# Voicebox

> A local, open source voice studio: cloning, TTS and dictation

- Category: Image, video & voice (`media-ai`)
- Tags: `open-source`, `free`, `mcp`
- Official: https://github.com/jamiepine/voicebox
- Wunderlust: https://wunderlust.xyz/r/voicebox
- Image: https://wunderlust.xyz/media/voicebox-poster.jpg
- Video preview: https://wunderlust.xyz/media/voicebox.webm
- Added: 2026-11-05

## What it is

A voice studio that runs on your machine to clone voices, generate speech and dictate text, positioned as a free alternative to ElevenLabs and WisprFlow.

## What it’s for

Use it for narration, voice cloning and dictation without sending audio to the cloud.

## Benefits

- Clones voices from short samples, with 8 TTS engines and 23 languages
- Dictation via a global shortcut with Whisper, and a multi-track story editor
- A REST API and MCP server so agents can speak in a cloned voice

## Caveats

- MIT license, no paid plans
- On Linux you have to build from source; there are no binaries yet
- Likely match, not yet confirmed by the author's links

## Similar on Wunderlust

- [Pocket TTS](https://wunderlust.xyz/r/pocket-tts.md): Kyutai's voice model that synthesizes and clones voice on CPU
- [VoiceStudio](https://wunderlust.xyz/r/voicestudio.md): Voice, cloning, transcription and dubbing
- [ElevenLabs](https://wunderlust.xyz/r/elevenlabs.md): AI voice and audio platform: speech, cloning, dubbing and agents
- [OpenCut](https://wunderlust.xyz/r/opencut.md): Privacy-focused open source video editor, available in the browser and on GitHub
- [Obsidian Local REST API](https://wunderlust.xyz/r/obsidian-local-rest-api.md): An Obsidian plugin that exposes the vault through a local API and MCP server
- [Open Notebook](https://wunderlust.xyz/r/open-notebook.md): An open source, self-hosted alternative to NotebookLM