Supertonic: open-source 99M-param local TTS beats ElevenLabs in tests
JafarNajafov · x · 2026-10-01
Supertonic is an open-source tool that generates studio-quality voiceovers locally on a laptop CPU via ONNX: 99M parameters, no GPU, no internet or API needed, outputting 44.1kHz audio in under a second.
- In real-world tests it correctly pronounced a phone number, a dollar amount, and a technical unit on the first try, while ElevenLabs, OpenAI, and Gemini TTS all failed the same test
- Supports 31 languages out of the box (English, Japanese, Hindi, Arabic, etc.), with expression tags for inline laughs, sighs, and breaths to sound more human
- Already integrated into Chrome extensions, iPhone apps, and even an e-reader in airplane mode
- MIT code, open weights, 13k GitHub stars
Even if you never ship a voiceover, having a free local TTS on your machine is just useful.
More from Infra
- Korea September exports surge on chip demand: semiconductors up 263% to record $60.3B — firstadopter · 2026-10-01
- turbopuffer rewrites storage engine v3 in public, grinding query plans live — vboykis · 2026-10-01
- BofA Raises Micron Targets on Durable AI Memory Cycle, Sees $60-100B Buyback by FY28-29 — firstadopter · 2026-10-01
- llama.cpp PR adds MTP support for Qwen Flash Next, GGUF quants now on Hugging Face — jacek2023 · 2026-10-01
- Distilled 260M text-to-image model hits <190ms on a T4, generating per keystroke — Gradio · 2026-10-01
- Agent spins up 322 Hugging Face Jobs in 90 minutes to test code, total bill ~$4 — vanstriendaniel · 2026-10-01