# Voicebox

> Local-first OSS voice studio — clone, TTS, dictate, MCP voice for agents

**Category:** Libraries  
**Pricing:** open-source  
**URL:** https://voicebox.sh  
**GitHub:** https://github.com/jamiepine/voicebox  
**Added:** 2026-09-17

---

## The Hook

Voicebox is the open-source ElevenLabs + WisprFlow combo that keeps models and captures on your machine.

## What's Good

MIT (jamiepine/voicebox). Desktop voice studio: zero-shot cloning, 7 TTS engines (Qwen3-TTS, Chatterbox, Kokoro, and more), 23 languages, global hotkey dictation into any field, effects, and MCP-aware agent voice hooks. Docs at docs.voicebox.sh. GH trending signal with ★~55k — privacy-first local voice I/O for builders, not another cloud TTS signup.

## The Catch

Local GPU/CPU and model downloads are on you; macOS/Windows desktop footprint is heavier than a REST TTS call. Large open-issue count — treat cutting-edge engines as experimental. Vs ElevenLabs-class cloud TTS (paid APIs) and Whisper-only STT stacks: Voicebox owns both ends of the loop locally, with agent voice as a first-class surface.

## Verdict

MIT local voice studio. Install when agents and humans need on-device TTS/STT without a cloud voice vendor.