Audition Podcast Breathing fix

[gianlucas99782867]

Hey guys,

I am currently editing a 2-hour long podcast. I worked with 4 mics and have got a mixdown file for all of them that I am using to edit. I have matched the loudness to -16LUFS and have a horrible amplified breathing sound sometimes before people started speaking. Is there any way I can get rid of that? Thanks in advance

[SteveG_AudioMasters_]

This invariably happens when you have some sort of automatic level control, or compression that's been over-applied to a track; the level gets pulled up to try to compensate for the silence that's happening between words. Yes I realise that this is a particularly useless thing for it to do, but that pretty much sums up what's happening...

How did you achieve the loudness matching? Also, that doesn't appear to be rather a low number compared to the -23LUFS or thereabouts that most broadcasters are using. If your original mix sounds okay, then it's whatever is happening after this. If the individual mic contributions are okay, then it's something that's happening at the mixdown point. Unfortunately, if the breathing sounds are nearly as loud as the wanted audio, then you have no option but to reduce them manually - there's no way that any software can distinguish between wanted and unwanted sounds at the same level.

[gianlucas99782867]

Thanks for your reply. I have found that while in Waveform, the audio is completely fine. It's once I put it into the podcast preset in the multi track form, the levels get messed up. So I imported the edited wave form into a new multi track sequence without preset, and that seemed to have fixed it.

Thank you for your reply

[SteveG_AudioMasters_]

All I can say, without saying too much, is that I would not recommend the use of the podcast preset under any circumstances!

[gianlucas99782867]

yeah, it has been causing me a lot of headaches. As a novice with audition, could you shortly explain why I shouldn't use it, please?

[ryclark]

Basically because it applies a lot of overdone effects to your audio by default which aren't really applicable in most cases as you have found. At least not without making a whole load of re-adjustments to the FX settings to match your audio. So you might as well start from scratch and only do what is necessary for your recordings.

[gianlucas99782867]

yes! Thank you so much for this.

[Trevor_Asq]

-16 LU does seem to be as close to a standard for podcasts as there might be. You need to be hitting this or any external adverts/interstitials will be considerably louder.

I‘ve had good results mixing to -23LU (aiming at PPM 5.5/-9dBFS for peaks - with a brickwall catching anything above -6dBFS) as normal then using a -16LU Loudness normalisation for final export. Maths say that my brick wall should be at -7.1 dBFS but nothing distorted so far.

The native Loudness Radar plug-in is good for monitoring LU, also zplane PPMulator has good LU options.

[Trevor_Asq]

The other part of this workflow is that the producer makes the edit for content/time using the raw files whilst the recordist concurrently does a balance mix for the whole recording.

The resulting exports are relinked in producers session for final review, music balance & export. 

Wouldn’t work for everything but it does for a simple speech based multi-contributor podcast.