Bright

• 

  Lost in the symphony of colors, she sees the world in a way only she can hear.

Bright

Inspiration

I am Nigerian, and my country enjoys an inventive and successful music industry. Making music in Nigeria, especially today, can be surprisingly democratic. College students produce beats on the weekends. Local church choirs release their own albums.

Recently, I read about The Sound Color project by Kevin Groat and Derek Torsani, which was created to use color to bring the experience of music to people without great hearing.

I thought visualization could be more than a way to allow people to see music. It can also be a way for them to create music. And because color and sound have infinite functions that map between their many representations, this was also a technically fascinating question.

Project

Bright takes in a natural language prompt and, in response to it, generates a song you can see. I built it in JavaScript with the D3 library, the Web Audio API, and the Suno API. I also learned a lot about visualization and the math of sound.

Some challenges we ran into were authorization hiccups with the generative AI API. Plus, its API limits got hit fast.

But the project was complete. It looks beautiful and is a great starting point for a more ambitious undertaking.

Next for Bright

1. Simultaneous video generation alongside the audio.
   
2. A canvas feature that will let users create music by painting onto a screen.
   
3. There is no incredible, open-source sound-to-visuals converter. I want to build one, especially incorporating ideas and feels from both synthesia realities and VJ-ing practices.

Built With

• d3.js
• javascript
• suno
• web