Inspiration
This project was inspired by an idea I had in high school where you would open the app and play a sound based on what felt like the right moment. The only thing you knew was the name of the sound effect before you played it so it would be a surprise for everyone involved. With this web app, I revisited this initial idea and expanded upon it.
What it does
The app asks you to capture 3 different moments. Once you're ready, you are given a chance to capture the moment through a video recording. Once the recording starts, you have access to the sound effect button that will play the unique sound based on the challenge given. After recording all three challenges for the day, each moment is combined into one video with the challenge and time captured at the top of the screen in various styles. The video can then be exported and shared with friends or on social media.
How I built it
The entire web app was created with Codex using GPT 5.6 Terra. The sound effects were mostly gathered through Fressound.org with the Creative Commons 0 license. Some sound effects were generated with Adobe Firefly.
Challenges I ran into
While the app worked fine in the ChatGPT browser within Codex, issues did arise on iOS and Android once more features were added. I prioritized the Android build over iOS for this challenge as it seemed to run into fewer roadblocks with the React web app that was generated.
What I learned
This is the first app I have made using generative AI and it was a unique learning process for me. I was surprised by how the codebase was a lot smaller than I would have thought for an app like this and I was able to focus on the unique aspects of the app without getting bogged down on the technical implementations.
What's next for In The Moment
The web app is stable on Android as submitted but I want to continue work on refining the app for iOS as well. I'll explore converting the app to a native application for both platforms as well as a way to generate daily themes and sounds and integrate with a cloud server to send those to the devices. Future versions would add more styles to the exported video and direct integration with social media platforms rather than the current export and upload solution that exists in this initial release.
Built With
- codex
- openai

Log in or sign up for Devpost to join the conversation.