Verified project record

jamiepine/voicebox

VoiceBox is a local-first, open-source voice I/O application that combines voice cloning, multi-engine text-to-speech, global dictation, and AI agent output. It runs entirely on your machine, ensuring voice data and captures are never transmitted to the cloud.

Project overview

Project type: MCP · Model Runtime · Audio & Speech
Use cases: Meeting Notes · Audio & Speech · Automation
Deployment: Refer to project documentation
License: MIT

Best for

Developers, creators, and AI engineers who need a completely local voice I/O stack without paid cloud dependencies.
Users on macOS who need accessibility-verified voice dictation and auto-paste capabilities.

Key capabilities

Zero-shot cloning from a reference sample, or 50+ curated preset voices via Kokoro and Qwen CustomVoice.
Generate speech across 7 TTS engines and 23 languages, switchable per-generation.
Voice input via a global hotkey with push-to-talk and toggle modes, plus accessibility-verified auto-paste on macOS.
One tool call (voicebox.speak) and any MCP-aware agent can speak to the user in a cloned voice.
Attach a free-form persona to any voice profile, then Compose, Rewrite, or Respond via a bundled local LLM.
Runs OpenAI Whisper for transcription backing dictation, the Captures tab, and the /transcribe API.
Multi-track timeline for conversations, podcasts, and narratives with drag-and-drop.
8 audio effects including pitch shift, reverb, delay, chorus, and compression powered by pedalboard.

Limitations and risks

Auto-paste for dictation is currently verified and available only on macOS.
Pre-built binaries are not yet available for Linux, requiring users to build from source or use Docker.
Running multiple local models may cause GPU contention or require significant memory.

Getting started

Rated as easy to set up due to pre-built binaries available for macOS and Windows. Download the installer for your OS, install, and open the application.

Evidence and sources

GitHub project description: The open-source AI voice studio. Clone, dictate, create.
README: Native performance — built with Tauri (Rust), not Electron
README: Voicebox is a **local-first AI voice studio** — a free and open-source alternative to **ElevenLabs** and **WisprFlow** in one app. Clone voices from a few seconds of audio, genera…
Release: v0.5.0
README: Windows / Linux auto-paste | Dictation paste parity — `SendInput` on Windows, `uinput` / AT-SPI on Linux

Reviewed at: 2026-07-17T07:38:42Z