01110111 01101111 01110010 01100100 01110011
by Aiswarya Sudhir
As someone who struggles to express themselves on the spot, I have always felt comfort in words. When you ink your words, you can take your time. You can fix them to show more accurately what you meant. Words are forgiving. Words are patient. I found my solace in words. In books, in poems, in lyrics and in simple online chats. From reading to writing to typing. Words are the art that I thrive in. In the end, words are never enough but they help. I started coding because I like the idea of being able to create things through my words. Granted, no part of coding is like literature, but to me they were similar enough.
Words can build images. When I read a poem, I can visualize my interpretation of it in my head. Even if they share a few similar characteristics, your visualization will end up being different from mine. I wanted to explore how a computer would interpret the same words that we read. How do poems ‘look’ to an assembly of complicated code? What features of the text was it more likely to pick up on? I gathered a few of my favorite descriptive poems and fed lines from them to a text-to-image generator. And yeah, they ended up looking very different from what I saw. These are some of them, including famous classics to little known gems and one children’s poem.
“About, about, in reel and rout/ The death-fires danced at night;/ The water, like a witch’s oils,/ Burnt green, and blue and white.’
‘The angels, have crouched too long in/ The bruising darkness’
‘the clouds a lucid grey-white canopy/ subduing summer-heat/ to a steady uterine warmth’
‘ My artificial flowers/ I send them to you./ My small bronze lions/ I set up at your door./ I myself sit down on the steps –/ a lost oriental pearl/ in the big city’s noisy sea.’
‘You are the electric indigo/ in the bolts of lightning/ which split the whole sky in half/ breathtaking yet dangerous to watch’
‘Tell me about her silken hair,/ Her rainbow-coloured wings./ Tell of her bag of magic stars;/ Sing me the song she sings.’
‘Like the wild hyacinth flower which on the hills is found,/ Which the passing feet of the shepherds for ever tear and wound,/ Until the purple blossom is trodden in the ground.’
‘I make the netted sunbeam dance/ Against my sandy shallows.’
‘Hanging their bluegreen bellies and their wing panes in a Chinese screen/ The honey-feast of the berries has stunned them; they believe in heaven./ One more hook, and the berries and bushes end.’
Explaining the title and header
The title of this work is ‘Words is 01110111 01101111 01110010 01100100 01110011‘. The word ‘words’ is written in binary code as 01110111 01101111 01110010 01100100 01110011. Binary code is the language we most associate with computers and AI.
The header is a combination of running the word ‘words’ through the text-to-image generator five times. While the images look similar, some parts change every times. While this is just the way the generator works, I felt like it was important as words can seem to change every time you see them, although they remain the same as well.
Tools Used:
- Runway ML text-to-image generator
- https://www.text2speech.org/
- Active Presenter (for Instagram cover)