After dominating the photo editing arena for years, Adobe Systems Incorporated has decided to venture into another sector to work their photoshop magic. Last week during the Adobe MAX 2016 event, VoCo was revealed- This is a program that allows you to edit recorded speech, so that you can alter what that person said, or create an entirely new sentence from their voice.
So how does it work? VoCo takes in a large amount of voice data (20 whole minutes worth) breaking it down into phonemes. From these phonemes the program creates a voice model of the speaker- which will probably include cadence, quirks and similar speech patterns. Now when editing someone’s speech, VoCo can either find that word somewhere within the 20-minute clip; or if the word hasn’t been uttered, it is constructed out of raw phonemes.
This is a great leap in technology for audio engineers and others who deal with audio for a living- correcting productions that have audio errors and such like in post has always been a pain. The only thing that worries me is what exactly will happen when underhanded persons get their fingers on this software and decide to target the innocent.