RMS normilize

emmrecs wrote on 5/16/2018, 10:54 AM

Hi, welcome to the Magix forums.

First please read this post and give us all the information requested there.

Second, you need to understand what the Normalise process actually does. To Normalise an audio file means to set a level, chosen by the user (you), which is to be the "peak" (="loudest") sound of the whole file. So, if you set the Normalise level to be -3dB, quite a common setting, the "loudest" peak of the file will be at -3dB. If the file previously had levels that varied between -30dB and -10dB, the normalised file will now have levels between -23dB and -3dB. In other words the dynamic range (the difference between the quietest and loudest sounds) will still be the 20dB that was present in the un-normalised file. OTOH, if the file previously had levels ranging from -15dB to 0dB, after normalising to -3dB it will now be between -18dB and -3dB, i.e. subjectively "quieter".

But, from what you write, I wonder whether you actually need to keep the range of levels within your file between -18 and -23? If so, that is quite unusual because it is quite a narrow dynamic range (only 5dB difference between your "loudest" and "quietest" sounds).

Your mention of -18 and -23 makes me wonder whether you have been given a "Target Loudness" figure to achieve? if so, Normalise is most definitely NOT the process you need to be looking at.

If you come back with all the details about exactly what you are trying to achieve (along with the details of your computer and audio set-up) we may well be able to give you more advice.

Jeff
Forum Moderator

luke-castle wrote on 5/16/2018, 11:02 AM

Yes I need to meet ACX QA process Wich is for audiobooks the link explqnis everythibg https://www.google.com/url?sa=t&source=web&rct=j&url=

emmrecs wrote on 5/16/2018, 11:18 AM

Hi.

Thanks for the link. Various things strike me about that specification, such as noise floor, consistency of sound, etc., etc, but I also noticed "-3dB peak level" so that immediately tells you that you must normalise your file to -3dB (I assume SF 12 has that facility).

As to meeting the requirement for the "-18dB to -23dB RMS", that is the "Target Loudness" they expect. You need a user of SF to guide you on this process, which I am not. I think you mentioned in another post that you have iZotope software; which precise program?

Jeff

luke-castle wrote on 5/16/2018, 11:20 AM

I can use izotope but I prefer not to, however i am not opposed to using it

emmrecs wrote on 5/16/2018, 11:28 AM

Unless you say which iZotope program you have it is impossible to know whether it is likely to be helpful to you to use it!

luke-castle wrote on 5/16/2018, 11:30 AM

I am so sorry I am new to all of this ozone 8 elements

emmrecs wrote on 5/16/2018, 11:40 AM

Hmm... Ozone 8 Elements is a cut-down version of the full Ozone 8 and i can't remember which "features" of the full version are absent in the Elements version. I have to go out now, but if you are able to list the constituent features of Ozone 8 Elements it should be possible to suggest one or more of them which might be useful to your needs.

luke-castle wrote on 5/16/2018, 11:43 AM

Sure here is the link to the feature list

https://www.izotope.com/en/products/master-and-deliver/ozone/features-and-comparison.html

rraud wrote on 5/17/2018, 11:42 AM

After you apply the 'necessary' processing, EQ, compression/limiting, use the 'Statistics' process to obtain a read out of the peak and RMS values. I would use peak normalization to insure your program audio does not exceed the stated -3.0 dBFS peak.

Though it probably will not apply to your submission, most broadcasters these day use the LU/LK loudness scales to determine appropriate levels (ATSC -24 LUFS) . Most non-broadcast spoken word material is recommended to be around -18 LUFS (+ or - 2, integrated).