Highly controversial topic. Would you/Did you use successfully AI generated voice overs in your audio projects (radio jingles, DJ drops, podcast intro, narration etc - nothing here is off the table)? How did that go? Post an example in the comments below. What’s your take on the whole AI voice over scene?
1 Like
I surely do not hope that Voice over artists are over. I work with both on a regular basis. The choice that we make is often based on the how multilingual your content is. I produce video content for e-learning purposes. We always have a big plus for the real voice. And when it’s only produced in one language (most of the time) a real voice is in my opinion always better (still). But I must say that I use Synthesia in a lot of cases as well. Just because switching between languages and adjusting your AI voice over is a lot easier and less time consuming. And the results are improving a lot. So I disliked it at first, but it makes my work in some cases a lot more scalable, cost effective and less time consuming.
1 Like
Thanks for sharing! Yes - I can imagine that cost effectiveness is HUGE when it comes to multi-language projects. And as you say - the AI tools do get better all the time.
YES,look at this:https://www.dalxe.com/
1 Like
I have used AI generated voice overs, but only to show the technology on some of my YouTube videos. I believe that AI can be used effectively for certain tasks, such as creating voice overs for automated systems or for creating simple audio projects with minimal effort. However, I think that AI-generated voice overs are not suitable for more complex projects that require a more human touch.
In my opinion, the use of AI generated voice overs can be a great way to save time and money when creating audio projects, but it should be used carefully and with caution. AI technology is still in its infancy and still has a long way to go before it can truly sound as natural and lifelike as a human voice. In addition, if used incorrectly, AI generated voice overs can end up sounding generic and robotic, which can be off-putting to listeners.
Overall, I think AI voice overs can be a useful tool for certain audio projects, but I believe that their use should be limited and used in combination with human talent for the best results.
1 Like
And just a few months later…
I have been using AI voices for my online radio station set in a video game/sci-fi metaverse for years. The best voices I found (though pricey) are from WellSaid Labs from Seattle.
Recently I’ve had ChatGPT writing scripts for the news on the hour, jingles, sweepers, teasers and different commercials for companies and organizations in that universe. Sometimes these scripts have been sent to MRC to voice but more and more I’ve gone the all-AI route unless I am looking for a very specific distinctive voice like Mike, Izzy or Robin. In other words voices which early on established our brand identity.
Human touch you say? Well it may be time to fear AI is getting that right too. Here are some examples of me creating all-AI produced commercials and jingles. Here is one for a starship customization shop https://www.youtube.com/watch?v=OmVOnQIHRmo and another one where AI even sings the commercial’s song: https://www.youtube.com/watch?v=pIUGw4pRMzs
Make no mistake, AI very soon is going to disrupt everything radio related in a big way and it has allowed me to create this very niche station at a lower cost than would otherwise have been possible.
By the end of the decade I predict most commercials and jingles will be either written or voiced by AI, possibly both. I also predict that there will be a #1 pop song sung by AI and AI assisted production will be the norm for most music and audio content.
The whole AI space is rapidly advancing in a way that making further predictions as to what will absolutely require a human is hard. I didn’t think a year ago I’d be able to write a catchy song with AI which people wouldn’t know was AI unless I told them it was but I’ve done several now.
You may also have heard of RadioGPT which is an all-in-one AI solution for real world stations to provide AI generated live content based on what is going on in a local area via monitoring news feeds, social networks, etc.
1 Like
BTW: Here is the full song which was used in that commerical again completely sung by AI: https://www.youtube.com/watch?v=zRqtJvr2yVE
and here AI conducts on-the-street interviews with other AI voices in this promo for the station: https://www.youtube.com/watch?v=acctXH1qxvI
1 Like
Thanks so much for sharing @Jade - you are absolutely spot on with your comments. However we like it or not - AI is creeping up on our industry and it’s just a matter of time before the human voices will become obsolete. The examples you sent are a great showcase of what is possible now with AI. It’s scary but the future is truly here now…
1 Like
Well Izzy, you and Mike will never be obsolete on our station. What I WOULD advise though is that you, Mike and all MRC voice talent copyright their likeness so that cloning your voices will be against the law. Right now, it would be very simple to clone you and Mike based on all of that Free Jingle Friday content which an AI can train on to mimic you or Mike. It’s simple but I view it as unethical however unless you copyright your likeness it is not illegal. You may have heard that some prominent voice actors have already done this and some even offer their voices as AI voices.
In fact you might also consider creating AI clones of your voices for a totally new service in which people could do text to speech with your voice without you having to record anything. It could be a subscription service or a cheaper one time fee than having any of you get in front of a mic reading a script in the traditional, legacy way. That’s what I’d do if I were MRC.
It’s all about evolution and adapting to this disruptive technology. Blockbuster used to be a big video store chain in the USA and they dismissed the possibility that streaming would replace people going to a store to rent a movie. Today there is only one Blockbuster left, it’s in Bend, Oregon. Meanwhile we all know what happened with the growth of Netflix, Amazon Prime, Paramount+ etc.
I really hope 10 years from now you and Mike will be AI voices which I can subscribe to use. The bonus is that no matter how old you two get your AI will always be in it’s vocal prime! Much love to you two and all MRC voice talent! <3
1 Like
I couldn’t agree more with you Jade. In fact we are already working on cloning Mike’s voice - so watch this space!
1 Like
That’s great news!!!
1 Like
Hi Jade,
New to the forum. I am a film producer working on a feature film project currently in post production in the US. We need to add a slight Polish or European accent to the recorded dialogue of at least one American male voice in our film. In Leu of looping much of the existing pre-recorded audio, does anyone know if there is a software plug-in available or in beta development, similar to the BritishAlyzer Adobe Audition VST plug-in, that might help us accomplish this goal? Thank you for any leads!
Hi Mike,
New to the forum. I am a film producer working on a feature film project currently in post production in the US. We need to add a slight Polish or European accent to the recorded dialogue of at least one American male voice in our film. In Leu of looping much of the existing pre-recorded audio, does anyone know if there is a software plug-in available or in beta development, similar to the BritishAlyzer Adobe Audition VST plug-in, that might help us accomplish this goal? Thank you for any leads!
Hi Isabella,
New to the forum. I am a film producer working on a feature film project currently in post production in the US. We need to add a slight Polish or European accent to the recorded dialogue of at least one American male voice in our film. In Leu of looping much of the existing pre-recorded audio, does anyone know if there is a software plug-in available or in beta development, similar to the BritishAlyzer Adobe Audition VST plug-in, that might help us accomplish this goal? Thank you for any leads!
I hate to break it to you 10 years later but that…that was a very well executed April Fool’s Day joke by Mike. 
That level of AI just did not exist in 2013 and I am not even sure it exists now. Though I would not be shocked that if you ask this again in 2033 there will a plethora of choices 
Are Voice Artists Going To Be Out of Jobs Soon?
No, in fact I feel real human voice overs will be valued even more so.
There will always be those that undermine their product with $5 logos and AI voiceovers. But that is ignorance. Penny wise but pound foolish.
Don’t let the AI fear mongers fool you. AI can be felt miles away.
Also, just because it’s human doesn’t make it great. I think people putting on “announcers voices” will also (hopefully) go away in favor of voices that sound natural and authentic.
Quality is always rewarded.
1 Like
Well you may want to check out ElevenLabs: Free Text to Speech & AI Voice Generator | ElevenLabs - you can clone your voice and then speak many different languages so I guess our April’s Fool is no longer so funny 
As a VO and Radio Imaging/Commercial Producer of over 25 years, I remind all talent to ALWAYS read the fine print on ANY web site that you choose before uploading a “demo” for cloning puposes. Make sure that you, as a VO are protected from that site/company from using your voice (or a clone of it) in the future without paying you.
With social media sites making AI generated voices “the norm”, it’s becoming acceptable to “hire” AI or cloned voices.