Verified project record

debpalash/VoiceStudio

A local voice-workflow application with cloning, design, video dubbing, transcription and in-app model management, plus API/MCP interfaces. Remote services and telemetry are optional; the source advertises broad language coverage to evaluate with the chosen engine.

Project overview

Combines voice workflows on user-owned hardware with model management and local API/MCP integration; optional remote paths have separate data handling.

Project type: MCP · Video · Audio & Speech
Use cases: Meeting Notes · Translation & Subtitles · Audio & Speech
Deployment: Refer to project documentation
License: AGPL-3.0

Best for

Creators and developers who want voice cloning, dubbing, transcription and audiobook generation running locally on their own hardware instead of a closed paid cloud service.
Users working across many languages who need voice workflows in up to 646 languages.
Agent builders who want to connect voice workflows via a local API and MCP server.

Key capabilities

Clone a voice from a clean reference recording and generate speech from entered text.
Design a custom voice by describing it in the voice design workspace.
Dub videos with timed speech in many languages.
Speech-to-text transcription of audio.
Expose voice workflows through a local API and MCP server for agents to connect.
Install and manage local speech models within the app; the default engine is k2-fsa/OmniVoice with selectable alternative engines.

Limitations and risks

Electron is the only desktop app and web UI; the Tauri shell was removed after version 0.5.3, so there is no separate web UI.
Voice cloning requires permission, and the models carry their own licenses that must be reviewed before commercial use.

Getting started

Use the documented install command or a platform installer, open the app and the voice-cloning workspace, select a voice or clean reference recording, enter text and generate, installing the required model when prompted. Explicit coding and cost policies remain unspecified.

Evidence and sources

README: Open-source voice cloning, voice design, video dubbing, dictation, transcription & audiobook creation in 646 languages.
README: | Clone a voice or design your own | Dub videos with timed speech | Local API & MCP for agents |
README: ![A tour of the Electron app: voice cloning, voice design, dubbing, and model management](docs/media/electron/voicestudio.gif)
README: Local workflows run on your hardware. Remote services are optional; usage analytics requires consent.
README: | No dedicated GPU (Intel/AMD integrated graphics, older PCs) | Fully usable on the CPU, slower; setup installs the small CPU build of PyTorch (about 5 GB free disk) |

Source updated at: 2026-10-03T18:28:00Z

Open original GitHub project