v4.0.0
kyutai-labs/pocket-ttsv4.0.0Oct 20, 2026by gabrieldemarmiesse
Upgrading?
AI Summary
This release introduces six new small English models optimized for speed and size, alongside a Rust-based CLI for improved performance. The update also enhances audio quality with fading techniques to remove clicking sounds and expands multilingual support with new community models in Greek, Turkish, and Japanese.
Key Highlights
- Added 6 new small English models (18m, 7m, 2m) optimized for speed and size.
- Introduced Rust CLI binaries for direct execution and improved performance.
- Improved audio quality with fade-in effects to eliminate onset clicks.
- Expanded multilingual support with new community models (Greek, Turkish, Japanese).
- Retrained major European languages (French, Italian, German, etc.) with full-length voice prompts.
New Features
- Six new small English models (english_18m_2026-10, english_7m_2026-10, english_2m_2026-10).
- Rust-based CLI executable for direct execution.
- Audio quality improvements including fade-in effects and pause detection.
- New community models: Greek (Ελληνικά), Turkish, Japanese (日本語), and Multilingual.
- Retrained models for French, Italian, Dutch, German, Spanish, and Portuguese.
- API updates to serve the new small English models.
Full Release Notes
## Highlights: 6 new models were added, really strong for their sizes! Select them with `--language` * `english_18m_2026-10` * `english_18m_refl_2026-10` * `english_7m_2026-10` * `english_7m_refl_2026-10` * `english_2m_2026-10` (experimental) * `english_2m_refl_2026-10` (experimental) To leverage the speed of those new models we also provide binaries of the pocket-tts runtime, that you can download directly from this Github release page. When you download the executable, it has the same API as the existing CLI of pocket-tts (the python one). ## What's Changed * Add Libratory to projects using Pocket TTS by @subev in https://github.com/kyutai-labs/pocket-tts/pull/327 * Update README.md by @manukyutai in https://github.com/kyutai-labs/pocket-tts/pull/328 * README: update xn Rust port links to gradium-ai/xn-ptts by @pratikgx in https://github.com/kyutai-labs/pocket-tts/pull/330 * Fade in the first 5 ms of each chunk to remove the onset click by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/332 * Fade in the whole first frame of each chunk by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/333 * End voice prompts on a short pause by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/334 * Back to a 5 ms fade-in at chunk starts by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/335 * Drifting sampler head: training recipe and english_drifting_26-09 by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/329 * Add the new languages to the api by @gabrieldemarmiesse in https://github.com/kyutai-labs/pocket-tts/pull/336 * Add Pocket TTS Greek (Ελληνικά) to community models by @antoniosmakro in https://github.com/kyutai-labs/pocket-tts/pull/340 * Training: smaller loader defaults and an upfront manifest size check for small finetunes by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/342 * Add ToBe SAID tool information to README by @lookbe in https://github.com/kyutai-labs/pocket-tts/pull/343 * Retrained French, Italian, Dutch, German, Spanish and Portuguese models (YODAS v3 polish, full-length voice prompts) by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/344 * Add Outmake to projects using Pocket TTS by @DaneBentley in https://github.com/kyutai-labs/pocket-tts/pull/349 * Limit tests: end repeats, numbers, short texts, bursts, pacing, prompt length, cut prompts, unpunctuated text by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/351 * Add Turkish community model by @alperiox in https://github.com/kyutai-labs/pocket-tts/pull/352 * Add english_2026-09_2 (re-aligned Dora data) and make it the default by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/353 * Fix cross-page links between the CLI and API reference docs by @pratikgx in https://github.com/kyutai-labs/pocket-tts/pull/354 * Add Pocket TTS Japanese (日本語) to community models by @takuma104 in https://github.com/kyutai-labs/pocket-tts/pull/368 * Add Chinese voice option to README by @yuexiajiayan in https://github.com/kyutai-labs/pocket-tts/pull/373 * Pin predefined voices to the restored pause-ended states (#377) by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/378 * Add multilingual Pocket TTS to community models by @Thanush101 in https://github.com/kyutai-labs/pocket-tts/pull/381 * Add a rust version of the CLI based on Gradium's xn-ptts by @gabrieldemarmiesse in https://github.com/kyutai-labs/pocket-tts/pull/383 * Small models: z_dim, attention options, low-rank text table, one-step heads, decoder dims, small English models by @manu-lm in https://github.com/kyutai-labs/pocket-tts/pull/360 * Rewrite dashes and curly apostrophes the models never saw in training by @quetion667 in https://github.com/kyutai-labs/pocket-tts/pull/387 * API: serve the six small English models by @gabrieldemarmiesse-ai-agent in https://github.com/kyutai-labs/pocket-tts/pull/389 * Rust CLI: run the small English models by @gabrieldemarmiesse-ai-agent in https://github.com/kyutai-labs/pocket-tts/pull/388 ## New Contributors * @subev made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/327 * @pratikgx made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/330 * @antoniosmakro made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/340 * @lookbe made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/343 * @DaneBentley made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/349 * @alperiox made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/352 * @takuma104 made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/368 * @yuexiajiayan made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/373 * @Thanush101 made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/381 * @quetion667 made their first contribution in https://github.com/kyutai-labs/pocket-tts/pull/387 **Full Changelog**: https://github.com/kyutai-labs/pocket-tts/compare/v3.3.0...v4.0.0
Related Repositories
public-apis/public-apis★ 491,146
A collective list of free APIs
NousResearch/hermes-agent★ 255,989
The agent that grows with you
yt-dlp/yt-dlp★ 200,094
A feature-rich command-line audio/video downloader
microsoft/markitdown★ 191,498
Python tool for converting files and office documents to Markdown.
Significant-Gravitas/AutoGPT★ 187,704
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.