan TTS Amashusho

A

Mukoraporogaramu
Wannaphong Phatthiyaphaibun
Inyandiko y'Iyemererakoresha
Apache 2.0
Itariki
standard
Umuvuduko
fast
Guhindura izina
Oya
Ururimi:
Thai
Inyuguti
500

KhanomTan TTS Amashusho

Bernard

Thai
Bisanzwe Male

Default

Thai
Bisanzwe Neutral

Kerstin

Thai
Bisanzwe Female

Linda

Thai
Bisanzwe Female

Thorsten

Thai
Bisanzwe Male

KhanomTan TTS -

Yes. KhanomTan is released under the Apache 2.0 license, so it can be used commercially — which is unusual for open Thai TTS, where most alternatives carry non-commercial restrictions.

On TTS.ai it is configured for Thai. The underlying YourTTS architecture is multilingual, but the deployed model is focused on natural Thai pronunciation and prosody.

No. KhanomTan does not support voice cloning here; it generates speech from its built-in speaker voices. For cloning, use a model like Chatterbox, GPT-SoVITS, or IndexTTS-2.
← Amashusho yose