Skip to main content
forums, blogs, song critique, support & more...

Volume Normalizer for wav Files?

Hi guys!

Some of you may have heard of a program called 'mp3 Gain', which allows you to balance the volume levels of your mp3 files so that they're all roughly the same. It doesn't remove the dynamics within each track (or each album even), but basically just allows you to control the general playback level of each file. Very handy tool, and I've been using it for years.

However, I'm looking for a program that does the same thing for wav files, as mp3 Gain (unsurprisingly) only works with mp3 files. I'm just looking for something simple, where you can load up your wav files, then set the desired 'average' volume level in decibels (or something similar). My multi-track recorder has a hard time mastering songs to an acceptably loud volume, and converting to mp3 causes too much loss in sound quality (so I can't use mp3 Gain).

I had a friend in Australia who used to casually master tracks for me (using Pro Tools), but he's in a bad state at the moment so he can't help. I know some amateur audio engineers who could help, but I want to do it myself. I do have Reaper, but I want to avoid mastering in DAWs because I like the objective standard of the decibel measurement in mp3 Gain, and am hoping to use something similar.

Thanks heaps in advance for any help you guys can provide!

Tags

Comments

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Sun, 02/08/2015 - 18:21

paulears, post: 424828, member: 47782 wrote: This is my process too - but I think he really wants each track to be the loudest it can be? I'm not sure why. The example of the solo guitar being as loud as the body tracks is in my humble view not the right way to do it. When I sit in the studio, I rarely adjust the volume. Each track lives at it's appropriate level to me. With 24 or 32 bits of depth to play with, and high bitrates, normalisation to squeeze the last drop seems counter intuitive. Loads of players give the option to up the average levels so that with your earplugs in, on a noisy train, even the quiet tracks sing out - but this is NOT how it should be, and isn't how mastering engineers do it. They listen to all the tracks and adjust each one to the next. I treat it like I mix live music. Some tracks shake the room and need to, but others hardly tickle the meters and are the ones where volume is not critical at all.

If you have recorded similar songs, content wise, then they should already be very similar. If you need to nudge the master up and down more than a tiny bit, then I'd suggest something inconsistent happened. I do see that you could end up with very misadjusted volumes if the tracks were recorded at different times, or maybe in different rooms etc - and then adjusting the volume between individual tracks is perfectly sensible - but if the songs are different, but just similar in content, I still think this adjustment is an aural thing and NOT something for normalisation to do automatically. Normalisation is maths, not music.

I've lost count of how many times I've said that I want the LOUDEST track on the album to be as loud as it can be - NOT all of them. I want to RETAIN THE (potentially large) DYNAMICS between the different tracks in the album, by applying the same amount of gain boosting to each one. I'm doing this because the Tascam's mastering tools usually don't master the tracks loud enough at the final stage.

I'm aware of the detriments of the loudness war, and how mixing as quiet as possible for as long as possible is the best way to ensure fidelity. But if I had a dollar for every mix I've handed to band members, only be told "Dude! It's waaaay too quiet!", I'd be able to buy an SSL desk. Yes, I know loudness isn't a recipe for good sound - I definitely like quieter music more than I used to - but this isn't about ME, it's about OTHER PEOPLE. I've found that trying to coach listeners into using their volume dial is futile; you just have to give them what they want. Is that the wrong ethos for mixing classical/jazz music? Audiophiles would say 'yes', but then they're not famous for their empathy. If I can make classical records for classical music lovers, and they like the sound, I've done my job. If a bunch of strangers on the internet want to crap on me because I didn't do it their way, I don't care. If I was doing this professionally, I'd be doing A LOT more research and I'd learn how to use Reaper properly.... but I'm not.

I'm sorry to be indignant, but I feel like only the first few responders in this thread actually read my first post and understood where I'm coming from. The rest have totally hijacked it, but whatever..... it's the internet. I am very grateful to those who provided good help - you've undoubtedly opened a few windows for me. I'm sorry if any of this post offended anyone.

Personally, I don't like 'wall of sound' mixes anywhere near as much as I used to. I love those sounds in sections, but each track has to have a dynamic shift away from that so that my ears (and mind) don't get fatigued. I don't go to rock/metal concerts as much as I used to for this exact reason - the sonic browbeating is monotonous (much like other browbeating....). That said, I don't think a record has to be generally quiet in order to achieve this. I see no reason why an solo acoustic guitar has to be quieter than an orchestra, as long as the relative dynamics are retained. I mean, for God's sake, if we're talking about TOTAL dynamic fidelity, the acoustic guitar would have be objectively SILENT in order to contrast with the orchestra 'correctly'. That's why we have electronic sound - it enables us to combine otherwise-discordant elements and adjust them so they exist on the same palatable spectrum. Our sense of dynamics is localised, hence why professional sound engineers have to take breaks while editing/mixing.

I've had so many people look at my settings for all kinds of things (including my live effects units), and sneer at me for using the wrong settings. But I only do it because I think it sounds good. A priori is superior to a posteriori in many fields, but music isn't one of them (IMO).

Profile picture for user bouldersound
Member for
11 years 5 months

bouldersound Sun, 02/08/2015 - 23:17

Blue_Whistle88, post: 424863, member: 48849 wrote: I've lost count of how many times I've said that I want the LOUDEST track on the album to be as loud as it can be - NOT all of them. I want to RETAIN THE (potentially large) DYNAMICS between the different tracks in the album, by applying the same amount of gain boosting to each one. I'm doing this because the Tascam's mastering tools usually don't master the tracks loud enough at the final stage.

I get what you're saying and my last suggestion does address these requirements. All the dynamics within each album should remain the same while the differences between albums should be evened out so the loudest song of each album is as loud as is practical.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Sun, 02/08/2015 - 23:36

bouldersound, post: 424874, member: 38959 wrote: I get what you're saying and my last suggestion does address these requirements. All the dynamics within each album should remain the same while the differences between albums should be evened out so the loudest song of each album is as loud as is practical.

Yes, that's exactly what I'm getting at :)

Profile picture for user paulears
Member for
7 years 4 months

paulears Mon, 02/09/2015 - 01:05

Indignant?

Some of you may have heard of a program called 'mp3 Gain', which allows you to balance the volume levels of your mp3 files so that they're all roughly the same.

Control the volume level so they're all roughly the same - this seems to me to NOT be what you are now saying. Maybe the trouble is some of us didn't actually understand what you really meant? If the question is a bit skewed when we read it, then mistakes in understanding creep in - yet you start to complain we aren't listening? Pardon me for trying to help! Do you want every track roughly the same or not?

Profile picture for user DonnyThompson
Member for
8 years 6 months

DonnyThompson Mon, 02/09/2015 - 03:23

Blue_Whistle88, post: 424863, member: 48849 wrote: The rest have totally hijacked it, but whatever..... it's the internet.

Touchy touchy...

"But whatever....it's the internet..."

You've come to a forum filled with professionals, many of whom have over 25 years of experience, and you are bitching because we've given you other things to think about that completely pertain to your situation ?
Okay, then. I'm outta here.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Mon, 02/09/2015 - 17:40

paulears, post: 424876, member: 47782 wrote: Indignant?

Control the volume level so they're all roughly the same - this seems to me to NOT be what you are now saying. Maybe the trouble is some of us didn't actually understand what you really meant? If the question is a bit skewed when we read it, then mistakes in understanding creep in - yet you start to complain we aren't listening? Pardon me for trying to help! Do you want every track roughly the same or not?

"It doesn't remove the dynamics within each track (or each album even), but basically just allows you to control the general playback level of each file."

That's the sentence immediately after the one you quoted. I also said that "I still want to retain the dynamics between each track on the album". Your interpretation of that was:

"I think he really wants each track to be the loudest it can be? I'm not sure why."

Therefore, I want the album as a whole to be as loud as possible, whilst retaining the dynamics between each track. Sorry if I offended you, but I thought it was clear that my opening statement was a general description of the program, whilst the second part elaborated on exactly how I use it (and what I'm aiming to do here). That's why I got frustrated and said "I've lost count of how many times I've said... [etc]". If it wasn't clear, that's my fault and I apologise. I do appreciate all the help you've provided.

Donny - To be honest, you've been more critical than helpful. You said you don't do mastering because you're not an M.E. Well, neither am I, so should I just give up? How is that pertinent advice, when I'm asking how to master levels properly?

Profile picture for user Davedog
Member for
19 years 6 months

Davedog Fri, 02/13/2015 - 00:40

Pre-mastering. His recording device doesn't have enough gain to keep it all pristine and clear when it gets to unity. He want's a device he can bring up all the individual tracks to the point they need to be to mix. And trying to keep out the additional processing sound....and then go to mastering.

The cure for this is called gain-staging at capture. I don't give a rat's ass which device you capture to, they all can sound great if the capture is at the level it needs to be. And that level for ANY style of music,sound effects,foley,etc on ANY kind of recording device, is that point at which the input starts to collapse minus 3DB.

Realizing this , you eliminate another step in the process which is going to add SOMETHING to the sound...SOME processing to whatever you apply it too...Its not going to be the SAME as when you captured it....No matter WHAT the manufacturer might tell you. So your REAL choice here is to find devices that will enhance the instrument track,vocal track, percussion track, any track, that you need to be brought up to the level you need to mix at. One device will NOT do this. When you add this mythical device it is going to add its stamp, no matter how benign or transparent it may be. adding the SAME device on different content tracks will ensure that your record "has that _____(fill in the blank) all over it". So whatever you choose be sure its something you like.

Its almost like you have to actually plan this stuff.......................

Profile picture for user Reverend Lucas
Member for
7 years 1 month

Reverend Lucas Tue, 02/03/2015 - 14:52

It looks like MP3Gain applies lossy processing, beyond simple normalization. I'm not familiar with any programs that are dedicated solely to normalization, but it is easy enough to do in Reaper. Just import the WAV file, then Right click>Item Properties>Normalize items. You can then use the volume slider to adjust. Importing all the files in question, then normalizing them all would get the peak signal levels the same for all of them. You could then individually adjust each if some are louder than others.

Metering can be accomplished by any number of meter plugins on the master bus.

With that being said, making your masters 'acceptably loud' may involve other processing.

I hope this helps. Others might know of a dedicated normalization program that would simplify the process a bit.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Tue, 02/03/2015 - 18:59

Well, the idea is to retain the dynamics between different tracks, and just balance them as an 'album' relative to other 'albums'. Mp3 Gain lets you do this with an 'Album Gain' option. My understanding is that all mp3 files contain a master gain setting in their metadata (along with media info, technical statistics, etc), and that Mp3 Gain merely adjusts that. Therefore, it isn't really processing the audio at all, but rather the metadata which modulates the gain. I could be wrong though.

I know basic volume adjustment is all I need. I've gotten things sounding great with Mp3 Gain normalisation (no worse than a regular compressed wav file anyway), but I just want the normalisation done BEFORE the mp3-compression stage because I'm going to be submitting the masters to a duplicator (as wav files). I do compression et al on my multi-tracker :)

I'll give Reaper a go though, because it'd be good for me learn a bit more about it. I haven't downloaded any extra plug-ins for Reaper, as I haven't used it much, so do you think the built-in normaliser would be good enough? Or can you recommend any others I should try?

Member for
20 years 9 months

unknown Tue, 02/03/2015 - 19:46

Therefore, it isn't really processing the audio at all, but rather the metadata which modulates the gain. I could be wrong though.

I'm pretty sure there is a program, that I've even tried, posted in a an older thread here. You can load any file and adjust the levels of each one for comparisons.
That's what you are looking for, right?
I do however, just load whatever in my DAW as the Reverend mentioned but I must admit, this worked pretty slick. If I find it, I'll be sure to post the link.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Tue, 02/03/2015 - 20:26

That'd be awesome, thanks!

I just got recommended a program called SoX, which seems to do file conversions as well as editing of various properties (including gain). It even seems to have an 'album gain' function, similar to the one in Mp3 Gain (based on screenshots). Unfortunately, I just tried running it a few times and it doesn't seem to work. I tried various compatibility settings but no luck.

I just had a crack at editing the gain in Reaper, and while it proved to be very effective, it also compresses the files a lot. Lower gain boosts are ok, but once you start going above +3dB the wave peaks start getting clipped. Perhaps this is just the nature of basic gain boosting, but I'm wondering if there's a plug-in out there that intelligently boosts gain in such a way that doesn't require clipping?

Profile picture for user Reverend Lucas
Member for
7 years 1 month

Reverend Lucas Tue, 02/03/2015 - 21:11

Sounds like SoX may be an easy fix to your issue. In general, one normalizing function won't do a better job than the next - they'll all set your waveform to the maximum amplitude amplitude that can be reached without clipping. You can go down from there, but adding any volume beyond that requires compression/limiting. Basically a limiter will boost beyond that, while keeping the peak waveforms from clipping. I believe Reaper's basic compressor has a limiter mode that will accomplish this.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Tue, 02/03/2015 - 23:20

As mentioned, SoX doesn't seem to work, even with compatibility settings. Unfortunate, cause it looked good.

But regarding Reaper, I think you have a very valid point. I suspect that Mp3 Gain clips at higher gain settings, so the gain adjustment in Reaper will achieve what I need it to. I can also do a bit more mastering in there I guess, so that might be handy. So you think Reaper's compressor would be able to boost the signal without clipping? Is that just the Reacomp, or a different one? Also, are there any Reaper plug-ins (including packs) that you think might be worth getting, including compressors?

Profile picture for user DonnyThompson
Member for
8 years 6 months

DonnyThompson Wed, 02/04/2015 - 03:45

Blue_Whistle88, post: 424671, member: 48849 wrote: Also, are there any Reaper plug-ins (including packs) that you think might be worth getting, including compressors?

I would think that, like any other DAW platform, there is already a compressor that comes stock with Reaper ? - I may be wrong about that, because I'm not a Reaper user, but most DAW's will give you basic EQ and compressor processors.

Any VST plug will work, you don't need to get a "Reaper" processor, unless they sell additional processors, like Harrison MixBus does, which are processors proprietary to that DAW.

In that respect, there are many very good VST processors and processor collections available. Waves, UAD, IK Multimedia, Steinberg... all make very good plugs....But that's not to say that DAW manufacturers don't also make very good processors for their programs. Samplitude has some of the most effective and best sounding plugs I've ever heard in terms of EQ, Gain Reduction and Modeling. Reaper may have some good processors that come stock with their program as well... again, I can't say, as I'm not a Reaper user.

I'd steer clear of shareware or freeware VST's. As with most things, the level of quality is directly related to what you pay for.

Profile picture for user Reverend Lucas
Member for
7 years 1 month

Reverend Lucas Wed, 02/04/2015 - 07:43

Blue_Whistle88, post: 424671, member: 48849 wrote: So you think Reaper's compressor would be able to boost the signal without clipping? Is that just the Reacomp, or a different one? Also, are there any Reaper plug-ins (including packs) that you think might be worth getting, including compressors?

Compressors 'create' headroom by making the loudest parts of the waveform less loud, which allows you to then boost the gain without clipping. ReaComp is Reaper's built-in compressor. It should accomplish what you're trying to achieve. A limiter is just a compressor with a high compression ratio, usually higher than 10:1 is considered limiting. The ratio can be set in ReaComp. Properly setting a limiter will eliminate clipping, but be careful as limiting comes at the expense of dynamic range.

The only other compressor in Reaper is called ReaXComp, which is a multiband compressor. Multiband compressors are more complicated to use, but provide control over individual frequency bands and can do things singleband compressors can't.

All of the Reaper plugins are included with the package - there are no add-on purchases made by Cockos. As Donny mentioned, there are a whole lot of commercially available compressors that are not made by Cockos, and work with virtually any DAW. Some of them are intended to change the sound as little as possible. Some impart a unique character that may or may not be desirable, depending on what you're doing. That's another can of worms.

Profile picture for user DonnyThompson
Member for
8 years 6 months

DonnyThompson Thu, 02/05/2015 - 03:44

Reverend Lucas, post: 424691, member: 48050 wrote: Multiband compressors are more complicated to use, but provide control over individual frequency bands and can do things singleband compressors can't.

MB compression can also destroy sonics - in the hands of someone who doesn't understand how to use it.

It's one of the most common culprits that I hear in hurting the mixes that I'm called upon to critique ( and fix) in the smaller home recording "studios" that I consult with from time to time.

Truthfully, I do know what I'm doing with MBC, and yet, I rarely use it myself.

Reverend Lucas, post: 424691, member: 48050 wrote: Some of them are intended to change the sound as little as possible. Some impart a unique character that may or may not be desirable, depending on what you're doing. That's another can of worms.

Rev hit the nail on the head with this... and this is very important to know.

There are some "classic " compressors, like the 1176, LA2 (3 and 4), Fairchild 670, etc. - along with several other very popular and classic GR (gain reduction) models - that not only apply gain reduction, but that can also add a distinct sonic "character" to the signal at the same time. Within these compressors, some are "faster" than others, (1176) they react quicker, and this imparts a particular "sound" as well. This doesn't mean that "slower" compressors aren't as good - to the contrary, models like the LA2 or the Fairchild 670 models can also sound wonderful because of the way that they react.

None of these models that I just mentioned - either in plug form or with the actual hardware, would be considered by most to be sonically "transparent", and by and large, the engineers who use these, are often looking for that sonic thumbprint that these models offer when they choose to use one.

There are a few other "classic" compressors that tend to be more towards the neutral side sonically....not that they are entirely transparent, but are more-so than the models mentioned above. The Focusrite Red is one of these, The SSL 2-Bus Compressor is another. But, don't be fooled into thinking that these are substandard, just because these don't have the same obvious "character" that the others mentioned have. These two fairly neutral-sounding compressors work wonderfully, and have been used on thousands of records.

Transparent GR devices will do what they are intended to do (see Rev's explanation on compressors above) but don't add any "color" to the sound, other than the reductive gain result that is intended. Most stock compressors tend to be transparent.

If you are looking for straight ahead GR, I would think that the compressor that comes stock with Reaper would probably do the job just fine. Most popular DAW's provide at least one stock compressor that work and sound just fine.

So, unless you are after a certain sonic character to the compression, I'd just stick with what you have now.

If you'd like to learn more, here are a few articles that you can check out - they are all from SOS, a trade mag that I trust:

[="http://www.soundonsound.com/sos/1997_articles/apr97/compressors.html"]Compressors - explanation[/]="http://www.soundonsound.com/sos/1997_ar…"]Compressors - explanation[/]

[[url=http://="http://www.soundonsound.com/sos/sep09/a…"]Classic Compressors[/]="http://www.soundonsound.com/sos/sep09/a…"]Classic Compressors[/]

[[url=http://[/URL]="http://www.soundonsound.com/sos/sep09/a…"]Compression Made Easy[/]="http://www.soundonsound.com/sos/sep09/a…"]Compression Made Easy[/]

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Thu, 02/05/2015 - 14:05

Cool, thanks! I remember reading that there's a particular order you're supposed to use when setting them (using ARRT as the mnemonic) - Attack, Release, Ratio, Threshold. I've only ever used stock settings and changed a couple of things to make it suit, and to be honest I don't really follow that order much (but I'm still new at this).

I usually use a simple MB-compressor in my Tascam DP-32 (which is my main recording device), and I totally agree with what you're saying about being careful not to use damage. I usually end up using almost identical settings on each of the three bands, because drastic changes can cause big problems when the dynamic of the track changes (e.g. loss in bass & gain in mid-range). I think the only thing I change is the gain for each band, usually with an extra 1 or 2 dB for the bass and treble (as I was taking very raw, acoustic recordings with lots of mid-range). But this this particular stage (which I guess is 'secondary mastering'), just a standard compressor will do.

I'll go ahead and use the one built into Reaper then. I'm still struggling to figure out exactly WHAT is is that differentiates a proper compressor from clipping via gain boosting; isn't 'peak reduction' just the clipping of peaks BEFORE they reach the threshold, so they don't get squared off? So there's still the same loss in dynamic range (i.e. distance between peaks and troughs)? Though I guess that WOULD preserve the tone better, as it retains the 'shape' of the peak.

I'll use the standard gain fader if I totally muck things up, but hopefully I can figure out how to use the ReaComp such that I do actually improve the sound a bit (though I'm sure it's pretty transparent).

Thanks heaps for all the help guys! I literally couldn't have asked for anything better :D

Profile picture for user DonnyThompson
Member for
8 years 6 months

DonnyThompson Fri, 02/06/2015 - 02:34

Blue_Whistle88, post: 424741, member: 48849 wrote: I've only ever used stock settings and changed a couple of things to make it suit, and to be honest I don't really follow that order much (but I'm still new at this).

Ooooo... bad move. Repeat after me... "presets are bad...presets are bad..."

Don't ever rely on presets. Always start at a null, a full reset. There's no way that anyone can pre-determine what compression schematic you need.
Learn more about gain reduction, Blue... I promise you it will not be useless information. In fact, it's crucial if you want to do any kind of audio engineering.

Profile picture for user Blue_Whistle88
Member for
6 years 4 months

Blue_Whistle88 Fri, 02/06/2015 - 11:02

Ok, fair enough. But in my defence, the mastering compressor in the Tascam comes up with some default settings every time you switch it on anyway, so you have to change it from SOME kind of preset no matter what. I only jump to a certain preset because I found it to be a better starting point than 'default', but I still change a few things in it. It's a relatively simple compressor too, so you can go a long way only changing 2 or 3 settings.

x