Roy wanted to understand how children learn language in order to build robots than learn in the same way – and what better test subject than his newborn son!? To understand his son’s acquisition of language, Roy set out to capture every word the baby and his caregivers spoke and record the environment and events around which these utterances occurred. He wired his house with cameras and microphones, resulting in 200 terabytes (90,000 hours!) of audio/video data, or “the largest home video collection ever made!”
Roy gave special recognition to his grad students during his talk, and I can imagine the gratitude was well-deserved after the months of footage they waded through!