# supertone-inc/supertonic

Lightning-Fast, On-Device, Multilingual TTS — running natively via ONNX.

Repository: https://github.com/supertone-inc/supertonic
Canonical: https://ross.abutalabs.com/products/supertonic
Homepage: https://huggingface.co/spaces/Supertone/supertonic-3
Language: Swift
License: MIT
License Family: permissive
Topics: cpp, csharp, go, ios, java, lightweight, nodejs, on-device, python, rust, swift, text-to-speech, tts, web, flutter, multilingual, onnx, onnxruntime, speech-synthesis, webgpu
Last push: 2026-07-24T04:00:17+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 94, release rhythm 33, longevity 20
- inputs: {"age_days": 288, "days_push": 40, "days_rel": 239, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 13734, forks 1520 (observed 2026-08-28T04:11:04.708823+00:00)

## What it is
Supertonic is a lightning-fast, on-device multilingual text-to-speech system powered by ONNX Runtime, with a compact 99M-parameter open-weight model supporting 31 languages. It runs locally on desktop, mobile, browsers, and edge hardware with no cloud dependency. Note: the repository is being archived with no further development or official support.

## Use cases
- convert text to speech offline on device
- run fast tts on raspberry pi or edge hardware
- synthesize speech in multiple languages locally
- add text-to-speech to a mobile app without cloud api
- generate audio from webpages in under a second
- privacy-preserving speech synthesis with no api calls
- run tts in the browser via webgpu

## When to choose
- you need low-latency, on-device tts with a small model footprint
- you want multilingual speech synthesis without cloud services
- you target resource-constrained or edge devices
- privacy rules forbid sending text to external apis

## When to avoid
- you need ongoing official support or updates - the repo is archived as of July 2026
- you need voice cloning after August 2026 when Voice Builder shuts down
- you want the highest-quality large-model tts regardless of size
- you need a maintained, actively developed solution

## Facets
- artifact type: library
- maturity: abandoned
- function: tts, speech-recognition, machine-learning, llm-inference
- domain: speech-processing, machine-learning, cross-platform
- platform: cross-platform, python, cpp, rust, go, jvm, browser, wasm
- tags: on-device-inference, onnx-runtime, speech-synthesis, voice-cloning, edge-devices, multilingual-tts, webgpu, flutter, natural-language-processing, swift, nodejs, ios, mobile, desktop, gpu

## Member repositories
- supertone-inc/supertonic (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:04.708823+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:13:04.534030+00:00, confidence not recorded.
  - readme: https://github.com/supertone-inc/supertonic (fetched 2026-08-28T04:11:04.708823+00:00, sha 0707256e0696)
  - homepage: https://huggingface.co/spaces/Supertone/supertonic-3 (fetched 2026-08-29T08:07:21.826059+00:00, sha b41ccc6beba7)
- Data as of 2026-08-30T08:39:29.467469+00:00.