tiny-voca specs: 4.5M params, 21MB ONNX, Apache 2.0 licensed
thisdudelikesAI · x · 2026-11-17
Full specs of tiny-voca: 4,537,291 parameters including the waveform generator; 18MB model, 21MB ONNX file; two fixed American English voices (F2 and M2); 24kHz mono output; trained from scratch; Apache 2.0 for both code and weights.
Related event: tiny-voca limitations: English-only, two voices, no streaming yet(2 posts)→
More from Models
- Haru 4B: open-weight Apache 2.0 full-duplex speech model that listens and talks simultaneously — thisguyknowsai · 2026-11-17
- Hermes App Screenshot Exposes Meta's Unreleased muse-wonder-1.0 Model with 1M Context — Teknium · 2026-11-17
- Agent token usage surges but price war leaves model vendors' revenue flat — FinanceYF5 · 2026-11-17
- Claude secretly banned 'blockchain' and every word needed to explain the project — RileyRalmuto · 2026-11-17
- 18MB browser-run TTS model tiny-voca makes paid English TTS look silly — thisdudelikesAI · 2026-11-17
- After MazeBench gets benchmaxxed, developers turn to secretive AI benchmarks — amplifiedamp · 2026-11-17