Repolode
Submit
debpalash

debpalash/VoiceStudio

Open-source, local-first voice studio for voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation in 646 languages.

Open-source alternative to:ElevenLabs

View on GitHubWebsite
Stars
76,537
Forks
8,363
Open issues
262
Last commit
21 hours ago

What is it?

VoiceStudio is a desktop application for generating and editing speech entirely on your own hardware. It covers voice cloning from a reference recording, voice design from a text description, dubbing of video with timed speech, floating-widget dictation, transcription, and audiobook or batch narration. The default engine is OmniVoice, and other speech engines can be chosen instead.

The app is built on Electron, with a Python backend that runs the models. It supports NVIDIA CUDA on Windows and Linux, Apple Silicon via Metal (MPS), and CPU-only operation on machines without a dedicated GPU, with slower performance. It can be installed with a shell or PowerShell script, from release packages, or via Docker, and it exposes a local API and MCP server so scripts and coding agents can drive it.

It is positioned as a local alternative to hosted voice services such as ElevenLabs. Model weights and generated-output conditions are governed separately from the AGPL-3.0 application license, and cloning voices is meant to be done only with permission.

Key features

  • 01Clones a voice from a clean reference recording.
  • 02Designs new voices from a text description.
  • 03Dubs videos with timed speech output.
  • 04Provides floating-widget dictation and transcription.
  • 05Creates audiobooks and runs batch narration jobs.
  • 06Exposes a local API and MCP server for scripts and agents.

Use cases

  • Dubbing a video into another language with timed speech.
  • Producing audiobook narration from long-form text.
  • Dictating notes or text by voice on your own machine.
  • Letting a coding agent generate speech through the local API or MCP.

Topics

#text-to-speech#voice-cloning#speech-to-text#dubbing#local-first#audiobook#transcription#ai#cuda#elevenlabs-alternative

FAQ

What is VoiceStudio?
Open-source, local-first voice studio for voice cloning, voice design, video dubbing, dictation, transcription and audiobook creation in 646 languages. VoiceStudio is a desktop application for generating and editing speech entirely on your own hardware. It covers voice cloning from a reference recording, voice design from a text description, dubbing of video with timed speech, floating-widget dictation, transcription, and audiobook or batch narration. The default engine is OmniVoice, and other speech engines can be chosen instead.
Is VoiceStudio free to use? What license does it have?
VoiceStudio is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
Is VoiceStudio still maintained?
Actively maintained. The last commit to VoiceStudio was on Oct 19, 2026.
What can I use VoiceStudio for?
Dubbing a video into another language with timed speech. Producing audiobook narration from long-form text. Dictating notes or text by voice on your own machine. Letting a coding agent generate speech through the local API or MCP.
Where can I get VoiceStudio?
The source code and installation instructions are on GitHub: https://github.com/debpalash/VoiceStudio

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects