Inspiration

In messaging apps like Facebook Messenger, nicknames for friends in individual or group chats add to the humor. Adding emojis only adds to the humor as now a fun image is associated with the friend. However, it can be difficult to associate emojis directly to names without serious thought. We wanted to create something that would automatically generate a nickname and corresponding emoji to make chats as fun as can be.

What it does

Namoji uses OpenAI APIs to convert an inputted name to emojis. After inputting a name, the AI associates the name with an English word, or multiple words, and then associates one or several emojis with the newly created word(s). The feature is on a website for ease of access and use.

How we built it

Based on OpenAI tutorials, we mode a website that changes a given name into a series of one or several emojis. To train the system, we used a stepwise approach by first assigning similarly spelled English words to the name, and then assigning emojis to those words. After doing each of these steps in the OpenAI Playground, we combined the code in Google CoLab to test the comprehensive code (in Python), and then transferred the code to a Glitch website and reformatted to create a user-friendly interface.

Challenges we ran into

We ran into some bugs when we were training the data, as we edited Stop Sequences and ensured that each name was converted first to a word, and then to an emoji. Combining and executing the two operations together also posed some challenges in terms of trying to only show the final product. The biggest challenges occurred when implementing the code on Glitch, as that involved writing everything in JavaScript.

Accomplishments that we're proud of

We’re most proud of the ease of use and simplicity of our project. Adapting the API code to JavaScript took some reformatting, but it allowed for a more presentable and simpler user experience. Adapting the code also improved our JavaScript skills, and getting the APIs to work not only actually made the project function, but showed how our skills grew.

What we learned

We developed a basic understanding of how APIs work and can be used for both text completion and text-to-image functions. Specifically, we learned how to do API calls so people can actually use the app. We also gained more experience in JavaScript and Web Development for when we were creating the user-friendly platform to use our project.

What's next for Namoji

Namoji could be expanded to include full names, and differentiate between first, last, and full names. Moving forward, we’re also looking for how to make the applet more user-friendly and mobile-compatible so that individuals on different devices can use the application as well.

Built With

Share this project:

Updates