# VoiceStudio

> VoiceStudio is an open-source desktop studio and local API for voice cloning, dubbing, transcription, and text-to-speech. It runs on Windows, macOS, and Linux and supports 14 speech engines and 646 languages.

VoiceStudio was previously named OmniVoice Studio. The open-source edition is free. Pro is a separate, closed-source build with extra features (more devices, remote devices, early access, and more on the way), sold as a paid per-user Commercial License for client work. VoiceStudio Cloud is a hosted service in early access.

Identity note: "VoiceStudio" and "OmniVoice Studio" refer to the same project, not separate products. VoiceStudio is the current product name. OmniVoice may also appear as an engine name and in older installer filenames.

## Primary links

- [Homepage](https://voicestudio.sh/): Product overview and feature demos
- [Name change](https://voicestudio.sh/omnivoice-studio): Former-name history and product identity
- [Download](https://voicestudio.sh/download): Desktop installers and Docker
- [Documentation](https://voicestudio.sh/docs): Setup, workflows, and API guides
- [Local API reference](https://voicestudio.sh/docs/api-reference): Task-oriented index for the open-source desktop application's local endpoints
- [Cloud API preview](https://voicestudio.sh/docs/cloud-api): The unreleased public Artifact and Job contract
- [Voices and cloning](https://voicestudio.sh/docs/voices): Voice profiles and local cloning
- [Dubbing](https://voicestudio.sh/docs/dubbing): Multilingual video dubbing workflow
- [Pro licence](https://voicestudio.sh/pro): Commercial License plans, per user
- [Why professionals get Pro](https://voicestudio.sh/blog/why-voicestudio-pro): what a seat adds and what it does not
- [Cloud early access](https://voicestudio.sh/cloud): Hosted API waitlist and free usage credits
- [Partners](https://voicestudio.sh/partners): Partner gallery and application to integrate, sponsor or resell
- [Source code](https://github.com/debpalash/VoiceStudio): AGPL-3.0 repository

## Product facts

- Local-first: audio, transcripts, and voiceprints stay on the user's machine by default.
- Cross-platform: Windows, macOS, Linux, and Docker.
- API: OpenAI-compatible speech and transcription routes plus VoiceStudio cloning, dubbing, model, project, and system endpoints.
- Licensing: GNU AGPL-3.0 for open-source use. Pro and Lifetime are paid per-user Commercial Licenses for the closed-source Pro build; closed-source distribution and private hosting are an Enterprise enquiry.
- Hosted service: VoiceStudio Cloud is coming soon; it does not replace the desktop app.

## More detail

- [Complete LLM reference](https://voicestudio.sh/llms-full.txt)
- [Local OpenAPI document](https://voicestudio.sh/voicestudio-openapi.json)
- [Cloud public OpenAPI preview](https://voicestudio.sh/cloud-openapi.json)
- [MCP server](https://voicestudio.sh/docs/mcp): connect Cursor, Claude Code or Codex to your local VoiceStudio to speak, clone and transcribe
- [Docs as Markdown](https://voicestudio.sh/docs-md/index.md): every docs page is also plain Markdown at /docs-md/<page>.md
- [Model library](https://voicestudio.sh/library): every speech and transcription model VoiceStudio runs, with languages, capabilities, engines (llama.cpp, audio.cpp, MLX, ONNX, PyTorch) and supported hardware
- [License summary](https://voicestudio.sh/license)
- [Privacy](https://voicestudio.sh/privacy)
- [Terms](https://voicestudio.sh/terms)
- [Refunds and cancellation](https://voicestudio.sh/refunds)
- [Acceptable use](https://voicestudio.sh/acceptable-use)
- [Data policy](https://voicestudio.sh/data-policy)
- [FAQ](https://voicestudio.sh/faq)
- [Contact](https://voicestudio.sh/contact)