Instructions to use christoo23/glance-voices with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Pocket-TTS
How to use christoo23/glance-voices with Pocket-TTS:
from pocket_tts import TTSModel import scipy.io.wavfile tts_model = TTSModel.load_model("christoo23/glance-voices") 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
Glance voices
French voices for Kyutai Pocket TTS (French, 6 layers), used by the Glance assistant on macOS.
| File | Voice | Source recording |
|---|---|---|
voices/claire.safetensors |
Claire | CML-TTS fr, 12205_11650_000004-0002 |
voices/lea.safetensors |
Lรฉa | CML-TTS fr, 12977_10625_000037-0001 |
voices/margaux.safetensors |
Margaux | CML-TTS fr, 1591_1028_000108-0004 |
Each file is a Pocket TTS voice state (conditioning cache) computed with the
French model from the enhanced clip published in
kyutai/tts-voices (cml-tts/fr/).
No audio of the speakers is redistributed here.
License and attribution
Released under CC BY 4.0, like their source. Changes: the original clips were converted into Pocket TTS voice states.
- Recordings: LibriVox volunteer readers, via the CML-TTS dataset โ Oliveira et al., CML-TTS: A Multilingual Dataset for Speech Synthesis in Low-Resource Languages (2023), CC BY 4.0.
- Clips: Kyutai, kyutai/tts-voices,
cml-tts/fr, CC BY 4.0. - Model used to compute the states: Kyutai Pocket TTS (weights CC BY 4.0, code MIT).
- Downloads last month
- -