Voicebox is a generative AI model designed for speech synthesis, editing, and transformation. Unlike traditional text-to-speech tools that are trained for specific tasks, Voicebox can perform multiple speech-related functions using a single model.
It learns from raw audio and text, allowing it to generalize across tasks without needing retraining for each one.
Download Voicebox v0.5.0 - Software Mirrors |
|---|
Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 Download Voicebox v0.5.0 |
Voicebox v0.5.0 Release Notes:The Capture release.Voicebox stops being just a voice-cloning studio and becomes a full AI voice studio. Hold a key anywhere on your machine, speak, release — the transcript lands in the focused text field. Flip the primitive around and any MCP-aware agent — Claude Code, Cursor, Spacebot — speaks back through an on-screen pill in one of your cloned voices. A local LLM sits between the two, so transcripts come out clean and voice profiles can carry a personality that reshapes what the agent says before it gets spoken. Dictation — speak anywhere, paste anywhere
Personality — voice profiles that speak for themselvesVoice profiles now carry an optional personality — a free-form description of who this voice is, up to 2000 characters. When set, two new controls appear next to the generate button, each powered by a new Qwen3 LLM running entirely locally:
The same LLM doubles as the refinement model, so there's one local LLM in the app, not two. API surface. POST /generate, POST /speak, and the MCP voicebox.speak tool accept personality: bool. POST /profiles/{id}/compose powers the shuffle button. MCP client bindings carry a default_personality: bool that applies when personality isn't passed explicitly.Agents — any MCP-aware agent gets a voiceVoicebox ships a built-in Model Context Protocol server at http://127.0.0.1:17493/mcp so Claude Code, Cursor, Windsurf, Cline, VS Code MCP extensions — any MCP-aware agent — can call into your local Voicebox install. Four tools ship with dotted names:
Claude Code one-liner: RefinementA clean transcript needs more than Whisper. Each capture flows through a small Qwen3 LLM that strips fillers, fixes punctuation, and optionally rewrites self-corrections — all on-device.
Captures tab + settingsSettings → Captures is now the home for the whole dictation flow:
Stories — timeline editorThe Stories tab graduates from a TTS sequencer into a real timeline editor. Same generation-row backing, but clips now compose with imported audio, per-clip levels, and a flexible track stack.
Interface
Windows paritySame dictation flow on Windows. Right-hand default chord (Ctrl+Shift) avoids AltGr collisions on layouts where Ctrl+Alt is the compose key. Focus is captured at chord-start so paste lands in the original field even if focus drifts during transcribe/refine. |
Install on Linux
We're currently working through CI issues that prevent us from shipping a reliable pre-built binary for Linux. In the meantime, building from source is straightforward and takes just a few minutes.
Prerequisites
just — install via
cargo install justTauri system deps — see Tauri docs
Build from source
git clone https://github.com/jamiepine/voicebox.git
cd voicebox
just setup
just buildThe built app will be in
tauri/src-tauri/target/release/bundle/
Or run in dev mode
just devKey Features
1. Advanced Speech Generation
Voicebox can generate natural-sounding speech from text while mimicking tone, style, and context. It produces high-quality audio that closely resembles real human voices.
2. Voice Cloning with Minimal Input
With just a few seconds of audio, Voicebox can replicate a speaker’s voice and use it to generate new speech content.
3. Speech Editing and Noise Removal
One of its most impressive features is the ability to edit audio without re-recording:
Remove background noise
Replace words in an existing recording
Fix mistakes seamlessly
This works like an “eraser” for audio.
4. Multilingual Capabilities
Voicebox supports multiple languages and can even transfer voice style across languages, meaning a person’s voice can be recreated in another language naturally.
5. Cross-Task Generalization
Unlike older models, Voicebox can handle tasks it was not explicitly trained for, thanks to in-context learning. This makes it more flexible than traditional speech AI systems.
Real-World Use Cases
Voicebox opens up many possibilities:
Content creation and voiceovers
Audio editing for podcasts and videos
Accessibility tools for visually impaired users
Multilingual communication with natural voice output
Gaming and virtual assistants
It can generate speech up to 20 times faster than some older models, making it practical for real-time applications.
Open Source vs Research Version
There are currently multiple “Voicebox” implementations, which can be confusing:
Meta Voicebox (Research Model)
Not publicly released for full use
Focused on advancing AI speech technology
Open-source Voicebox tools
Local voice cloning apps inspired by similar concepts
Run on your own machine with full privacy
Support multiple TTS engines and editing tools
Enterprise Voicebox platforms
Focus on voice analytics and customer insights
Used by businesses for feedback and automation
Pros and Cons
Pros
Extremely realistic speech generation
Multi-functional AI for voice tasks
Supports multilingual output
Can edit audio without re-recording
Highly flexible and future-ready
Cons
Full Meta version not publicly available
Potential misuse concerns with voice cloning
Requires strong hardware for local implementations
Still evolving and not fully standardized
Who Should Use Voicebox
Voicebox is ideal for:
AI developers and researchers
Content creators working with audio
Companies building voice-enabled products
Users interested in voice cloning and TTS
It is less suitable for casual users looking for a simple plug-and-play app.
Final Verdict
Voicebox is one of the most advanced AI speech technologies available today. It goes far beyond traditional text-to-speech, offering editing, cloning, and multilingual capabilities in a single system.
Voicebox represents the future of voice AI. While still evolving, its capabilities already show how powerful and flexible speech generation technology can become.
If you are exploring cutting-edge voice AI, Voicebox is one of the most powerful technologies in this space. Originally introduced by Meta as a research breakthrough, Voicebox represents a major leap in how machines generate and edit human-like speech.
