FFAStrans forum

Hello

I am currently having an issue understanding audio conform setting within the encode processor.

Using the current workflow:
I have a watch folder that watches a s3-to-smb share that initiates a copy from s3 compatible storage to job workfolder, does a volume analysis to table & creates a report file, does an encode with a volume conform to -24, does another volume analysis and appends to the report and then pushes back to s3 compatibile storage.
Report thats made by workflow: Heres the issue I'm facing:
When I run the command: on the output file I get the following output: Which at at the bottom shows: "mean_volume: -31.5 dB"

However, the report that I attached and linked above, show that the channels are all conformed to -24... How is that?
Should "mean_volume" be -24dB if I have conformed volume to -24?
Or do I not understand what conform volume means?


This is the type of file I am running through this workflow:
Regards,
Lue

Hi lue,

from my understanding we analyze the ebur128 program loudness into your variable loudness_pre. If you set Conform volume to -24, the encoder will calculate the difference between the current r128 loudness and -24 and raise/lower the volume accordingly.
Not sure how much the output of volumedetect filter would be influenced by an r128 correction.

What exactly is your goal? Are you not working against r128 specs?

I suspect the volumedetect computes mean volume differently than ebu 128. I ran an EBU test file through your workflow and it definitely normalized to the level in the encode block. This is a pretty good free plugin (I used it in audacity) so you can see levels measured to EBU specs https://www.tbproaudio.de/products/dpmeter.