VoiceStudio
Self-described "open ElevenLabs alternative" — local voice studio under AGPL
A self-described "open ElevenLabs alternative": a local voice studio under the AGPL license.
Text-to-speech, speech-to-text, and voice cloning, open source.
Key features
- Voice cloning, voice design, video dubbing, dictation, transcription, audiobooks
- 646 languages supported
- Local workflows run on your hardware; optional remote workers and local API/MCP
- GPU acceleration on NVIDIA CUDA, Apple Silicon Metal, or CPU fallback
Good for
- Creators making dubbed videos and audiobooks locally
- Users dictating with a floating widget in 646 languages
- Developers wiring voice into AI agents via local API or MCP
Stack & tags
PythonNode.jsBunRustElectronvoice-cloningttsdubbingdictationaudiobooksttdesktop-appmcp
Quick start
curl -fsSL https://voicestudio.sh/install | shFrom the project README on GitHub.
Screenshots
Screenshots from the project’s GitHub README.
Similar open-source projects
Whisper
★ 109,795OpenAI's open-source speech recognition model — the benchmark for multilingual STT
GPT-SoVITS
★ 62,271TTS/cloning trainable on 1 minute of audio — a legendary Chinese-community project