# Voicebox

> Voicebox (v0.5.0 release April 2026, changelog to v0.6.0 in Oct 2026, ~59k stars, last commit 7 Oct 2026) is a Tauri desktop app with a Python backend that runs on your machine. It clones voices from a short sample, generates speech in 23 languages with engines such as Qwen3-TTS, Chatterbox and Kokoro, and has a multi-track Stories editor, audio effects, Whisper transcription, a dictation hotkey, a REST API and an MCP server so agents can speak in your voices. Installers for macOS and Windows, Docker, Linux from source. MIT.

- Repository: https://github.com/jamiepine/voicebox
- Official docs: https://docs.voicebox.sh
- Checked: 2026-10-29

## What it replaces
- ElevenLabs: Speech generation, Voice cloning, Speech recognition (partial). Covers local speech generation, voice cloning and Whisper transcription; no hosted conversational voice agent platform, only an MCP tool that lets coding agents speak.
  - Use it instead of paying for ElevenLabs? Personal use: Yes. Business: Partly (Commercial use with conditions: MIT code allows commercial use; output from the OmniVoice engine must not be used commerc…; Model weights: Each TTS engine downloads its own weights under its own licence (Kokoro and Qwen3-TTS Voi…).
- Descript: Voice cloning, Speech generation (partial). Voice cloning and text-to-speech with a multi-track timeline for audio; no transcript-based audio or video editing.
  - Use it instead of paying for Descript? Personal use: Yes. Business: Partly (Commercial use with conditions: MIT code allows commercial use; output from the OmniVoice engine must not be used commerc…; Model weights: Each TTS engine downloads its own weights under its own licence (Kokoro and Qwen3-TTS Voi…).

## Facts, with sources

| Question | Status | Answer | Source | Checked |
|---|---|---|---|---|
| Runs on my machine? | Confirmed | Runs entirely on your machine; an optional remote mode runs the backend on another machine such as a GPU server | https://github.com/jamiepine/voicebox#what-is-voicebox | 2026-10-29 |
| Code licence | Confirmed | MIT | https://github.com/jamiepine/voicebox/blob/main/LICENSE | 2026-10-29 |
| Model weights licence | Partial | Each TTS engine downloads its own weights under its own licence (Kokoro and Qwen3-TTS VoiceDesign are Apache-2.0); the opt-in OmniVoice engine has CC-BY-NC weights and is labelled non-commercial | https://github.com/jamiepine/voicebox/blob/main/CHANGELOG.md | 2026-10-29 |
| Commercial use | Partial | MIT code allows commercial use; output from the OmniVoice engine must not be used commercially, and other engines follow their own weight licences | https://github.com/jamiepine/voicebox/blob/main/CHANGELOG.md | 2026-10-29 |
| Paid services needed? | Confirmed | None; speech generation, transcription and the bundled Qwen3 LLM all run locally | https://github.com/jamiepine/voicebox#what-is-voicebox | 2026-10-29 |
| Hardware | Confirmed | Minimum 8 GB RAM, 5 GB free disk and a modern multi-core CPU; 16 GB+ RAM and a CUDA GPU recommended. Also uses Apple Silicon (MLX), ROCm, DirectML and Intel Arc | https://github.com/jamiepine/voicebox/blob/main/docs/content/docs/overview/installation.mdx | 2026-10-29 |
| Operating systems | Confirmed | macOS 11+ (Apple Silicon and Intel) and Windows 10+ installers; Linux builds from source (no prebuilt binaries yet); Docker image | https://github.com/jamiepine/voicebox#download | 2026-10-29 |
| Setup effort | Confirmed | Easy on macOS and Windows; download the DMG or MSI and the first generation downloads its model (about 350 MB to 8 GB). Linux needs a source build | https://github.com/jamiepine/voicebox/blob/main/docs/content/docs/overview/installation.mdx | 2026-10-29 |

GitHub: 59296 stars, 7432 forks, last push 2026-10-07, archived: no.

## Alternatives to Voicebox
- [VoiceStudio](https://repoquarry.dev/projects/voicestudio.md): Local desktop app for voice cloning, voice design, video dubbing, dictation, transcription and audiobooks, built on OmniVoice and other engines.
- [GPT-SoVITS](https://repoquarry.dev/projects/gpt-sovits.md): Few-shot voice cloning and TTS WebUI; 5-second zero-shot or 1-minute fine-tune, five languages, MIT code and weights.
- [Fish Speech](https://repoquarry.dev/projects/fish-speech.md): Fish Audio's multilingual TTS and voice cloning stack (Drama 3, S2-Pro) under a source-available community licence.
- [F5-TTS](https://repoquarry.dev/projects/f5-tts.md): Flow-matching text-to-speech research model that clones a voice zero-shot from a reference audio clip.
- [Chatterbox](https://repoquarry.dev/projects/chatterbox.md): Resemble AI's family of open TTS models (Turbo, Nano, Multilingual) with zero-shot voice cloning from a reference clip.
- [Coqui TTS](https://repoquarry.dev/projects/coqui-tts.md): Python TTS library with many models (XTTS-v2, VITS, YourTTS, Bark, Tortoise); original repo unmaintained, continued by idiap fork.
---
Source: RepoQuarry (https://repoquarry.dev/projects/voicebox). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.