TRANSCRIPT:   Jamie Dupree is a radio reporter with Cox Media Group in the US, for WSB: and two years ago he lost his voice.   As a story on the BBC will tell you… he’s got it back. Kind of. A company based in Scotland has sampled his voice from a set of recordings: and he’s now back filing full reports for the company with the aid of voice synthesis.   The technology isn’t 100% perfect, but it’s certainly well on the way; and that’s - of course - interesting to radio.   If consistency is one thing that makes radio work - and most radio programmers agree - then it’s an obvious next step to have not just consistent music but consistent radio presenters. Indeed: the same radio presenters, twenty-four hours a day.   There are, already, radio stations like this. I wrote about the excellent, tight-sounding Carolina Classic Hits a while ago, which has the same radio presenter on-air twenty-four hours a day. The station sounds great, with crunch-and-roll links and an almost relentless up-tempo sound: and Rick Freeman has been on the air every hour of every day since 2012, thanks to a bank of well over 5,000 voicetracks.   But recording voice-tracks only gets you so far. What if you could go a little further. If voicetracked weeks in advance, your favourite radio personality can’t talk about yesterday’s World Cup game. But with this type of technology he can.   Indeed, this could even make famous radio presenters come back from the grave. (Indeed, this has already been done: Bob Monkhouse, a UK comedian, famously did a TV ad for prostate cancer awareness, four years after he died of the disease).   A number of years ago, I saw a presentation by a German radio company, who’d added a personalised clock alarm to their radio app. You’d wake up in the morning to hear the breakfast radio presenters say “Good morning James!” - clever, but it turns out that you only need record a hundred first names to cover the vast majority of your audience.   With voice synthesis, personalised radio could be something else.   “Now, here’s a great new song from Taylor Swift: and you’ve ten minutes, James, before you need to leave for work, so crank this song up while you get ready!”   “Should be a warm day today, but maybe some rain this afternoon. In your task list it says you need to visit the dry-cleaner, so maybe you want to do that this morning?”   The opportunities are endless.   And for television? Well, the technology to produce fake videos - perhaps to make Barack Obama say something he didn’t really say - is already out there - and getting much more realistic.   The preceding column was written by a human being. At least, this week it was.  Support the show.This podcast uses the following third-party services for analysis: OP3 - https://op3.dev/privacy