It was penned by an AI developed at the University of Toronto, which builds on deep learning capabilities like those of Google's DeepDream. Instead of turning images into hallucination nightmare dogs, this project took seasonal inputs and kicked out its best approximation of a real human carol. It's called "Neural Karaoke," because good science is unnerving science. To prepare, the AI was given 100 hours of Christmas music, 50 hours of song lyrics, and video feed from the game Just Dance. So the thing can dance too.

A team of researchers at Oxford University have coaxed an artificial intelligence program into an impressive leap forward and towards our own obsolescence. The program, known as LipNet, is showing particularly promising ability to read lips in video clips, thanks to machine learning and a novel way of approaching the data. The key difference is that rather than try to teach the AI the mouth shapes of single words and phonemes, the LipNet is asked to interpret whole sentences. Using GRID, a huge bank of 3 second videos featuring brightly lit forward facing speakers, LipNet has learned to translate speech to text with a 93.4% accuracy rate. Compare that to humans' 52.3%.