Supertonic 3 — verbatim mirror of Supertone/supertonic-3

This repository is an unmodified mirror. Every file under onnx/, voice_styles/, audio_samples/ and img/, plus config.json and LICENSE, is byte-for-byte identical to Supertone/supertonic-3 at upstream commit 3cadd1ee6394adea1bd021217a0e650ede09a323 (upstream last modified 2026-05-18). Each file was checked against the upstream tree's SHA-256 digests before upload. The only additions are this README and README.upstream.md, the original model card exactly as Supertone published it.

Why a mirror exists

Supertone Inc. (Seoul) resolved to dissolve at an extraordinary shareholders' meeting on 2026-07-15. Its hosted services — Supertone Play, the Supertone API and the Supertonic Voice Builder — were discontinued on 2026-08-31, and the upstream GitHub repository announced on 2026-07-23 that it will be archived with no further development or official support. Applications that download the open weights straight from the Supertone organisation have no guarantee that the organisation, or its files, stay online once liquidation completes. This mirror is maintained by the Sokuji project so its users keep a stable download source.

Nothing here adds a voice-cloning capability. Custom voice_style.json files were only ever produced by Supertone's Voice Builder, which no longer exists; files you already downloaded from it keep working with these weights.

License

The weights are licensed under the BigScience Open RAIL-M License (August 18, 2022), reproduced verbatim in LICENSE. Redistribution here is permitted by that license on the condition that the same license and its use-based restrictions (Attachment A) travel with the files — they do. In particular the license forbids using the model or its derivatives to impersonate others without their consent, to generate content without disclosing that it is machine-generated, or in any way that violates applicable law.

Copyright (c) 2026 Supertone Inc. "Supertonic" and "Supertone" are names of Supertone Inc.; this mirror is not affiliated with or endorsed by Supertone Inc. or HYBE.

---

The upstream model card follows unchanged.

Supertonic 3 | Lightning Fast, On-Device, Accurate TTS

Supertonic is a lightweight text-to-speech system for local inference. It runs with ONNX Runtime entirely on your device, with no cloud call required for synthesis.

Supertonic 3 expands the open-weight release from 5 to 31 languages, improves reading stability, and reduces repeat/skip failures.

Quick Start

Install the Python SDK and generate speech immediately. On first run, the SDK downloads the model assets from Hugging Face.

pip install supertonic

from supertonic import TTS

tts = TTS(auto_download=True)
style = tts.get_voice_style(voice_name="M1")

text = "A gentle breeze moved through the open window while everyone listened to the story."
wav, duration = tts.synthesize(text, voice_style=style, lang="en")

tts.save_audio(wav, "output.wav")
print(f"Generated {duration:.2f}s of audio")

What's New in Supertonic 3

• 31 languages: expanded from the 5-language Supertonic 2 release.
• More stable reading: fewer repeat and skip failures, especially on short and long utterances.
• Higher speaker similarity: improved similarity across the shared-language set compared with Supertonic 2.
• Expression tags: supports simple tags such as <laugh>, <breath>, and <sigh>.

Custom Voices and Audio Samples

The open-weight package includes fixed preset voice styles for immediate local inference. If you want to hear how Supertonic 3 performs with zero-shot custom voice styles, visit the Audio Sample Demo to compare reference audio and generated speech across several use cases. To create your own Supertonic 3 voice-style JSON from reference audio, use Supertonic Voice Builder; purchased Voice Builder styles include downloadable embeddings for both Supertonic 2 and Supertonic 3.

Here are a few reference/generated pairs from the audio sample demo:

Call center, English
Text: Good morning, thank you for calling. How can I help you today?

Reference voice	Supertonic 3 output

Character voice, Japanese
Text: ふふっ、退屈してたところなの。ちょうどいい遊び相手、見つけたかも♪

Reference voice	Supertonic 3 output

Elder character voice, Korean
Text: 혼자 떠나기엔 길이 험하구나. 이 낡은 검을 가져가거라. 언젠가 어둠이 네 이름을 부르더라도, 부디 빛을 잊지 말거라.

Reference voice	Supertonic 3 output

Audiobook, English
Text: I was not afraid of silence. I had lived with it long enough to know that, sometimes, it speaks more honestly than people do.

Reference voice	Supertonic 3 output

Audiobook, Japanese
Text: その朝、ロンドンの霧はいつになく低く垂れこめていた。私はただの訪問者だと思っていたが、ホームズの目はすでに別の結論にたどり着いていた。

Reference voice	Supertonic 3 output

News, English
Text: Here’s a story worth paying attention to. Supertone has released Supertonic 3, its on-device TTS model. This version expands support to thirty-one languages and improves reading stability.

Reference voice	Supertonic 3 output

Performance Highlights

Supertonic 3 is designed for practical on-device inference: compact enough to run locally, while staying competitive with much larger open TTS systems.

Reading Accuracy

Across measured languages, Supertonic 3 stays within a competitive WER/CER range against much larger open TTS models such as VoxCPM2, while preserving a lightweight on-device deployment path. Asterisked languages use CER; the others use WER.

Supertonic 2 to Supertonic 3

Compared with Supertonic 2, Supertonic 3 reduces repeat and skip failures, improves speaker similarity across the shared-language set, and expands language coverage from 5 to 31 languages.

Runtime Footprint

Supertonic 3 runs fast on CPU, even compared with larger baselines measured on A100 GPU, and uses substantially less memory. It does not require a GPU, which makes local, browser, and edge deployment much easier.

Model Size

At about 99M parameters across the public ONNX assets, Supertonic 3 is much smaller than 0.7B to 2B class open TTS systems. The smaller model size is a practical advantage for download size, startup time, and on-device inference.

Supported Languages

Code	Language	Code	Language	Code	Language	Code	Language
en	English	ko	Korean	ja	Japanese	ar	Arabic
bg	Bulgarian	cs	Czech	da	Danish	de	German
el	Greek	es	Spanish	et	Estonian	fi	Finnish
fr	French	hi	Hindi	hr	Croatian	hu	Hungarian
id	Indonesian	it	Italian	lt	Lithuanian	lv	Latvian
nl	Dutch	pl	Polish	pt	Portuguese	ro	Romanian
ru	Russian	sk	Slovak	sl	Slovenian	sv	Swedish
tr	Turkish	uk	Ukrainian	vi	Vietnamese

License

This project's sample code is released under the MIT License. See the GitHub repository for details.

The accompanying model is released under the OpenRAIL-M License. See the LICENSE file in this repository for details.

This model was trained using PyTorch, which is licensed under the BSD 3-Clause License but is not redistributed with this project. See the PyTorch license for details.

Copyright (c) 2026 Supertone Inc.