Sharp Recurring Background Noise

Hi Everyone,

I was recording for a podcast episode, but you can hear my neighbor’s AC unit dripping on mine in the background. I tried noise reduction but it massively distorts audio.

Any advice?

Post some of the noise so we can analyze it.

Couldn’t you hear the noises in the headphones?

You should have a good pair of wired headphones connected to your microphone or interface during production. You can’t connect them to the computer and you can’t use wireless headphones.

Screen Shot 2022-05-23 at 19.19.28.png

I could hear it but unfortunately this was the only place and time I could record. R

aw audio attached.

Thanks for the sample.

I can make it a little different, but not any better.

Audacity thinks that dripping is a performer in your show. Audacity has no good tools to separate individual voices, instruments, or sounds from a mixed performance.

There is a Hollywood way out of this, but it’s not fun. Write down the words in script form and present it again in a quiet room. If you can’t get the guest to perform again, find someone with a similar voice to act the part.

In the case of a damaged video (or movie) they sit in a quiet studio while watching the noisy performance and announce the clean work while matching the lips moving.

I’m not kidding.


Sorry to say, but that recording was pretty bad.
Not only was the dripping very loud, but extremely broadband.
Lastly, some of the dialog was clipped, whilst at other times, much softer, almost like the person speaking was not keeping
a constant distance from the mic.

The noise can be reduced but at a cost, reduction in the vocal quality, although not as bad as running it through the standard noise reduction.
I have attached a cleaned version, as can be heard, the dripping is greatly reduced but will be impossible to completely remove.
The best that can be hoped for, is make it less objectionable.
The dialogue just needs some leveling and some EQ to get rid of some of the high end to match the OP’s taste.

The only other solution is to do what Koz suggested in the previous post.

BTW, to make the file smaller, I have made it a mono file, no need for stereo for dialogue.
The OP can then duplicate the track if need be.