voicebox
The open-source AI voice studio. Clone, dictate, create.
Voicebox is an open-source AI voice studio that includes a built-in MCP server for integrating text-to-speech and speech-to-text capabilities into AI agents and applications. It provides voice cloning, multi-engine speech generation across 23 languages, global dictation, and bidirectional voice I/O that runs entirely on your local machine. The MCP server exposes tools for agents to speak in cloned voices, transcribe audio, and manage voice profiles and captures. Developers can connect it to Claude Code, Cursor, Windsurf, and other MCP-aware clients via HTTP or stdio transport to give their agents natural voice output and input capabilities.
Key Features
Use Cases
- 01Give Claude Code or Cursor a cloned voice for speaking task completions and notifications
- 02Build voice-enabled chatbots and assistants with custom personality-driven speech
- 03Automate podcast or dialogue generation using the multi-voice Stories editor
- 04Enable hands-free coding workflows with global dictation transcribed via Whisper
- 05Create accessible voice interfaces for users who need speech assistance
- 06Integrate text-to-speech into custom applications via the REST API or MCP tools
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
voicebox — FAQ
What is the Voicebox MCP server?+
Voicebox includes a built-in Model Context Protocol server that exposes text-to-speech, speech-to-text, and voice profile management as tools for AI agents. It allows MCP-aware clients like Claude Code, Cursor, and Windsurf to speak in cloned voices, transcribe audio, and access voice capabilities through standardized tool calls.
How do I install the Voicebox MCP server in Claude Code?+
Run 'claude mcp add voicebox --transport http --url http://127.0.0.1:17493/mcp --header "X-Voicebox-Client-Id: claude-code"' after installing the Voicebox desktop app. The MCP server runs automatically when the Voicebox app is running on port 17493.
Which AI clients support the Voicebox MCP server?+
The Voicebox MCP server works with Claude Code, Cursor, Windsurf, Cline, VS Code MCP extensions, and any client supporting HTTP or stdio MCP transports. Per-client voice bindings can be configured in the Voicebox app's Settings → MCP section.
Do I need API keys or cloud services to use Voicebox?+
No API keys are required. Voicebox runs entirely locally with all models, voice cloning, and processing happening on your machine. You need to download the Voicebox desktop app for macOS, Windows, or Docker; Linux users must build from source.
Is the Voicebox MCP server free?+
Yes, Voicebox is completely free and open-source under the MIT license. All TTS engines, voice cloning, dictation, and MCP server functionality are included at no cost.
What hardware do I need to run Voicebox?+
Voicebox runs on CPU without GPU acceleration, though performance is significantly faster with Apple Silicon (MLX/Metal), NVIDIA GPUs (CUDA), AMD GPUs (ROCm), or Intel Arc. Minimum requirements depend on the TTS engine, with LuxTTS requiring only ~1GB VRAM while larger models like TADA 3B need more resources.
How do I install voicebox?+
Open the source repository on GitHub and follow its README. voicebox is a mcp server — MCP Agents Market links you directly to the official repo.
Is voicebox free?+
voicebox is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.