Higgs Audio Hackathon returns to Toronto with MScAC
October 23–24, 2026. One build day. Your next big idea, out loud.
Build something
worth hearing.
October 23–24, 2026
Sign up on Luma Explore the weekend ↓
Boson AI and the University of Toronto’s Master of Science in Applied Computing (MScAC) program reunite for the Higgs Audio Hackathon, October 23–24, 2026—the weekend before UofT reading week. Friday kickoff. Saturday building, demos, and celebration.
Meet us in Toronto.
Join us at MScAC headquarters, University of Toronto: 700 University Avenue, 9th Floor, Toronto, ON M5G 1Z5.
All are welcome, including participants from beyond UofT. Come solo or with a team; Friday evening includes time to meet collaborators and form teams.
We’re hiring. Boson AI CEO Alex Smola and many of our team members will be there to meet participants, see your work, and discuss opportunities to join us.
Build for real conversations.
Show what Higgs makes possible through multilingualism, natural interaction, and thoughtful integration.
Avatar experiences
Feature Higgs Avatar in an expressive experience that brings a face to speech, with lip-sync and facial expression that follow the voice.
Voice interaction
Build a voice-only experience with natural pacing, turn-taking, and interruptions. Show how it handles different languages.
Tools and agents
Integrate tool calling or an agent into a polished application. Demonstrate a useful action and its result.
Present a clean, visually appealing, informative demo that makes the use case, integration, and result easy to understand. Explore TTS text normalization and word-level timestamps for naturally spoken numbers and synchronized captions.
From first idea to final demo.
October 23–24 · Roughly 24 hours · Tentative schedule
Get a head start.
Participants receive early API access and sufficient credits one week before the hackathon. Explore the models and tutorials so you can hit the ground running when the event starts.
Kick off. Learn. Connect.
Registration opens at 5 p.m. An evening of kickoff talks, tutorials on model serving and building with Higgs Realtime and Avatar, networking, and team formation follows.
Build. Demo. Celebrate.
The main build day. Presentations start around 3 p.m.; final rounds and judging wrap up by 6–7 p.m. Then we head to pub night.
All times are local to Toronto. Final timings will follow.
A voice. A face. A new possibility.
Get set up with Higgs Realtime and Avatar using your early API access and credits. Work through the tutorials before kickoff so you can arrive ready to build.
Start with the Realtime tutorial for voice and tool calling, or the Avatar quickstart for talking video from an image and speech.
Last year set the stage.
200+ builders. 50+ projects. Last year’s Toronto winner, EchoCharlie, turned silent video into speech. Your turn.


