Peak Level Value Report for multiple audiofiles

Hi everybody,

I am looking for a way to make WL7 display the peak level values of several audio files as numeric values in a large report etc…

I will explain it in more detail.

Suppose that I have several hundreds of individual audio files and want to determine the respective peak level value for each of them.
Off course, I can do it by hand…when I open each file individually with WL7, retrieve the peak level value via the “Normalizing” or “Global Analysis” dialog, and put that value into a text file / table etc.
In this list/table the file name along with its corresponding peak value would be recorded.
Then I would have a nice overview, and this list could also be useful for other purposes (statistics, graphs, etc.).

Unfortunately, I don’t have the time and perseverance to accomplish this rather complicated procedure.

So, is there a way to accomplish this via a batch job?
Just measure the peak level value of a file and then automatically transfer this data to a text file?

I mean, the data itself is already there anyway… I just would like to have it in another form and as a “bunch”.

This must be possible somehow… or should be implemented quite soon.

I hope you can help me here with this problem.

CU

P.S. the only tool that can, to some extent, do this job is the TT Dynamic Range Meter Offline Tool …, but unfortunately that only works on 16Bit/44,1 KHz, mono/stereo files…any other format than the “CD-Audio” compliant one is denied. :frowning:

This is currently not possible. Batch analysis report is something I have in mind, for the future, though.

In your case, what is the final use?

Hi PG,

i have a sound library with several hundreds of files in 24Bit/96kHz quality. After checking the files manually with a spectrum analyzer i found nothing above 22-23 kHz. It’s just upsampled stuff to advertise with the higher samplerate…i guess…hehe.
So i decided to convert the samplerate down to 44.1kHz to save some space and computing power.

But the problem is here.
When converting samplerates you often have to deal with clipped audiofiles after the conversion process.
So i have to lower the peak levels of some these files by max. 1-2 dB (before the conversion process) to avoid the clipping.
Unfortunatly, not all of the files are normalized to -0.1 dBfs or similar, most of them have levels around -2 or -3dBfs (the ones i checked manually so far).
I just need to find out which files have the highest peak level. These files then have to be preprocessed to prevent any clipping from the samplerate conversion process.
With this report it should be quite easy to identify all of these files and preprocess them isolated from the rest.

I think it shouldn’t be necessary to preprocess all files just because only a few are at full scale level.

Of course, i can just lower the level of all files by 1 or 2 dB prior to the samplerate conversion, but this involves a little quality reduction for all files.
Also i have to think about the final bit resolution, since the level adjustment process is a 32bit FP operation.
All this processing then makes me think about truncation and dithering…but that’s another story…hehe.


Btw. such a batch analysis report function might be useful for EBU R128 compliant level matching purposes and other things i don’t think of now.


CU

When converting samplerates you often have to deal with clipped audiofiles after the conversion process.

This is why I recommend adding the Peak Master plugin after the resampler, because it will remove the possible over-peaks while keeping the rest of the signal strictly 100% the same.

Philippe

I would also use a limiter, as PG suggested.
Alternatively you could use the demo version of Nugen Audio’s LMB:

http://www.nugenaudio.com/LMB_Loudness_batch_processor.php

It can give you a log of the audio data, including True Peak Level.

THX PG and LutzR for your hints.
I will try it with a limiter behind the Resampler to avoid clipping and leave the rest of the files untouched.

Also i have checked out some processing tools like the before mentioned Nugen Audio LMB, Grimm Audio LevelOne and the opensource software R128GAIN.

The main problem with Nugen Audio LMB is that you need an iLok just to try out the demo. And i don’t have one of these…hehe.
Grimm Audio LevelOne is completly useless because the demo version doesn’t show any results of the analysis it does.
It writes PDF reports+BWFv2 tags for each file, but they’re also useless because all reporting functionality is deactivated in the demo version. This is the most useless demo i’ve ever tried…hehe.

R128GAIN somehow works a bit better since it shows the peak level/true peak level in a dos-command window while it works, but there’s no way to get these infos into a nice text format or similar.
It might be possible…but not with my non-existent programming/scripting skills…hehe.

I also checked out the opensource software SoX, but this thing is way to much for me as it needs heavy scripting skills to get it working the way i want it to…hehe.

So to get to a conclusion…
I going to make it the limiters way.


CU

DonQ-Juz

P.S. Also i hope that PG is really considering all this batch analysis+report stuff to be included in WaveLab 8…he will surely blow away all these overpriced processing tools mentioned above. :wink:

Hi everybody…

I found a new and easy to use solution for (my) problem.

Foobar2000 with the Dynamic Range Meter addon/extension.

This nice little addon implements the algorithms of the DynamicRange Meter we all should know…hehe.
But the best is the fact that you can use it on all file/audio formats that Foobar2000 can play…24Bit/96kHz…no problem at all…and they’re both free of charge.

The DR Meter plugin does an offline scan of the highlighted files in your playlist and puts a text report file into the corresponding folders.
In this report file you can find the Peak and RMS values of every analysed track nicely arranged in a table…plus the official DR values…if you need them.

Here are the links to Foobar2000 and the Dynamic Range Meter plugin/addon/extension

http://www.jokhan.demon.nl/DynamicRange/

I hope some of you find these infos useful.

CU

DonQ-Juz