Recorded voice sounds robotic/tinny?

I’m editing a podcast recording- but the voices all sound oddly distorted/robotic. I don’t think it’s clipped, and I raised the rate to 4.8 (i think?) but there’s still some robot-ness. I’ve tried playing around with a lot of the settings, and the only other thing that helped was pitching it up a little bit? The voices are closer to their natural levels, but there’s still some distortion that I’m not sure how to fix.

It sounds like you’re too far from the microphone. Are you using a separate mic? The mic built-into your laptop?

Maybe you only think you’re recording from a separate mic… If that’s a possibility, scratch on it with your fingernail to make sure you’re recording from the correct mic.

This was recorded in a studio on two separate mics- we tested both and they both work and were being picked up from the correct source. Is this fixable at all? Two episodes were recorded this same day, and they can’t really be redone because they were interviews with guests.

I don’t recognize that distortion.

Was the original recorded as two separate microphones on two separate tracks, say one voice on the left and one on the right of a stereo track? Then in post production, mixed down? It’s stereo (two tracks) but they’re identical.

How did you get there from two microphones?

Do you still have the two original separate microphone tracks?

This clip fails the Koz Test. If somebody wrote me a big check and said I had to do this intentionally…I couldn’t do it.

I think you’re down to Desperation Methods: Re-record it, or type out the words and get AI Voices to read it.

Koz

How far apart were the two performers? Were the backs of the two microphones pointed at each other?

Koz

I’m not sure- the setup is at a local library. It’s two mics, but they record on the same track. I don’t know how to get them separate in the first place.

The mics weren’t facing the same way, but their backs weren’t facing each other. I’d say each one was maybe 3, 4 feet apart?

Unless somebody else posts, I think this is the end of the ideas.

Koz

I would contact the library and tell them of your troubles. They may have something broken and be continuously churning out damaged recordings.

Koz

It sounds like an electronic delay, rather than an acoustic one.
I can re-create it …
[before-after]
But I can’t fix it.
Maybe AI can fix it ? …

So there’s the answer to all those “Help Me Clean Up” posts that usually fail.

Echoes are usually fatal because they are the performer’s own voice arriving at the microphone multiple times. So you are, in effect, asking the solution or product to remove the performer from themselves. Now, if you generate a solution that “knows” what a human voice is…

That kind of thing is most unusual in an electronic pathway. That’s why I wanted to know how they got from two independent voices to a single two-channel mono show.

Some Thing Is Broken.

Koz

This topic was automatically closed after 30 days. New replies are no longer allowed.