Microsoft reopens VibeVoice: 10-second voice cloning, 90-minute local TTS
solyarisoftware · x · 2026-11-04
Microsoft re-released VibeVoice as open source after pulling it over misuse concerns. The model clones any voice from 10 seconds of audio, generates up to 90 minutes of multi-speaker (up to 4) conversation in one pass, supports real-time streaming input, and runs fully locally — no per-minute pricing. Technically it uses ultra-low-frame-rate (7.5 Hz) continuous acoustic/semantic tokenizers plus a next-token diffusion framework (LLM for context, diffusion head for audio detail) for speaker consistency and natural turn-taking. Microsoft disabled the repo in Sept 2025 after out-of-scope uses; it's now back with watermarks and safety controls. Demos show spontaneous emotion, singing, podcast with background music, and Mandarin-English cross-lingual speech.
More from Multimodal
- Seedance 2.5 AI video of a wild Malfoy is turning heads — nikola_mr64990 · 2026-11-04
- Free course covers the 4 hard problems of production generative media agents — Al_Grigor · 2026-11-04
- PixelBench: testing multimodal models pixel-by-pixel instead of with QA — bdsqlsz · 2026-11-04
- Kitten-TTS 2 offers free voice cloning on laptops, no GPU needed — solyarisoftware · 2026-11-04
- "We'll Take Your RAM": a fully local-hardware cyberpunk meme from 2030 San Francisco — Wonderful-Yak-3348 · 2026-11-04
- Openfilm: open-source AI video agent turns coding agents into video makers — solyarisoftware · 2026-11-04