That’s “acting” and there’s no effect for it.
There’s AI voice production tools that can simulate that, but fair warning, there’s no turnoff more efficient than a “perfect” announcer mispronouncing a normal English word.
I recently encountered a product promotion where the AI announcer mispronounced the name of the product.
I was not looking for an easy effect. But in analyzing the voice of a person who is excited vs. bland, there is a pitch difference and speed difference. I was wondering if anyone tried to play with the tracks to simulate the difference.
It does not have to be perfect.
" Upward Inflection is a change in pitch from a lower to a higher note specifically in a vowel. Most often, this change in pitch indicates questioning, insincerity, surprise or suspense."
What I would need to do is change the pitch at certain points in the speech. Quite a challenge.
Correct, but the voice doesn’t “switch” between those effects. It slides gracefully and it’s the graceful sliding that’s super difficult.
If the show is long enough, you could be hiring a voice actor to re-read it. If it’s too long for that, then it’s a retirement project. You’ll be applying effects when you retire.
I wasn’t kidding about getting an AI voice production tool to re-read it.
As a voice actor AND audio engineer for 10+ years: There’s NOTHING that’s gonna be worth messing around with. By the time you invest the time, tools, work, effort and thousands of miniscule adjustments to make it kinda sound somewhat passable you could have had the original narrator (or a soundalike) re-read it and be done with it.
Even “cleaning up” a 30-second segment like that would take HOURS of work.
Natural is not an issue. It is a voice in a pinball machine sound track. The original recorded voices and music are highly compressed - 1990 technology. If the people who own the copyright have the original recordings, they are not releasing this. I believe that they cannot be cleaned up as they are distorted from the compression.
Since it is for my own personal use, playing with it should not be an issue.
I am using a movie soundtrack of the actor (who is now deceased) and finding many of the same lines, but not the original levels of inflection.
I can just edit and use the movie voices. Or I can play with them to make them sound better.
Though it’s a difficult one. You MIGHT look into Melodyne, but the easiest way to do that may be to hire a soundalike.
As a voice actor, we do get called in to handle that exact thing: the original voice actor can’t/won’t reprise the role, so they select soundalikes. It may not be a road you’re willing to go down, but I did want you to know it’s an option.
I have looked, briefly, into AI. That is a pretty deep rabbit hole, as far as I can tell. But the sound snippets that I am getting from the movie(s) would be useful for training the AI if I decide to go in that direction.
“As a voice actor, we do get called in to handle that exact thing”
That would be another option - thanks for that. Not an option I had thought of.
A very interesting project indeed! My favourite pinball table at the time was “Stones 'N Bones” from the game “Pinball Fantasies”
Re-recording is the best option, I totally agree.
@LexVAPin
I’d like to add an idea:
Record the lines yourself (or get someone from your family or friends to do it for free).
Just to test if it’s worth hiring a voice actor. And it might turn out that a professional voice actor is not necessary at all, as it is a personal / private fun project.