• Get involved.
    We want your input!
    Apply for Membership and join the conversations about everything related to broadcasting.

    After we receive your registration, a moderator will review it. After your registration is approved, you will be permitted to post.
    If you use a disposable or false email address, your registration will be rejected.

    After your membership is approved, please take a minute to tell us a little bit about yourself.
    https://www.radiodiscussions.com/forums/introduce-yourself.1088/

    Thanks in advance and have fun!
    RadioDiscussions Administrators

Using Artificial Intelligence in Developing Broadcast Programming

davideduardo

davideduardo

Moderator/Administrator
Staff member
From David Oxenford's blog at Using Artificial Intelligence in Developing Broadcast Programming – Watch for Legal Issues

Using AI to provide compelling local content to distinguish a broadcaster from its digital competition promises to help broadcasters maintain the relationships they need to survive with their local communities, even at a time when broadcast budgets are stretched thin, making the hiring of staff to provide local service difficult. If AI can deliver content relevant to audiences at a lower cost, it might seem like a winning proposition for broadcasters. But, as always, broadcasters need to be sure that they recognize the risks that arise from such services and ensure that the services that they are buying have safeguards to minimize these risks. Plus, they should look for indemnifications if the risks are not properly managed.
This is a free, non-firewalled blog and worth reading by those involved with broadcasting.
 
We will see. I dont know. I'm just not there yet that it will be the thing that necessarily makes much of a difference. But it might.
 
DavidEduardo said:
Using AI to provide compelling local content to distinguish a broadcaster from its digital competition....
Click to expand...

That feels like a contradiction to me. It's not as though digital competition doesn't have access to AI.

I was driving in West Virginia, scanning the FM dial to hear what was there, and found an automated classic rock station using what sounded like AI to back announce songs. I say it sounded like AI because it was an unemotional voice giving the artist name and song title. That was it. TBH I've heard automated stations use human voices to do the same thing. But the precise way the information was given sounded very AI. I'm not exactly sure how it was being done, if they used an outside service or a plugin in their automation. But it caught my attention.
 
TheBigA said:
That feels like a contradiction to me. It's not as though digital competition doesn't have access to AI.

I was driving in West Virginia, scanning the FM dial to hear what was there, and found an automated classic rock station using what sounded like AI to back announce songs. I say it sounded like AI because it was an unemotional voice giving the artist name and song title. That was it. TBH I've heard automated stations use human voices to do the same thing. But the precise way the information was given sounded very AI. I'm not exactly sure how it was being done, if they used an outside service or a plugin in their automation. But it caught my attention.
Click to expand...
How much emotion do you need in a voice back-selling "Californication" for the hundredth time in the past month?
 
To answer CTListener: as a former air talent before moving on to management and sales, your back announce is to sound as if you are playing your favorite songs because for the listener that might just be and they didn't hear the prior 99 spins of the song. It's not just back announcing. We'd run promos for contests and other things that had to be announced. Our PD made us present each promo it in a different way every hour of our shift for an entire week (4 hours M-F doing afternoon drive and 5 hours Saturday middays. By Thursday your brain hurt trying to create a new way to say the same old stuff and with the same emotion you had during the first hour on Monday, but you did so because it might be the first time the listener heard it.
 
TheBigA said:
That feels like a contradiction to me. It's not as though digital competition doesn't have access to AI.

I was driving in West Virginia, scanning the FM dial to hear what was there, and found an automated classic rock station using what sounded like AI to back announce songs. I say it sounded like AI because it was an unemotional voice giving the artist name and song title. That was it. TBH I've heard automated stations use human voices to do the same thing. But the precise way the information was given sounded very AI. I'm not exactly sure how it was being done, if they used an outside service or a plugin in their automation. But it caught my attention.
Click to expand...
The Fish in Atlanta has have a female voice back announce a lot of their songs for many years.

Was the station around Beckley WV? If it is the one I am thinking of it has back announcing of songs with a bland voice before COVID.
 
CTListener said:
How much emotion do you need in a voice back-selling "Californication" for the hundredth time in the past month?
Click to expand...
I’d be happy not hearing any voice tracking. Just move to the next song. It’s how a lot of the channels on SXM are. No banter, just a promo ID every few songs.
 
And Don CT, that's exactly what radio gets a black eye for doing. For some reason, just playing the music is the worst radio can do. I don't get it. Those that complain about radio doing this and say that's driving people to go online to stations that, by the way, just play the songs and that's it. So, somebody needs to explain how radio is driving people away to online music streams if the radio station does the same as the online stations.
 
b-turner said:
And Don CT, that's exactly what radio gets a black eye for doing. For some reason, just playing the music is the worst radio can do. I don't get it. Those that complain about radio doing this and say that's driving people to go online to stations that, by the way, just play the songs and that's it. So, somebody needs to explain how radio is driving people away to online music streams if the radio station does the same as the online stations.
Click to expand...
I wonder if SXM has done research on what channels to go voiceless vs the ones to do the “radio” format.
 
Don CT said:
I wonder if SXM has done research on what channels to go voiceless vs the ones to do the “radio” format.
Click to expand...
I know XM did this over 20 years ago. The research was not as absolute as a "yes" or "no" to talk/chat/DJs, but had to do with how much and how often listeners liked it in different music formats.
 
TheBigA said:
I was driving in West Virginia, scanning the FM dial to hear what was there, and found an automated classic rock station using what sounded like AI to back announce songs. I say it sounded like AI because it was an unemotional voice giving the artist name and song title. That was it. TBH I've heard automated stations use human voices to do the same thing. But the precise way the information was given sounded very AI. I'm not exactly sure how it was being done, if they used an outside service or a plugin in their automation. But it caught my attention.
Click to expand...
It might be as simple as using the text to speech function available through Amazon Web Services (and probably through its competitors) -- that actually predates the current AI hype and is good enough that at its best it doesn't sound all that much different from the automated back announces in automated radio from the seventies and early eighties.

Alternatively, maybe they've captured the spirit of John Borders and are using him to back announce their music. (He was the voice of the "TM Stereo Rock" format" for much of that format's existence)
 
Did the "voice" back announce more than one song? If not it back announcement was most likely recorded when the song was put in the system.

Having ran several business in WV, I am sure whatever they are doing is the cheapest or free. The business environment is so poor there, the State of West Virginia is offering up to $20k for at home workers to move there.
 
They most likely have a category of songs in the back announcement recorded. That category is always the last song in the music sweep. If they could have an extra large hard drive and have all songs with or without the back announcement but you music generator program would have to know to pick the voiced version as the last song.. AI could easily do if you are going to have A I, you should have a working A I voice too. There are a lot of other things it could do.
 
While people in broadcasting are wringing their hands about AI, the folks at Spotify appear to be embracing it, and offering AI DJs for their users. Because we all know Spotify is in such dire financial shape that they can't afford actual humans. But here's how they plan to utilize AI for their subscribers.


newsroom.spotify.com

Spotify Debuts a New AI DJ, Right in Your Pocket — Spotify

Personalization is at the heart of what we do at Spotify—just think of fan-favorite playlists like Discover Weekly, or our annual Wrapped campaign. The beauty of these experiences is our ability to deliver the right piece of music for that exact moment in time, and maybe even connect you with...
newsroom.spotify.com newsroom.spotify.com
 
TheBigA said:
There's enough of a difference that it stood out to me. You might also ask if there's even a need to back-announce a 30 year old song.
Click to expand...
And there you have it...

Human emotion, which AI will never be able to replicate. That's why I believe the human voice will still have a place on radio. The AI that I've heard on radio has pretty much been the weather report from Weatherology.com for example. It always sounds robotic like, with no emotion, or pausing.
 
I heard a news broadcast on WRMI, and while it sounded more or less like a voice, the inflections and pauses were indeed very mechanical and robotic, and it droned on for at least 10 minutes with no pauses or breaths.

I'm pretty sure it was either good old-fashioned text-to-speech (which has been around in some form since at least the 70s) or "AI" (which is more or less the same thing, but bound up in layers of hype), and if that is the future of anything called "radio", I'm not impressed.

c
 
Status
This thread has been closed due to inactivity. You can create a new thread to discuss this topic.


Back
Top Bottom