In a world where deepfakes videos are a real thing, it's not surprising to read that researchers have developed software that can create a video lecture from an audio narration (though one wonders: why?). In an arXiv preprint (9 page PDF) researchers Byung-Hak Kim and Varun Ganapathi "present LumièreNet, a simple, modular, and completely deep-learning based architecture that synthesizes high quality, full-pose headshot lecture videos from instructor's new audio narration of any length." This article summarizes the paper.
Today: 7 Total: 123 [Direct link] [Share]
