Hey everyone. We open-sourced our product Assemblix, and just added TTS and AI-avatar integrations, so your agents can now talk with a voice and a face, not just text.
It is a self-hosted, source-available engine for building chat workflows visually, an n8n-style tool aimed at chat and voice agents. Handy for client work: runs on their own infra with their own keys, multi-provider, and approachable enough for non-devs to maintain after you ship.
For context, we run it at our own communication school, around 2,000 exercises a month go through the engine. Voice and avatars are a fresh addition; the core has been in production about half a year.
If you find it useful, I'd really appreciate a star on GitHub) it genuinely helps us a lot.