Threads @soundslikeclickclack — Dc8aGPsIHh4
VoiceStudio is a fully local, open-source ElevenLabs alternative that just crossed 19k GitHub stars in days. One desktop app on Mac, Windows or Linux handles voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation across 646 languages, no account, no API key, no per-minute meter. It switches between 16 TTS engines and 11 speech-recognition engines under the hood, with vocal isolation and speaker ID baked in. Everything runs on your own hardware. Default engine is OmniVoice (built on k2-fsa) for TTS and WhisperX for transcription, but you can swap in 16 TTS and 11 ASR engines total. Demucs handles vocal isolation, Pyannote does speaker ID, and AudioSeal watermarks generated audio. Runs on CUDA, Apple Silicon, ROCm or plain CPU. It’s free and AGPL-3.0 licensed on GitHub (debpalash/VoiceStudio), formerly called OmniVoice Studio. One-command install scripts for Mac, Windows and Linux (including ARM64), or try the Google Colab notebook first if you don’t want to install anything locally. GitHub - debpalash/VoiceStudio: VoiceStudio is the open-source, fully-local ElevenLabs alternative — voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.