Gene Kogan is a programmer and artist who is interested in generative systems, machine learning, and the application of emergent technologies into artistic and expressive contexts. He writes software for live music, dance, theater, performance, and visual art.
He contributes to numerous open-source software projects and frequently gives workshops and demonstrations on topics related to code and art.
The subject is a series of experimental works in producing images, video, sound, and text using recent advances in unsupervised deep learning research. They include:
Similar approach to text generation (next section) but trained on audio (in frequency domain) instead of characters, producing noisy audio which nevertheless resembles the source it was trained on. Audio generation is less developed than text and image generation but very promising. Some samples: