Kyutai
TTS 1 model no stock voices based in France open weights only data from provider API synced 2026-11-18
Non-profit AI research lab in Paris that releases speech models as open weights (Moshi full-duplex speech model, streaming STT and TTS), mostly under CC-BY-4.0. Its commercial spin-off Gradium sells a hosted API and is listed separately.
Text-to-speech 1
| Compare | Model | Price |
|---|---|---|
kyutai/tts-1.6b-en_frtts-1.6b-en_fr |
– |