Noise floor issue when combining more than one audio file

After finding this forum I was very happy to find out about the Audiobook-Mastering-Macro! It seemed to have solved all of my issues!

Files were created, passed ACX check and everything was great, until… the author I recorded the book with requested that instead of having the files broken up in to sub-chapters (CH 1.1; CH 1.2; CH 1.3, etc) he would like all of the CH 1 files amalgamated.

Easy enough, right?

Once I added two files together, all of a sudden the noise floor was not registering anymore and was too low/ unnatural.

My brain hurts trying to understand this… any ideas why the larger file won’t pass anymore? By larger I mean they range between :45 - 3:00 each so together they are less than 10 minutes.

I further went to the extreme of recording my studio empty for 5 minutes and threw it underneath, and it still didn’t pass.

I have used this PC for zoom a couple of times. Maybe something in there is messing me up?

Doesn’t explain why the files pass and then don’t when strung together. I have created much longer files that did pass using the same audio.

To delete the unwanted noises and breaths, I used punch & copy/ paste so that the background was consistent.

Looking for any suggestions!

You probably added some digital silence between sections.

The only digital silence added was generated at the very beginning and very end (as per the audiobook requirements) using the silence generator.

The rest was punch copy/paste or the ambient room room noise from the recording.

WONDERING IF MAYBE THERE IS SOMETHING IN THE MACRO THAT IS KEEPING MY NOISE FLOOR SO LOW? I don’t know much about them (macros)… followed the instructions and it just worked like magic.

Check for any automatic normalization processes that might be affecting the volume levels. When using copy/paste for editing, be cautious of introducing artifacts that could trigger problems during the ACX check.

That’s NOT what the ACX requrement says:

You should be able to generate some pink noise at around -70dB (or maybe lower… I’m not sure what the minimum is) and mix it in. It shouldn’t have to match the actual room tone to pass.

If you know where the absolute-silence is, you can just mix it in there but it’s probably OK to mix it into the whole file.

Thanks, however my author is not publishing through ACX, he is going with Author’s Republic — and they DO ask for “silence”:
Contain 1 to 5 seconds of silence at the beginning and end of each track.
(Maybe I am interpreting this incorrectly?)

How might I generate this pink noise? Is that a feature of Audacity?

That sounds like a great alternative!

What DVDdoug said. ACX uses “dead silence” as an evaluation trigger. If you have dead silence, it’s a good bet you also used other filters, effects, and corrections that can negatively affect the quality of the performance.

Odd distortions are distracting and ACX says very clearly, they hate distractions.

There’s another note here, if you can pass ACX conformance, you can successfully post anywhere else. So don’t poo-poo their process.

You don’t have to generate pink noise. You can use actual Room Tone. Instead of jamming Generate Silence here and there, record several seconds of microphone with no movement or sounds from you. That should match perfectly all the rest of the background sounds in the performance.

If you don’t have a quiet, echo-free room, that can start a cascade of patching, correcting, and manipulating. That room is a really big deal.


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