Open-Source Voicebox Studio Launches with Local Voice Cloning, Challenging ElevenLabs with 5x Faster Apple Silicon Performance
Summary
Voicebox Studio launches as a free, open-source voice cloning platform that runs locally on users' computers, delivering speech synthesis 5x faster than competitors on Apple Silicon while challenging cloud-based services like ElevenLabs with professional editing tools and developer APIs.
Key Points
- Voicebox launches as an open-source voice synthesis studio that runs locally on users' machines, offering voice cloning and speech generation powered by Qwen3-TTS as a free alternative to cloud services like ElevenLabs
- The platform features a DAW-like interface with multi-track timeline editing, voice profile management, conversation mixing, and API integration for developers building voice-powered applications
- The software achieves 4-5x faster inference on Apple Silicon using MLX backend with Metal acceleration, while supporting Windows and Linux through PyTorch, with plans to add real-time synthesis and additional voice models