It wasn't that long ago, before the days of Photoshop and Content-Aware fill, that we thought that the camera never lies. Those days are numbered for the movie camera now too. A recent paper from University of Washington researchers describes how, given only audio of someone's voice, they can take an existing video of someone's face (not necessarily the same person!) and lip-sync it perfectly to the audio track.
This can be used to replay an old speech in a new setting, or even replace speech in a video with that of a voice impersonator, saying something completely different. This technology has some positive applications, like re-sequencing a recorded speech to be shorter and in a different order, but I suspect it will get used for mischief as well. You can find more details here.
That's all from us for this week. Monday's a holiday here in the US, so we'll be back with more for the blog on Tuesday. See you then!