《 Draw a Memory! 》

OBJECTIVE

Allows users to interact by drawing out a recalled memory and watch it generate detailed descriptions, while listening to a frequency generated, based on the drawing.


HOW IT WORKS

A Digital Drawing Interface

Users interact by drawing out a recalled memory. Upon pressing the ‘improv!’ button, a frequency sound either low, mid or high could be heard and a paragraph of the memory-based drawing will be generated. This is done using Gemini API prompts in the p5.js code. Users will also alternate between a mouse, trackpad and wacom tablet to see which is most comfortable for drawing.

Through user-testings, it is said that a drawing interface retains a human touch (50% human 50% AI intelligence), making it an expressive experience. Every drawing is also unique, one of a kind and personal. The generated paragraphs based on drawings are fun to read, encouraging engagement. The generated sound frequency is a nice touch to an audio-visual experience. Sometimes the sentences can be so accurate, it hits close to home. Other times, a miss in accuracy can trigger a good laugh.

There are limitations, where sound generated is fixed with only 3 variables: low, mid, and high frequency. This prototype does not have a physical element.

Tasks:

• ✶ Do a pre-survey
• ✶ Test trackpad, mouse & wacom and pick the most comfortable
• ✶ Pick a soundscape from 12 tracks
• ✶ Draw a recalled memory
• ✶ Do a post-survey