Voicebox
Open-source local AI voice studio for cloning voices, generating speech, dictation, and giving MCP-aware agents custom voices.
🛡️ AgentReady threat assessment
MAESTRO 7-layer threat model + OWASP AIVSS risk score for Voicebox, derived from its capabilities.
These scores are auto-generated from public information (the agent's own listing, docs, and repository) using the canonical OWASP AIVSS formula and the MAESTRO framework — an estimate for guidance, not a penetration test, audit, or certification. See the scoring methodology — every score is re-derived by the same automated method as an agent's public evidence changes.
Overview
Voicebox is a free, open-source, local-first AI voice studio from the jamiepine/voicebox GitHub project. It is described as an alternative to ElevenLabs and WisprFlow in one app, combining voice output and input workflows. Voicebox can clone voices from a few seconds of audio, generate speech in 23 languages across seven TTS engines, provide global-hotkey dictation into text fields, and let MCP-aware AI agents speak using voices the user owns. The project emphasizes local execution and privacy, with models, voice data, and captures running on the user's machine.
Key features and capabilities
- voice cloning
- text to speech
- speech to text
- dictation
- local first
- mcp
- open source
- multilingual
- tts
- privacy
Use cases
- Cloning a voice from a short reference audio sample
- Generating multilingual text-to-speech locally
- Dictating into text fields with a global hotkey
- Giving MCP-aware AI agents a selected voice
- Creating speech with preset voices
- Running private voice workflows on a local machine