Audio distortion inside same track

Yello everybody,

my very first posting here, so hope nobody mind if add extra complementary here, like how nice it is to meet everybody

My technical challenge.
I have video (sorry if this is not most accrue term) on my editing table, it is filmed as one time shot.
Talent behind camera is same all the time, also is wireless mic he is using. For some reason which I cannot comp out, I have distortion every time when talent start new scene.. Root cause is naturally anatomy of microphones, but since damage done (talent has left building) it would present me lesser dim of light, if I could found way to fix this on my editing table.

So basically I have one audio tracks on my editing table, which have very different levels on audio. Audio is just fine when talent has been speaking couple of second or some, but when ever he start some new sentence after ten second silence, I have this seriously exception on my ears... I have try apply basic level fix's, like using EQ, add "Voice Channel" etc, but I just cannot get those scenes to equal levels.

How long is the take?
Can you post an original audio sample to hear and see what is going on?

That sounds like you were using automatic gain control rather than manual on the camera when recording. It is likely that the distortion is baked into the audio. Rather than a compressor you could try using a limiter to even out the levels.

Hello Robert, hello Charles!

Sorry I cannot share much of my nice tries. Not since lack of any member on this board or any, simply because environment where I'm working, is ultra regulated. Talent is health care professional and audience is limited to very same science. If I share any sentence on the air, I would be marketing medicine to consumer, which is grand scale no no..

Technically.
HW stays same all the time.
Talent present sale speech, sentence after another. For reason which is so unclear to me (HW origin most obviously), is why last sentence is so out of this planet.
Hope pic explain a little better. From start everything goes just fine, talent speech sentence after another jaadi jaadi.
After green bar, there is some general noise, since talent did lost his nerves (some old school chap) and there was three dude(tte) behind camera together to ask him to just to cool down and keep performing...
After red bar, talent start another sentence like all another ones before, but audio level is just so very different with high echo etc...

Anyhow, I was able to put something bearable together by cutting that last high volume sentence to tiny fractions and slide volume from down to up.

https://aijaa.com/n5Igls

Hello Robert, hello Charles!

Sorry I cannot share much of my nice tries. Not since lack of any member on this board or any, simply because environment where I'm working, is ultra regulated. Talent is health care professional and audience is limited to very same science. If I share any sentence on the air, I would be marketing medicine to consumer, which is grand scale no no..

Technically.
HW stays same all the time.
Talent present sale speech, sentence after another. For reason which is so unclear to me (HW origin most obviously), is why last sentence is so out of this planet.
Hope pic explain a little better. From start everything goes just fine, talent speech sentence after another jaadi jaadi.
After green bar, there is some general noise, since talent did lost his nerves (some old school chap) and there was three dude(tte) behind camera together to ask him to just to cool down and keep performing...
After red bar, talent start another sentence like all another ones before, but audio level is just so very different with high echo etc...

Anyhow, I was able to put something bearable together by cutting that last high volume sentence to tiny fractions and slide volume from down to up.

https://aijaa.com/n5Igls

Hmm, you said a wireless system was used.
My guess it is some bad settings in the transmitter / receiver.

For example the Sennheiser EW100 system does have a pilot tone function:
"The pilot tone has an inaudible frequency that is sent from the transmitter and evaluated by the receiver. It supports the receiver’s squelch function."

Should be normally set to "ACTIVE".

And then there is the squelch threshold:
"If you set the squelch threshold to a very low value, a very loud hissing noise can occur in the receiver. This hissing noise can be loud enough to cause hearing damage or overload your system’s loudspeakers."
The function of this circuit is to mute or silence the audio output of the receiver in the absence of the desired radio signal. When the received signal strength falls below this level the output of the receiver is muted.

"Ideally, the squelch level should be set just above the background radio noise level or at the point where the desired signal is becoming too noisy to be acceptable. Higher settings of squelch level require higher received signal strength to unmute the receiver."

I found most of the time it is better to set the threshold to high.

There also could have been some other transmitter in the room messing with the signal...

So it could be that the receiver was cranking up the volume while talent did pause and when talend started to speak it would not dial down fast enough.

It could also be an auto level setting in the camera messing with the volume at that point.

All that is just an educated guess by me as I was neither their nor could I hear your audio.

If you still want me to have a look at it, you can contact me in private and upload a sample to my company's private upload space. I'll have a look for analysis and will delete the file afterwards.

BTW two things:
1. You might have a look at deepl.com for a better translation from your mother language into English
2. Instead of chopping those high volume sentence into fractions, you could use key-framed fader settings.

Thank you Charles, thank you Robert.

I was (think so) able to comp this out, beginner mistake I did perform, was item you both mentioned, autogain on camera..

Lesson learn, thank you again both!

I've recently been working with Izotope RX9 on some really bad audio (I recorded it with a cheap cardioid mic). It had wind noise, traffic noise and distortion. I used it to eliminate and repair the unwanted noise, which was fantastic compared to the original sound file. I then used the effects: modulation, pitch and vocal channel to further enhance sound. I hope my suggestions help!