Supertonic: Open-Source Local TTS Released
JafarNajafov · x · 2026-07-14
Supertonic is a text-to-speech model that runs entirely on local devices, requiring no cloud, API keys, or per-character billing. The author states it is 100% open-source under the MIT license and has around 2700 GitHub stars.
The reported performance metrics are highly aggressive: it achieves 167x real-time speed on an M4 Pro with only 66 million parameters. In speed comparisons, Supertonic hits 1263 characters/second, compared to 287 for ElevenLabs Flash and 55 for OpenAI TTS-1. It can even run on a Raspberry Pi or offline e-readers.
Functionally, it handles currencies, dates, phone numbers, and technical units well without preprocessing. It supports 11 platforms and 5 languages, offering a Chrome extension to convert webpages to audio in under a second. The author concludes that the cloud API-dominated TTS market may soon be disrupted by such on-device models.
More from Infra
- Sail equivalence proof and riscv-coq cross-checks uncover real RISC-V spec bugs — xennygrimmato_ · 2026-10-27
- xAI's Colossus 1 hits ~14.6 exaflops FP64, ~5x Top500 leader — cis_female · 2026-10-27
- Lambda runs 19 B200 nodes in a 16-node power budget, boosting throughput 24% — TheZachMueller · 2026-10-27
- Podcast: Micron's HBM roadmap, Meta's AI infra, and NVIDIA's supply chain lock-in — BenBajarin · 2026-10-27
- SGLang MegaKernel + PD Pushes the TCO-Latency Frontier on GB300, SemiAnalysis Benchmarks Show — AccBalanced · 2026-10-27
- Scaling self-hosted Convex: compiling TypeScript Convex functions to WASM — leventov · 2026-10-27