Photoshopping Audio with Adobe
Adobe demonstrated a crazy audio feature at their annual MAX event last week that enables users to replace words of dialogue simply by typing in the text you want instead of having to actually edit the audio. Adobe says it requires about 20 minutes of spoken word from the speaker to have enough material to pull from.
Imagine using this for music. With technologies for voice recognition, word replacement, and autotuning/melodyning, users could create multiple album’s worth of melodies and harmonies with only one 20-minute session with the vocalist. Or, you could gather an artist’s existing vocal stems and write and create new songs using their voice.
For simpler applications, adjustments could be made to the lyrics of a song long after the singer is gone, dialogue for movies or TV could be replaced very easily, and audiobooks could tell totally different stories than what was originally in their pages.
Of course, there will be nefarious uses as well, such as using the software to alter presidential speeches or make a singer use profanity that they wouldn’t normally use.
It will be interesting to observe how this technology progresses. Watch the video to see it in action, with Keegan-Michael Key and Jordan Peele as the example.