Instructions to use richardr1126/Supertonic-3-CoreAI with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Supertonic
How to use richardr1126/Supertonic-3-CoreAI with Supertonic:
from supertonic import TTS tts = TTS(auto_download=True) style = tts.get_voice_style(voice_name="M1") text = "The train delay was announced at 4:45 PM on Wed, Apr 3, 2024 due to track maintenance." wav, duration = tts.synthesize(text, voice_style=style) tts.save_audio(wav, "output.wav")
- Notebooks
- Google Colab
- Kaggle
Supertonic 3 Core AI for OpenReader
This repository contains the four Core AI inference stages converted from
Supertone/supertonic-3 revision
724fb5abbf5502583fb520898d45929e62f02c0b, along with its tokenizer,
configuration, and ten unchanged voice styles. The export fixes 512 text code points
and 512 latent frames. OpenReader downloads and verifies every file before
installing the complete model on an iPhone or iPad.
PROVENANCE.md describes the conversion, SHA256SUMS pins the files, and
LICENSE.txt carries the original BigScience Open RAIL-M license. Conversion
source and validation instructions are in OpenReader's
tools/supertonic directory.
Fixed padding has a known difference from the official variable-length pipeline. See PROVENANCE.md and validation.json for the conversion-only gate and phone scope.
Model tree for richardr1126/Supertonic-3-CoreAI
Base model
Supertone/supertonic-3