Instructions to use kyutai/pocket-tts-without-voice-cloning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Pocket-TTS
How to use kyutai/pocket-tts-without-voice-cloning with Pocket-TTS:
from pocket_tts import TTSModel import scipy.io.wavfile tts_model = TTSModel.load_model("kyutai/pocket-tts-without-voice-cloning") voice_state = tts_model.get_state_for_audio_prompt( "hf://kyutai/tts-voices/alba-mackenna/casual.wav" ) audio = tts_model.generate_audio(voice_state, "Hello world, this is a test.") # Audio is a 1D torch tensor containing PCM data. scipy.io.wavfile.write("output.wav", tts_model.sample_rate, audio.numpy()) - Notebooks
- Google Colab
- Kaggle
Ctrl+K
english_{18m,7m,2m}_2026-10 and their _refl variants replaced (Dora-text fine-tunes: swear words, rare words, numbers, acronyms; REFL redone on top); new languages/english_2026-10, also the default languages/english (english_2026-09_2 recipe with the unaligned-last-word loader fix, reads sentence-final numbers); recomputed predefined voice states
a7056f5 verified