Open-source local VoiceStudio hits 11k GitHub stars with voice cloning and dubbing in 646 languages
adnan_hashmi · x · 2026-09-05
VoiceStudio (formerly OmniVoice Studio) is a fully local, open-source voice AI desktop app positioning itself as an ElevenLabs alternative, now at 11k GitHub stars and 220k+ downloads. It offers 3-second voice cloning, voice design from a sentence, timed video dubbing, multi-voice audiobooks from scripts or EPUBs, and transcription with speaker diarization, across 646 languages with a local API—all data stays on-device. Current release v0.5.0.
More from Multimodal
- LTX2.5 runs video gen fully locally on Macs where MiniMax H3 needs 36GB RAM — cocktailpeanut · 2026-09-05
- Viggle-Animate open-weights: video character replacement from one repainted frame, no pose or mask pipeline — cocktailpeanut · 2026-09-05
- SVG Animations With Synced Sound, All From Prompts, Powered by QuiverAI's Model — stuffyokodraws · 2026-09-05
- H3 Ref2V workflow notes: 7 references break cohesion at low steps, 10 steps worked — R34vspec · 2026-09-05
- Watch GPT Sol (ultra) Sculpt a Banana in Blender, Start to Finish (32x Speed) — D3VAUX · 2026-09-05
- KREA2 struggles with two-character LoRAs; users detour via nano banana pro — breakallshittyhabits · 2026-09-05