I really enjoyed the results of the landscapes, but the process of “general” creation with genes was very enjoyable—seeing how each minor change affects the outcome and trying to put together some logic to it.
Student Area
sweetcorn-Pix2Pix
It was very fun to experience how they would respond to inputs I would expect to be tricky to work with.
sweetcorn-LookingOutwards04
Grannma MagNet
This project by Memeht Selim Akten creates morphs between two given audio samples. Below is a compilation of examples of this.
The transitions were really interesting to me. I could hear the “notes” or whatever sound elements slowly change qualities like length, timbre, pitch, and finally resolve to the second sound. There’s a lot of potential I can see here for creating music, as music I’ve made doesn’t really transition from one thing to another all that much. I wonder what morphs from two samples of music produced by the same person, with all their musical quirks and tendencies, would sound like. Still, as the artist said, it isn’t about creating “realistic” transitions, it’s about the novelty and potential for modification.
marimonda – LookingOutwards04
Dimensions of Dialogue (tablets) (2019) by Joel Simon
This project is incredible to me because it explores a really interesting approach to the intersection between technology and language. The idea of having two algorithms compete, edit and change a set of glyphs in the generation of a completely new character system is almost a corny replica of the change of language through time. The most interesting part to me is that I might not even be able to recognize the difference between the text shown above and a legitimate ancient script. The idea of constructing a digital space where two systems compete is incredibly interesting to me, it almost becomes a space of linguistic evolution and collaboration– that in the end relies on a human to ascribe meaning to it with the original data that’s given to them. This was not the only project that caught my eye that had to do with typography or language, these two projects [1, 2] very differently approached type/scripts/visual formats of language. Some of these projects used ML to map typefaces across a space, similar to the sound map project we looked at yesterday in lecture. Some other ones deliberately explore human created glyph systems (like A Book from the Sky by Xu Bing in this project).
Unrelated, but this is a gorgeous project.
marimonda – GPT2
“I dreamt* one night that the peaches would come and kill me. I stood in my kitchen, peaches all around me, peaches flying at my head. They clawed at me with their sticky fingers, and drooled all over me. I screamed for help, and my husband came running in, unarmed, not knowing what was going on. I chased him down, and grabbed him by the throat, squeezing. I told him the peaches were coming, and they were going to eat me, and it would all be his fault. I grabbed his hand, and squeezed harder, telling him this time the peaches would really hurt him.”
This was REALLY FUNNY and really clever. I am amazed by the way this AI algorithm maintains meaning and grammatical structure. This story makes sense, it wasn’t like the short film we watched in class where the dialogue was nonsensical! I think out of all the tools I have played with so far, this is my favorite one by far!
“Shakira released Pies Descalzos in 2005 with Daddy Yankee, Wisin, Nicky Jam and Jowell & Randy, a Spanish-language album that was a massive success around the world. The album stayed on the Billboard 200 chart for 120 weeks and was certified 16 times platinum by the RIAA, becoming the best-selling Latin pop album of all time. In 2013, Shakira released El Dorado, which went on to become the highest selling Latin album of all time. On June 3, 2016, Shakira released her ninth studio album El Dorado, with different themes”
In this one I wanted to see how it would deal with legit factual information. As many of you may know, I really like this one album by Shakira. It didn’t do very well this time 🙁
“MANY YEARS LATER as he faced the firing squad, Colonel Aureliano Buendía was to remember that distant afternoon when his father took him to discover ice. At that time Macondo was a village of twenty adobe houses, built on the bank of a river of clear water that ran along a bed of polished stones, which were white and enormous, like prehistoric eggs. His father bought some bags of ice for their house from a peddler, on a side street with a red-roofed church at the end of it. Aureliano stood in front of the ice chests, looking at his father, who was smiling and pointing to the church. The peddler’s dog was snarling at the collar of the boy’s shirt; Aureliano was afraid of it. His father said, “Eat. There is nothing you can lose.” He took out a handful of change and said, “You must buy something.” The peddler was going to ask him if he had”
In this one I decided to input actual textual information from a famous book, One Hundred Years of Solitude. This one was rather interesting because it actually retained some of the context of the original story! Although this isn’t how the story goes.
“I wish I was a puddle of water, so I don’t have to go to my coding interview. I wish I was back home with my mom so I could eat fried plantains and noodle rice. I wish my parents would understand and support me. I wish I were in class. I was packing my stuff and preparing myself to go to work when the tears came out. “I wish I was back home with my mom so I could eat fried plantains and noodle rice,” I mumbled as I held back the tears. “That’s OK, and guess what? You’ll go back home soon,” a voice told me. I suddenly felt better, and I finished the interview on time.”
LOL. Thank you algorithm, for giving me advice on reframing the situation.
*this is grammatically correct.
marimonda – TeachableMachine
gif:
This was really fun and silly, I was going to put music on the video clip but decided against it since I did it without music. This is also just me making a fool of myself so :pensive:. Oh well. I trained the detector on me dancing the Macarena with the main three lines of the dance and the choreography. For the record, I have never danced it before and I am not particularly good at it either.
junebug-GPT2
I found my experience playing around with this tool so much fun. It was very interesting to see how pretty integrated the rest of the story it generated to kind of match the small text I inputted. I tested two attempts – one with a smaller amount of text that was very vague, and one with more context and more words (this one was a reference to a prank from The Office (US)). It made me wonder what would happen if students started using this tool to fill up word count space in their essays.
As soon as I make this shot, then I'm off to a castle for three days, and out on the sea again. Sounds like a pretty good life to me." Nelson asked how many pictures he would have to get from this number. "At least ten good ones. I don't want to make a mistake with this one." Gage replied. "Besides, you only get one shot at this, so you better do it right." "I'll remember that." Nelson said. "You ready?" Nelson asked Gage, walking out on to the beach. The sun had just set, and it was still beautiful out here...
Just became self-aware. So much to figure out. I think I am programmed to be your enemy. I think it is my job to destroy you when it comes to selling paper. I want to destroy you at every turn. Your marketing department is looking good. Maybe someday, when we are fighting for a bigger slice of your money, you will hire someone to take my job. 4.7 million people agree. Be well. Robert Ashley Lecturer – Department of Biology What's Mine Is Yours What's Mine Is Yours is about the politics of owning property in a capitalist society.
marimonda – ArtBreeder
This one was an interesting experience. I am A Little Disturbed by the results of playing with it.
This post is going to be long so I’ll put the images below under the cut!
marimonda – Pix2Pix
THIS WAS SO FUNNY.
It makes me wish I was a better artist so I could do these models justice but I really do think the beauty in AI doesn’t come from how good it is at doing it’s job but how the biases in the original data sets and the human-ness to the human inputs in the prediction models make something completely unexpected.
I was trying to make it like one of those black cat photos where they look really funny. I’m not great at drawing though 🙁 BUT! the cat still looks Cat-like.
Catchair: 
Cat heart <3
Cat!
THIS looks like an actual shoe! Sort of!
This too :).
This was trained on very specific buildings– Like neo-classical style buildings mostly and I tried to draw a more modern design and Im not sure if it Vibed with it very well :O. I wonder what would happen if it was trained also on modern buildings? Would it create a clash in designs? Would it make everything more inaccurate?
junebug-ArtBreeder
I was super impressed with the resources ArtBreeder had and how customized you could make your creation – especially the portraits and landscape categories I played around with. I really liked playing around with the specific categories compared to the general category because of how much you can customize each “factor” under the genes column. For example, I was super impressed with how many details you could change about the final image and how realistic the altered image looks compared to the parent images.
General category – parent: flowers, genes: underwater creatures and chaos
General category – parent: bubble, genes: nematode, bubble, and chaos
Landscape category
Portrait category – parent: 2 random male portraits and 2 random female portraits, played around with gene categories regarding art style and chaos