Inspiration
To bring joy in a joyless time.
What it does
We take the keywords of the input text and turn the font and colours into something that reflects the personality of each word, like how Geronimo Stilton books were typeset.
How we built it
We take the input text and run each word through Merriam-Webster's API to determine word type. If it's a suitable type (noun, adjective, adverb, or verb), then we run it through IBM's tone detection API, then take the cosine similarity between the tone of the word and our self-generated dataset of dozens of fonts and 18 colours. We then change the text to the closest font and colour match and then display the resulting work of art.
Challenges we ran into
As our first node project, it was a bit of a struggle figuring out deployment and such. Also it was tedious to wade through dozens of API's for both the word info and the NLP tone detection.
Accomplishments that we're proud of
We're happy to have manually curated a ton of fonts and rated them along the 7 tone axes that IBM uses. Also pretty pleased with how the design turned out.
What we learned
How to run a Node app. How to use IBM's Tone API.
That Geronimo Stilton is the world's 20th best selling series with almost 200 million copies sold (yes, really).
What's next for Stilton Space
Making the world more whimsical, one noun, adjective, adverb, or verb at a time.
Log in or sign up for Devpost to join the conversation.