Voicebox

What it is

Voicebox is a free, open-source, local-first voice cloning studio that runs entirely on your own machine. It clones a voice from a few seconds of audio and generates speech with five different TTS engines, a multi-voice timeline editor, and no cloud dependency at all.

Why it matters

ElevenLabs charges by character and processes your audio on their servers. Voicebox runs locally, so there is no per-character cost, no monthly plan, and no audio leaving your device. For podcasters, audiobook producers and anyone working with sensitive material, that changes what is possible.

What you get

• Voice cloning from roughly three seconds of reference audio.
• Five TTS engines: Qwen3-TTS, LuxTTS, Chatterbox Multilingual, Chatterbox Turbo and TADA.
• 23 languages covered.
• A multi-voice timeline editor for podcasts, dialogues and audiobooks.
• Unlimited length through automatic chunking with crossfade.
• GPU acceleration (CUDA) plus Apple Silicon support via MLX.
• A REST API and an MCP integration, so an AI agent can call voicebox.speak.

Where it falls short

Local models mean local hardware: a capable GPU makes a large difference, and CPU-only mode is slow. First-time setup also involves downloading several gigabytes of models, and voice cloning carries obvious ethical and legal obligations around consent.

Read the full review

We wrote a longer piece covering what Voicebox can do, the engine comparison, and where it fits in a local AI toolkit: Stop Paying for AI Voiceovers! This Free App Runs Entirely on Your PC.