tiny-voca: 4.5M-param TTS trained on synthetic speech, 70x realtime on CPU
aigclink · x · 2026-11-16
Developer @n4ze3m trained tiny-voca, a 4.5M-parameter TTS model built entirely on synthetic data: 17.6 hours / 16,018 English utterances generated by Qwen3-TTS. It runs on a regular CPU at 70x realtime (4 threads), with total GPU training cost of $9.56. Instead of a neural vocoder it uses a Source-Filter architecture — predicting pitch, energy, and voicing, then synthesizing waveforms via a source signal and dynamic filter — which is what keeps the model so small. A custom phoneme converter handles context-aware pronunciation of numbers, dates, and amounts. English-only for now, no voice cloning or streaming, but it demonstrates distilling lightweight on-device models from high-quality synthetic speech.
Related event: tiny-voca: 4.5M-Parameter TTS Trained Fully on Synthetic Speech(4 posts)→
More from Multimodal
- LTX 2.5 keeps crashing on AMD 9070 XT with driver timeouts while MiniMax H3 runs fine — ZestycloseReading504 · 2026-11-22
- Stripe residency dev draws all friends in JavaScript with GPT 6.2 Astra, results stun — gabrielchua · 2026-11-22
- Fable 5.2 plus WebGL demoed as a tool to "create anything" — hewarsaber · 2026-11-22
- Cinematic UFO folklore music video built with Suno and SLOPCANNON — DarkroomSignal · 2026-11-22
- Photographer crafts AI-generated digital muse Nous Girl on faux magazine cover — alexcovo_eth · 2026-11-22
- Turn any Google Maps location into a cinematic drone shot with AI — aziz4ai · 2026-11-22