Taylor Swift-trained Neural Network Storyteller

1-63kjGm-IX1mBUnHJR3tIkw

Herr Samim (hier auf Twitter, vorher auf NC: Machine generated TED-Talks) hat ein Neural Network auf Taylor Swift-Lyrics und Kitschromane trainiert und generiert damit Foto-Beschreibungen. Ganz, ganz großartig. Oben ein bisschen Robot-Geknutsche für die Frisur des Satans, hier zwei Clips dazu:

Stories are a fundamental human tool that we use to communicate thought. Creating a stories about a image is a difficult task that many struggle with. New machine-learning experiments are enabling us to generate stories based on the content of images. This experiment explores how to generate little romantic stories about images (incl. guest star Taylor Swift). […]

This experiment started by running 5000 randomly selected web-images through neural-storyteller and experimenting with hyper-parameters. neural-storyteller comes with 2 pre-trained models: One trained on 14 million passages of romance novels, the other trained on Taylor Swift Lyrics. Inputs and outputs were manually filtered and recombined into two videos.

1-brmI0US0cs00qwg4y0lMvg

1-c8eKdZpvqCFcRjq_0dHozA

1-CCF_gTs16d_FmosXZAb_UA

Bonustrack: Reed Morgan Milewicz hat einer AI die Regeln zu Magic The Gathering beigebracht, daraus hat er nun den Twitter-Bot @RoboRosewater gebastelt, der am laufenden Band Algo-Karten generiert, hier der Code auf GitHub. (via Boing Boing)

magic