As a podcaster I often wish it was easier to edit my spoken words after I have recorded an episode of my podcast Observed.Tech.
I do each show in a single recording session - about 45 minutes each time - and to be honest trying to edit out all the little imperfections, missed words and incorrect words is time consuming. So most of the time my show is what gets laid down in that single recording session and I just release it like that to listeners.
After watching this demo I could potentially type my entire podcast instead of recording my actual words but would that still be an audio podcast in that case?
That is why this technology from Adobe, called VoCo, is both intriguing and frightening at the same time.
A few days ago at Adobe MAX Sneak Peeks, a demo by Zeyo Jin showed how this software project brings photoshopping to recorded voices.
Here is the seven minute demo video of Adobe VoCo:
As you can see the applications of this technology are quite handy for fixing mistakes in a recording or maybe adding a missed segment to a show. You could even replicate the individuals voice to create a new voice over segment for a project.
In addition, as shown in this demo, you can even delete words that were originally spoken on the recording.
While this is very exciting technology it could potentially be subverted and used inappropriately in any number of ways. Yes, I admit that most any technology can be used in nefarious ways. I mean who among us have not fallen prey to a good photoshopping job.
If this technology comes to full fruition then it will create an entirely new area that we would need to have an accompanying process in place to authenticate recordings as being genuine and only containing words spoken by the original speaker.
Basically, voice recordings would need their own form of a watermark like we see on many digital documents and images these days.
So what do you think about the potential of a technology like VoCo?