Deleting Breaths Out Of Podcast

After recording my podcast and going back to edit it, I notice that the recording certainly picks up times I inhale. I’m to the point I can see it on the graph doing back through to do my editing. Since that is a monotonous thing to listen to on a podcast, is there a way to highlight one of those breaths at the beginning of the podcast and then use a command to have the software go through and delete all the breaths that are evident which I don’t want to go out on the final production?
Thanks for your help.

tngolfer

is there a way to highlight one of those breaths

Nice try.

I think the best we can do is make manual pasting and covering-up easier.

This is one of the Punch tools. Theory has it you can “punch copy” some background sound and then select the “evil sound” and punch the background sound over it, covering it up. That’s basically it. You would think you need to carefully manage the injected sound and juggle timing and all, but no. It automatically configures the patch to the select evil sound.

Have you tried different microphone positioning and orientations? Are you asthmatic?


I poo-poo’d the idea of global breath removal, but it is possible to set up Effect > Noise Reduction and try it. The problem is going to be deleting some normal vocalizations, too, by accident.

Drag-select a breath (and nothing else). Effect > Noise Reduction > Profile.

Select a whole chapter and Effect > Noise Reduction > choose settings > OK.

The audiobook people have a nice, gentle Noise Reduction of the Beast (6, 6, 6) and you can try the Audacity default settings. 12, 6, 3.

I think there’s a breath removal in the Audiobook notes. Looking.

Koz

https://forum.audacityteam.org/t/breath-removal/53801/7

I think this is still valid.

Koz

https://forum.audacityteam.org/t/de-breath-plug-in/50956/1

More Notes

Koz

The trick is to use the (bubbly) Noise Reduction version, (with breaths as the noise-profile), as a template for an envelope-follower, (like Steve’s Dynamic-Mirror.ny), applied to the original audio.
So the breaths are selectively attenuated, but you don’t have the digital NR artefacts in the final product.

The trick is to use the (bubbly) Noise Reduction version

Or put up with lesser reduction and no bubbly damage with the 6, 6, 6, settings. That’s why they were chosen. Nobody can hear them working.

And in case we didn’t say yet, if anybody gets this to work, do post back. This is a Forum, users helping each other, not a Help Desk.

Koz