Best Practice for Volume Normalization of entire library

[jfkaess]

I have about 12,000 tracks, some are mp3, some are 256 - 320 kbs AAC, and some are ALAC. I have used registered dbpoweramp to rip all of them over the years. I currently have been using itunes and an ipod classic for playback. I have used a program called ivolume (http://www.mani.de/en/ivolume/) to analyze all my files for using the soundcheck option in itunes and on my ipod. I currently have registered versions of dbpoweramp v15 and perfecttunes 1.5.

I would like to change the actual volume level of all my files (mp3, aac and alac) so they will be consistent levels instead of using soundcheck so I can move to a non-apple music player (maybe a Pono).

What is the best practice for adjusting the actual playback level of all my files? iVolume will allow me to delete all the soundcheck settings in the tag for each file. After doing that, how do I use a dbpoweramp dsp or converter to adjust the playback volume on all these different file types with, and which one and with what settings and what is the process? I also have seen something about R128 vs Replay Gain. Which should I use and what is the process and will it work on all my file types. This is going to be a lot of work, so I'd like to just do it once.

[Spoon]

Re: Best Practice for Volume Normalization of entire library

You would not need to remove the sound check, to alter the volume all the tracks have to be converted to a new format, with a volume normalize DSP effect (set to replaygain).

You would keep the original files and create a 2nd library based off that with the volume altered, because changing the volume means you cannot go back to the original files.

[garym]

Re: Best Practice for Volume Normalization of entire library

If the player you will use can read the ReplayGain tags, then just add those to all your files rather than permanently changing the volume of the files. Many players can read RG tags. To add RG tags to your files, just load them up and use batch converter to convert to REPLAYGAIN DSP (not the REPLAYGAIN(apply). Have it add ALBUM and TRACK RG tags. And you can leave the SOUNDCHECK tag value as both RG and Soundcheck can coexist.

[Porcus]

Re: Best Practice for Volume Normalization of entire library

I agree with garym, try to tag the files rather than altering them.

If you have to change volume, then http://mp3gain.sourceforge.net does this losslessly for mp3 files.

(OK, it is only nearly lossless - you can get errors if you change by a couple of hundred dB's, but hey ...)

[jfkaess]

Re: Best Practice for Volume Normalization of entire library

My goal is to actually alter the files, not add tags to them. I want my files to play with volume normalized regardless of the player I use.

Here is the solution I have come up with:

For all my mp3 files, I am re-ripping these CD's to 320 kbps AAC with r128 volume normalization using DSP Effects 11 beta so that they are set to -18 lufs instead of -23.

For all my AAC files I am also re-ripping these CD's to 320 kbps because I had previously ripped them to 256. I am using the same r128 volume normalization.

For my ALAC files, I am just using music converter (or batch comverter) to convert them using r128 volume normalization to ALAC files and then replacing the originals with the converted files.

I am using 320 kbps AAC because I am completely unable to tell any difference between these files and the CD originals (and I have some pretty decent audio equipment and headphones). My reason for originally using mp3's was because AAC and ALAC were proprietary. Now that they are public, there is no reason not to use these formats which are superior to mp3. I choose not to use FLAC because I am currently using Apple products, but even if I change later to Pono or another player, the Apple formats will be supported now that they are public, while Apple will likely never support FLAC.

[garym]

Re: Best Practice for Volume Normalization of entire library

My goal is to actually alter the files, not add tags to them. I want my files to play with volume normalized regardless of the player I use.

Here is the solution I have come up with:

All fine for your purposes. But I would advice you to figure out if your work flow is modifying the volume using TRACK RG value or ALUBM RG value. I'm not sure which one you would prefer (although I assume if you are doing lots of playlists that mix albums you'll prefer TRACK RG). But with TRACK RG, you'll then permanently lose the intentional difference between quieter and louder songs within the same album.

p.s. But if you're reripping everything anyhow it seems a shame not to create a lossless archive with *no* volume normalization, only RG tags (use ALAC or FLAC or any lossless version). Then you can create the lossy (AAC) versions of everything using permanent volume normalization. And this way if you change your mind in the future, you'll have the lossless to play or to use to create different sorts of lossy files (i.e., some new and improved lossy encoder comes out). Storage space is cheap but the time to rip and tag your CDs is not something you want to waste.

[Porcus]

Re: Best Practice for Volume Normalization of entire library

I am re-ripping these CD's to

You asked for "Best Practice". Once you rip files for further modifications, then best practice is to rip to lossless.

That said, there are near-lossless ways to alter the volumes for MP3 and AAC. I mentioned mp3gain, but five seconds on google reveal that this is possible for AAC too: http://aacgain.altosdesign.com .

So to wrap it all up, your solution creates lots of extra work just to do something that is way off best practice.