// Official vendor repos
Kyutai
4 tracked official repos · 4 releases in the last 30 days.
01Latest releases
-
kyutai-labs/pocket-tts v4.0.1 Patch
Fix issue https://github.com/kyutai-labs/pocket-tts/issues/390 for the sub-20m models.
-
kyutai-labs/pocket-tts v4.0.0 Major New model
new models were added, really strong for their sizes! Select them with `--language`
-
kyutai-labs/pocket-tts v3.3.0 Minor
Default every shipped model to temperature 0.3 by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/324
-
kyutai-labs/pocket-tts v3.2.0 Minor
The french language has been updated, we now have `"french"` and a better `"french_24l"`!
02All repos
| Repo | Category | ★ | Latest | Licence |
|---|---|---|---|---|
| moshiMoshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec. | Speech & audio | 11.3k | rustymimi-0.2.22024-09-22 | Apache-2.0 |
| pocket-ttsA TTS that fits in your CPU (and pocket) | Speech & audio | 9,991 | v4.0.12026-10-21 | MIT |
| delayed-streams-modelingKyutai's Speech-To-Text and Text-To-Speech models based on the Delayed Streams Modeling framework. | Speech & audio | 3,043 | — | Apache-2.0 |
| hibikiHibiki is a model for streaming speech translation (also known as simultaneous translation). Unlike offline translation—where one waits for the end of the source utterance to st… | Speech & audio | 1,528 | — | Apache-2.0 |