Open-source voice infrastructure for everyone

We build open-source tools to make voice AI accessible, production-ready, and easy to integrate.

Voice should be a first-class interface in every application — not a luxury reserved for big tech. We're building the open-source infrastructure to make that happen.

Project	Description
Macaw OpenVoice	Production voice runtime — STT + TTS with OpenAI-compatible API, real-time streaming, and multi-engine support

pip install macaw-openvoice[server,grpc,faster-whisper]
macaw pull faster-whisper-tiny
macaw serve

That's it. You now have a voice API running locally with OpenAI SDK compatibility.

• Documentation — guides, API reference, architecture
• Contributing — we welcome contributions
• Releases — changelog and downloads

_{Apache 2.0 Licensed — Built in the open}