supertonic
Librarysupertone-inc/supertonic
Lightning-fast, on-device multilingual TTS using ONNX.
Overview
Supertonic is a high-performance text-to-speech engine that runs entirely on-device using ONNX Runtime. It supports 31 languages with low latency and a small footprint, suitable for edge deployment.
README Preview
# Supertonic — Lightning Fast, On-Device, Accurate TTS\n\n[](https://huggingface.co/spaces/Supertone/supertonic-3)\n[](https://huggingface.co/Supertone/supertonic-3)\n[](https://github.com/supertone-inc/supertonic/tree/release/supertonic-2)\n[-Demo-lightgrey)](https://huggingface.co/spaces/Supertone/supertonic#interactive-demo)\n[-Models-lightgrey)](https://huggingface.co/Supertone/supertonic)\n\n\n \n\n\n**Supertonic** is a lightning-fast, on-device text-to-speech system designed for local inference with minimal overhead. Powered by ONNX Runtime, it runs entirely on your device—no cloud, no API calls, no privacy concerns.\n\n### 📰 Update News\n\n- **2026.04.29** - 🎉 **Supertonic 3** released with **31-language support**, improved reading accuracy, fewer repeat/skip failures, and v2-compatible public ONNX assets. [Demo](https://huggingface.co/spaces/Supertone/supertonic-3) | [Models](https://huggingface.co/Supertone/supertonic-3)\n- **2026.01.22** - **[Voice Builder](https://supertonic.supertone.ai/voice_builder)** is now live! Turn your voice into a deployable, edge-native TTS with permanent ownership.\n- **2026.01.06** - 🎉 **Supertonic 2** released with 5-language support. The v2 code path is preserved on the [`release/supertonic-2`](https://github.com/supertone-inc/supertonic/tree/release/supertonic-2) branch.\n- **2025.12.10** - Added `supertonic` PyPI package! Install via `pip install supertonic`. For details, visit [supertonic-py documentation](https://supertone-inc.github.io/supertonic-py)\n- **2025.12.10** - Added [6 new voice styles](https://huggingface.co/Supertone/supertonic/tree/b10dbaf18b316159be75b34d24f740008fddd381) (M3, M4, M5, F3, F4, F5). See [Voices](https://supertone-inc.github.io/supertonic-py/voices/) for details\n- **2025.12.08** - Optimized ONNX models via [OnnxSlim](https://github.com/inisis/OnnxSlim) now available on [Hugging Face Models](https://huggingface.co/Supertone/supertonic)\n- **2025.11.24** - Added Flutter SDK support with ma
FAQ (3)
TroubleshootingWhy does Supertonic TTS produce incomplete audio with missing or repeated words?
Upgrade to Supertonic 3, which significantly reduces incomplete audio, skipped words, and repeated fragments. For earlier versions, regenerating the audio may temporarily fix the issue. On resource-constrained devices like Raspberry Pi, consider using a lower quality setting to reduce generation time and check audio for abnormalities.
TroubleshootingWhy does Supertonic TTS mispronounce acronyms like 'SDKs' or skip words?
Earlier Supertonic releases sometimes misread acronyms, skipped words, or produced unstable output. Upgrade to Supertonic 3, which focused on improving reading stability and reducing skipped words. Try it at: https://huggingface.co/spaces/Supertone/supertonic-3
TroubleshootingWhy does Supertonic TTS skip or drop words when generating audio from long sentences?
Upgrade to Supertonic 3, which improves reading accuracy and reduces skipping. If that’s not possible, split input text at punctuation like em-dashes and semicolons before feeding it to the model—this prevents overly long inputs that cause the model to underestimate speech duration and drop words. For multiple output candidates, generate only one (zero-shot) and use the first file, as later iterations are more prone to glitches.