Views
No views yet
1pip install mlx soundfile numpy
2pip install "nanocodec-mlx @ git+https://github.com/nineninesix-ai/nanocodec-mlx.git"1from tts_mlx.inference import synthesize
2
3synthesize(
4 checkpoint=".", # path to this repo
5 text="Здравейте, аз съм българска система за синтез на реч.",
6 output="output.wav",
7 speaker_id=0, # 0 = AI voice, 1 = audiobook narrator
8 temperature=0.25,
9 top_k=50,
10 top_p=0.8,
11)| Speaker | Description | Best text length |
|---|---|---|
| 0 | AI-generated, clear & fast | Any (20–500+ chars) |
| 1 | Real female, audiobook narrator | 250–320 chars |