Hugging Face links for four CPU-runnable micro TTS models
vanstriendaniel · x · 2026-11-07
Daniel van Strien shares the Hugging Face links for the CPU-runnable TTS models he rounded up: phonon-31m-en (31M), Paradee-8M (7.9M, 62x real time on one thread), Leokani-tts (7M, 4.5M variant) and PiVo (1.4M with vocoder), plus a larger collection list for browsing more.
Related event: Developer Rounds Up Four Tiny CPU-Runnable TTS Models(2 posts)→
More from Multimodal
- Building a sound effects library in seconds with ElevenLabs' sound effects v3 model — austin_malerba · 2026-11-07
- A reusable 'Digital Dystopia Overload' prompt for glitchy neon AI art — LudovicCreator · 2026-11-07
- KlingAI 4.0 bodybuilding video showcase wows Reddit with realistic muscle detail — Forsaken_Stuff_Ai · 2026-11-07
- One frame to 3D: Depth Anything 3 + SAM 3.1 let you fly inside a real gallbladder surgery — MaziyarPanahi · 2026-11-07
- Procedural modeling with Gaussian Splats: stack splats like blocks in your browser — janusch_patas · 2026-11-07
- AI short film 'THE AWAKENING' stuns with cinematic scale-texture realism — tanji_ri · 2026-11-07