An offline-capable stack that transcribes and translates a consultation on the spot. Audio never has to leave the site, and the workflow assumes a clinician reviews the output before it is used.
Working offline is the constraint that determines everything else here. Once you assume a site where connectivity cannot be taken for granted, sending audio to an external API stops being an option - and the clinical requirement that a patient's voice never leaves the premises is met by the same decision. One constraint answers two demands. Assuming a clinician reviews the output before it is relied on is the right caution for anything touching a consultation.
Some winners chose to assume no network and no cloud — and won precisely because of that handicap.
Good intentions do not win. Every public-good project in this archive nailed the operational details.
Projects that explicitly refuse to let a generative model decide tend to score higher, not lower.
Summaries are written by this site. Project pages include demo videos, screenshots and the team's own write-up (videos may autoplay).