Multilingual quality
Supertonic-3: specs & requirements
Supertonic-3 is Supertone's 99-million-parameter model with official browser support and 31-language coverage — character-level input means no GPL phonemizer anywhere in the stack. Browser integration here is still being evaluated. Supertonic is already available in the Echo Live desktop app, where you can pair speech generation with RVC character voice conversion.
Supertonic-3 support is being wired into the tool. Its specs and hardware verdicts are below — meanwhile, the live models are ready to use, and the compatibility check already tells you whether your device could run Supertonic-3.
Specifications
| Developer | Supertone |
|---|---|
| Parameters | 99 million |
| Download size | 398 MB (fp32) |
| Output | 44.1 kHz mono |
| Voices | 10 |
| Languages | English (US), Korean, Japanese, Chinese, Spanish, French, German, Italian, Portuguese, Hindi, Arabic, Russian, Vietnamese, Indonesian, Thai, Turkish, Polish, Dutch |
| Voice cloning | No |
| Streaming | No |
| License | OpenRAIL-M (weights), MIT (code) — OpenRAIL-M permits commercial use but carries use-based restrictions that pass through to your outputs. |
| Speed | Very fast on WebGPU (vendor: up to 167× real-time on Apple M4); slower than real-time on CPU-only. |
| Runtime | WebAssembly, WebGPU-accelerated when available |
Source: Supertone/supertonic-3
Strengths
- +31 languages from one model
- +First-party web/ONNX support from the vendor
- +44.1 kHz output — highest fidelity here
Limitations
- −398 MB download
- −OpenRAIL-M restrictions require a license notice for redistribution
Supertonic-3 — questions
Is Supertonic-3 free to use?
Yes. The weights are OpenRAIL-M (weights), MIT (code) and they run on your device. OpenRAIL-M permits commercial use but carries use-based restrictions that pass through to your outputs.
What hardware does Supertonic-3 need?
It runs on any modern CPU via WebAssembly at roughly real-time speed; with WebGPU it is several times faster. 8 GB RAM recommended.
Is my text private when using Supertonic-3 here?
Yes. The model executes inside your browser tab — the text you synthesize is never transmitted to a server.
Can I use Supertonic-3 output commercially?
Review the current model, voice, and source-dataset terms before commercial use. The catalog lists OpenRAIL-M (weights), MIT (code) for this model and notes: OpenRAIL-M permits commercial use but carries use-based restrictions that pass through to your outputs.