BetterZoom: a video call app that can tell when you're nervous

BetterZoom: a video call app that can tell when you're nervous

●1 ●5 ●13
calendar_today ago • schedule2 min read
— Originally published at dev.to

We've all been on a call where our heart was racing and our voice was shaking, and nobody noticed, not even us until it was over. I wanted to change that, so I built BetterZoom. It's a Windows desktop app that uses Presage's SmartSpectra SDK to read your pulse, breathing and facial expressions from an ordinary webcam. You don't need a smartwatch or any sensors, just the camera you already have. It turns those readings into simple feedback you can act on while you're still talking, not hours later.

The app has two modes. Self mode is for practice. Before an interview, a pitch or a presentation, you start a session and talk to the camera. A coach watches your numbers and gives you short tips every so often, written by Gemini from your live readings. You can pick a normal, friendly coach or Smokey, a loud cartoon boxing coach with a hat and a cigar who talks to you like you're about to step into the ring. He can even read his tips out loud. A meme-style face shows your mood, from "VIBING" to "SWEATING". When you finish, you get a debrief with your pulse, breathing and expressions over time that you can export.

Meetings mode is the main highlight. You create a room, share the code, and people join a peer-to-peer video call. Every participant gets a clear consent screen before anything is measured. Once they agree, only a small set of numbers goes to the host every 25 seconds: pulse, breathing, heart rhythm, calm state and expressions. The raw video is never shared. The host sees a simple dashboard of how the room is doing and can generate an AI summary at the end. Picture a teacher who can tell the class is getting tense, or an interviewer who sees the candidate needs a moment.

Setup is easy. Download the installer from GitHub (https://github.com/Msohaib-maker/SneakMeet/releases/latest), run it, and you're ready. You don't need any API keys, because the server handles that. The readings are wellness estimates, not medical advice, but they're a surprisingly honest mirror. The source code is open at https://github.com/Msohaib-maker/SneakMeet. I'd love for you to try it and tell me what Smokey said to you.

Want more #ai & #github deep dives?

Join 5,978+ developers sharing real benchmarks, code breakdowns, and projects. Free forever.

More Posts

Cisco's Amy Chang: A Model's "Passport" Doesn't Tell You Where It Actually Came From

Tom Smithverified - Aug 27

Dozens of email accounts: learning AI coding tools as a student

rahuldk1105 - Oct 27

I’m a Senior Dev and I’ve Forgotten How to Think Without a Prompt

Karol Modelski - Mar 19

Europe Just Dropped the Hammer on AI: A Wake-Up Call?

PrabashanaDev - Jul 15

How I Built a React Portfolio in 7 Days That Landed ₹1.2L in Freelance Work

Dharanidharan - Feb 9
chevron_left
269 Points • 19 Badges
4Posts
5Comments
7Connections
- Speaks urdu, russian and english
- Experienced programmer with over 3 years of professional experi... Show more

Related Jobs

Commenters (This Week)

5 comments
1 comment
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!