The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Killing Guyz, I'm back with my first fully no-cost python project called Jarvis in female voice, just completed the HUD, and complex features.
I've started past 2 months ago, and just completed and committed full project in github publicly, it works without any api keys, without any cost and all, I've used different different ollama models for brain, vision, coding agent, security etc. with openai whisper local model for stt and kokoro tts models with multi languages you can implement if you want, models are working smoothly.
Then make it like Tony Stark jarvis I've implemented hand gestures using mediapipe hand gesture python library produced by google, and also you can use groq api too, for some better and trained models too if you want. Currently I don't have image but I will post the complete video here of what how will it work, how can it fallback if some models is not working and change to alternate.
Are you guys excited for demo video and Github repo link? - If, yes than just like and comment on this post, I will personally send you the Repo link.
Don't forget to follow.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
DEPARTMENT OF HUMAN VERIFICATION 🗳️
warning: for full experience do the test in the link above, before reading or watching explanation below. No data is stored.
--
I worked in digital privacy for years, and it was hard because the issues are invisible, - you don't feel or see your privacy being violated. That's part of the trick.
Every CAPTCHA watches how you move - your hesitation, your timing, the way your cursor drifts. It just never tells you. This is the same checkpoint except with a narrator. I wanted to make the invisible visible, - just how much data we give away unknowingly.
But surprising to me - looking at the data - it turned out to be really about the thing everyone's panicking about. Am I going to be replaced with ai? The data, after comparing it to ai, turned out to be proof of how important humans are.
Ai needs our hesitation, our failing smiles, the unauthorized blinks, and it can't produce a single one of them on its own. The machine fails this test by being perfect. A human passes it by being, well, human.
--
Built in Figma Make, imagery in Weave, facial detection via MediaPipe - runs entirely in your browser, keeps nothing.
This was a fun proof that artists can execute their ideas fast, with the tools available. The idea took me almost week. I made the website in one afternoon.
Thank you to my friends that tested it out and recorded their first reactions so I can use them in the video 📼
Links: Figma Community, LinkedIn, TikTok.
A deep-dive on making this coming on my Substack next days.
Nice work
I designed EyeText 👁️, an eye-controlled communication experience for people who have difficulty speaking or using their hands.
Using eye tracking and intentional blinks, users can navigate a keyboard, create messages, and convert them into speech. The experience also includes personalized calibration and quick communication cards for everyday needs.
What excites me most about this project is how quickly an idea became reality. Using Figma Make, I was able to transform a concept into a working interactive experience with a single prompt.
Design, accessibility, and technology coming together to help people express themselves.
🔗 Live Demo:https://serif-modify-89640782.figma.site
🔗X / Twitter Post: https://x.com/Naghdaliyevaart/status/2067282836021776819?s=20
This hero section background lowered my cortisol, great work Nigar!
Project Name: First Frame
Live Project Link: myfirstframe.figma.site
Problem:
We often see successful people for who they are today, but rarely see where they started. Their first steps, struggles, and early journey are usually lost.
Solution:
First Frame is a digital museum of beginnings. Users share their past and present, and the system generates a personalized frame with growth scores, skill insights, and a unique artifact identity. Every story becomes part of a shared, explorable canvas.
Process and Figma Suite:
The project started with ideation in ChatGPT, followed by multiple iterations in Figma. I explored wireframes, liquid glass concepts, and AI-generated prototypes before settling on a clean flat design. The experience was built using Figma Design and Figma Make, with custom interactions, infinite canvas navigation, story cards, and AI-generated frame analysis.
Next Steps:
Improve frame analysis, enhance image exports, add richer social sharing, and expand the canvas into a living archive of personal growth stories.
Trending
Claude
Claude has entered the design space. How are you using Claude Design?
Contra University
Learn from expert creatives how to earn more using next-gen AI tools.
MagicPath
The canvas is infinite, and exploration is becoming the workflow. How are you using MagicPath?
creativeaiflow
Creative AI workflows are evolving. What tools do you use, and what are their strengths and weaknesses?
freelancerlife
Freelancer life is wins, pivots, and everything in between. What’s yours right now?