Automatic Gain Control

[rjharle]

I have a number of CDs I consolidated into one file, so now I have a file with about 100 recordings. The problem is that they all are not playing at the same volume. I have to change the volume every 1 to 5 songs. I searched among the many audio programs for an Automatic Gain Control effect I could apply to the files. I only found one; Wavepad. I tried it on the files and the volume evened out among the files but, the file now sounds like there isn't any dynamic range; all the music is at and stays at one volume. I though the gain control would be proportional and match the variations of the recording; not make every one volume.

Is there a program with automatic gain control that would correct the volume issue without messing up the content.

 

[TLS Guy]

I have a number of CDs I consolidated into one file, so now I have a file with about 100 recordings. The problem is that they all are not playing at the same volume. I have to change the volume every 1 to 5 songs. I searched among the many audio programs for an Automatic Gain Control effect I could apply to the files. I only found one; Wavepad. I tried it on the files and the volume evened out among the files but, the file now sounds like there isn't any dynamic range; all the music is at and stays at one volume. I though the gain control would be proportional and match the variations of the recording; not make every one volume.

Is there a program with automatic gain control that would correct the volume issue without messing up the content.

Yes, there is a standardized level for WAV files. However you need a DAW with advanced software to rewrite a Wav. File. My DAW has Wave Lab installed. If you bring up a Wav. file there is an icon you can click, which will normalize the whole Wav. to AES/EBU specification as a far as average volume levels.

The fact that you have this issue can be due to your source being mastered by someone who does not know their craft.

However, there is a caveat to this. If you take a compressed recording and compare it to a high dynamic range uncompressed one, then the compressed recording will have a higher average level, whereas the better uncompressed recording will have a lower average level. So the better high dynamic uncompressed recording will play at a lower average level and seem quieter. If you think about it this has to be, as if the average level of the high dynamic range recording would run out of bits, in the forte passages which would result in abrupt catastrophic distortion. Analog overloads somewhat gracefully with added THD. Running out of bits, causes sudden equipment busting artefacts. So when mastering I keep a very careful eye on the bit meter.

As you probably have realized I am a superannuated old curmudgeon. In times past I made many live analog recordings, and in order to maximize S/N ratio you held your nerve and pushed the crescendos to the limit, even pushing the VU meter a db. or two into the red. On going to digital recording, you made darn sure you kept the most dynamic passages below 0 db. If you did not, you could end up ruining your recording. As software improved you could correct volume levels by the batch. But if the master ran out of bits during a live recording you were out of luck. I don't think I ever ran a live digital recording out of bits.

So the short answer is you can rewrite a Wav. file to AES/EBU spec, BUT the higher dynamic range recording will always seem to be quieter than the poorer low dynamic range recording. That means that if you tried to make the high dynamic recording play as loud as the low one, you would run the high dynamic recording out of bits on the crescendos.

 

[lovinthehd]

Foobar2000 has such a gain control, but how it particularly works I'm not sure.

 

[TLS Guy]

Foobar2000 has such a gain control, but how it particularly works I'm not sure.

I have looked into it. It seems to be just that, an automatic gain control to keep average level constant. That sort of system is a crude compressor. That is not what you want. Reviews of the Foobar AGC are not encouraging, which should not surprise.

You just have to keep your volume control handy.

As I think about it that may be the reason for the vinyl revival, as it encourages playing the album, or at least a side. Making playlists from an assortment of albums, is not a good plan. That is something I have never done.

 

[lovinthehd]

I have looked into it. It seems to be just that, an automatic gain control to keep average level constant. That sort of system is a crude compressor. That is not what you want. Reviews of the Foobar AGC are not encouraging, which should not surprise.

You just have to keep your volume control handy.

As I think about it that may be the reason for the vinyl revival, as it encourages playing the album, or at least a side. Making playlists from an assortment of albums, is not a good plan. That is something I have never done.

Might be just is wanted by OP if pursuing something like this in the first place, which isn't something I'm particularly interested in, I don't use it. Audyssey Dynamic Volume could well be similar. I don't use either myself. I play albums mostly, so within an album makes little difference. If mixing various singles, YMMV

ps fwiw I use foobar2000 as my main organizer/player on my pc....

 

[rjharle]

Yes, there is a standardized level for WAV files. However you need a DAW with advanced software to rewrite a Wav. File. My DAW has Wave Lab installed. If you bring up a Wav. file there is an icon you can click, which will normalize the whole Wav. to AES/EBU specification as a far as average volume levels.

The fact that you have this issue can be due to your source being mastered by someone who does not know their craft.

However, there is a caveat to this. If you take a compressed recording and compare it to a high dynamic range uncompressed one, then the compressed recording will have a higher average level, whereas the better uncompressed recording will have a lower average level. So the better high dynamic uncompressed recording will play at a lower average level and seem quieter. If you think about it this has to be, as if the average level of the high dynamic range recording would run out of bits, in the forte passages which would result in abrupt catastrophic distortion. Analog overloads somewhat gracefully with added THD. Running out of bits, causes sudden equipment busting artefacts. So when mastering I keep a very careful eye on the bit meter.

As you probably have realized I am a superannuated old curmudgeon. In times past I made many live analog recordings, and in order to maximize S/N ratio you held your nerve and pushed the crescendos to the limit, even pushing the VU meter a db. or two into the red. On going to digital recording, you made darn sure you kept the most dynamic passages below 0 db. If you did not, you could end up ruining your recording. As software improved you could correct volume levels by the batch. But if the master ran out of bits during a live recording you were out of luck. I don't think I ever ran a live digital recording out of bits.

So the short answer is you can rewrite a Wav. file to AES/EBU spec, BUT the higher dynamic range recording will always seem to be quieter than the poorer low dynamic range recording. That means that if you tried to make the high dynamic recording play as loud as the low one, you would run the high dynamic recording out of bits on the crescendos.

Thank you.
I cannot see myself rewriting each file just to correct a volume issue. Doesn't seem worth it. I guess the best way to handle this may be to segregate the files by volume.

 

[rjharle]

I have looked into it. It seems to be just that, an automatic gain control to keep average level constant. That sort of system is a crude compressor. That is not what you want. Reviews of the Foobar AGC are not encouraging, which should not surprise.

You just have to keep your volume control handy.

As I think about it that may be the reason for the vinyl revival, as it encourages playing the album, or at least a side. Making playlists from an assortment of albums, is not a good plan. That is something I have never done.

Another question.
Could I string a number of files together to make one big file and then "Normalize" the file to level it out. Then separate the files. Would that work?

 

[TLS Guy]

Another question.
Could I string a number of files together to make one big file and then "Normalize" the file to level it out. Then separate the files. Would that work?

You could if you have an editing program that can handle Wav. files. However, you have not quite understood what I have said. You can not deal with the issue of the varying dynamic range of the the different selections. This is very true in the pop rock world, as dynamic range compression is all over the map. So normalizing to AES/EBU will still make the better recordings with the higher dynamic range sound quieter. If you think about this has to be as the number of bits on a peak is a very finite quantity.

Even in the well engineered classical world the dynamic range of the forces involved will defeat you. If you play a small chamber work it will have by nature of the composition and forces involved have an inherently lower dynamic range compared to a huge work with large orchestra chorus and organ, which has a dynamic range of over 100 db.

So the large forces work will seem to play quieter as headroom is required for the huge fortes. So that is why if you play solo or chamber music you can get away with smaller speakers and amps. If you want to play a large scale work, so for instance the end of Beethoven's ninth at concert volume, then you need huge powerful speakers and a lot of amp power, to reproduce the orchestra, chorus and soloists at full forte.

 

[rjharle]

You could if you have an editing program that can handle Wav. files. However, you have not quite understood what I have said. You can not deal with the issue of the varying dynamic range of the the different selections. This is very true in the pop rock world, as dynamic range compression is all over the map. So normalizing to AES/EBU will still make the better recordings with the higher dynamic range sound quieter. If you think about this has to be as the number of bits on a peak is a very finite quantity.

Even in the well engineered classical world the dynamic range of the forces involved will defeat you. If you play a small chamber work it will have by nature of the composition and forces involved have an inherently lower dynamic range compared to a huge work with large orchestra chorus and organ, which has a dynamic range of over 100 db.

So the large forces work will seem to play quieter as headroom is required for the huge fortes. So that is why if you play solo or chamber music you can get away with smaller speakers and amps. If you want to play a large scale work, so for instance the end of Beethoven's ninth at concert volume, then you need huge powerful speakers and a lot of amp power, to reproduce the orchestra, chorus and soloists at full forte.

Sorry I should have picked this up in your first reply. These are not WAV Files. These are files ripped from CDs in a M4a format. M4a is a file extension for an audio file encoded with advanced audio coding (AAC) which is a lossy compression. Since the original performance was destroyed by this process there isn't much dynamic range left. (maybe 20-30db ?) I thought that CDs/files had a coding or setting in the files to a certain loudness and keep it consistent; so little Johnny won't blow out his eardrums using ear plugs. But each Record Label may have their own setting. Not sure how they managed to mess it up.

 

[TLS Guy]

Sorry I should have picked this up in your first reply. These are not WAV Files. These are files ripped from CDs in a M4a format. M4a is a file extension for an audio file encoded with advanced audio coding (AAC) which is a lossy compression. Since the original performance was destroyed by this process there isn't much dynamic range left. (maybe 20-30db ?) I thought that CDs/files had a coding or setting in the files to a certain loudness and keep it consistent; so little Johnny won't blow out his eardrums using ear plugs. But each Record Label may have their own setting. Not sure how they managed to mess it up.

You are still confused. M4A is a Wav file extension. All PCM audio is Wav. file based. M4A is also known as MPEG4 and the compression codec is AAC, which stands for advanced audio coding. What you are confused about is that these compressive codecs, of which AAC is the best, have NOTHING to do with dynamic range compression at all. What these compressive codecs are all about, is eliminating data from the audio stream that the brain will not notice. This is to save data largely in streaming. However random noise always reveals the sleight of hand of these lossy codecs. They beat example is applause which is random, and sounds more and more abnormal as the bit rate of a lossy codec is reduced.

So when you have a Wav. file at say 44.1 and want to convert, you can do that in your program instantly and by batch. But you can not convert it back to lossless. On the other hand codecs, like FLAC are lossless, as they are converted back to the original Wav. file on playback. FLAC is a true endode/decode format. Lossy codecs like AAC and MP3 can not be decoded.

Dynamic range compression is a program that reduces the spl. between the loud and soft dynamics of the program. The compression is specified in db with attack and release times.

So lossless audio can be just as compressed as any lossy codec file. Digital compression and audio compression are totally different concepts and completely independent of one another. So you can have a dynamically highly compressed lossless file and an uncompressed lossy file in an AAC codec for instance.

If you are ripping and archiving music these are concepts you really need to understand and study, as it is clear to me you have serious and fundamental misunderstandings.

 

[rjharle]

You are still confused. M4A is a Wav file extension. All PCM audio is Wav. file based. M4A is also known as MPEG4 and the compression codec is AAC, which stands for advanced audio coding. What you are confused about is that these compressive codecs, of which AAC is the best, have NOTHING to do with dynamic range compression at all. What these compressive codecs are all about, is eliminating data from the audio stream that the brain will not notice. This is to save data largely in streaming. However random noise always reveals the sleight of hand of these lossy codecs. They beat example is applause which is random, and sounds more and more abnormal as the bit rate of a lossy codec is reduced.

So when you have a Wav. file at say 44.1 and want to convert, you can do that in your program instantly and by batch. But you can not convert it back to lossless. On the other hand codecs, like FLAC are lossless, as they are converted back to the original Wav. file on playback. FLAC is a true endode/decode format. Lossy codecs like AAC and MP3 can not be decoded.

Dynamic range compression is a program that reduces the spl. between the loud and soft dynamics of the program. The compression is specified in db with attack and release times.

So lossless audio can be just as compressed as any lossy codec file. Digital compression and audio compression are totally different concepts and completely independent of one another. So you can have a dynamically highly compressed lossless file and an uncompressed lossy file in an AAC codec for instance.

If you are ripping and archiving music these are concepts you really need to understand and study, as it is clear to me you have serious and fundamental misunderstandings.

I don't agree that my understanding is as serious as you think. In the literal sense, you could say all sound files (except DSD) started out as WAV but, once it is compressed, relabeled and does not sound like a WAV file is it still and WAV. It does not look like a WAV, talk like a WAV or walk like a WAV, yet it is still a WAV. Because everything is related by association.

I do know the difference between digital and sound compression. Audio compression reduces the range by attenuating the louder signals and boosting the quieter signals. Digital compression finds repeated characters in a data set and replaces them with tokens, or shortened sequences. Because I didn't separate them and explain them in detail didn't mean I didn't understand them.

So the .WAV file is adjusted (being a much larger file) by digital compression (removing unperceivable sounds and crunching repeatable data) and audio compression by (reducing the range by attenuating the louder signals and boosting the quieter signals) so little Johnny doesn't lose his hearing. Then the file is given a new file extension (in this case) .M4a. The file will now fit on a CD and can be safely played, loudly, on a Smartphone. Once this is done it can no longer be transformed back into a WAV file but is still is a WAV. We are not talking about FLAC extensions.

I'm sure I cannot respond to your posts with your level of detail to prove my worthiness.

But, I'm still looking for a VST or program that will correct the variations in loudness when a number of CDs are ripped and the ,M4a files are integrated.

 

[TLS Guy]

I don't agree that my understanding is as serious as you think. In the literal sense, you could say all sound files (except DSD) started out as WAV but, once it is compressed, relabeled and does not sound like a WAV file is it still and WAV. It does not look like a WAV, talk like a WAV or walk like a WAV, yet it is still a WAV. Because everything is related by association.

I do know the difference between digital and sound compression. Audio compression reduces the range by attenuating the louder signals and boosting the quieter signals. Digital compression finds repeated characters in a data set and replaces them with tokens, or shortened sequences. Because I didn't separate them and explain them in detail didn't mean I didn't understand them.

So the .WAV file is adjusted (being a much larger file) by digital compression (removing unperceivable sounds and crunching repeatable data) and audio compression by (reducing the range by attenuating the louder signals and boosting the quieter signals) so little Johnny doesn't lose his hearing. Then the file is given a new file extension (in this case) .M4a. The file will now fit on a CD and can be safely played, loudly, on a Smartphone. Once this is done it can no longer be transformed back into a WAV file but is still is a WAV. We are not talking about FLAC extensions.

I'm sure I cannot respond to your posts with your level of detail to prove my worthiness.

But, I'm still looking for a VST or program that will correct the variations in loudness when a number of CDs are ripped and the ,M4a files are integrated.

If you take an uncompressed Wav. file and convert it to M4A it is still a Wav. You give it a different file extension in the folder. Both are Wav. files. If you send the original to the trash bin, you can not get back your original Wav.file as it will not convert back as there is missing data, which is the point of lossy codecs.

 

[Eppie]

I only have two solutions but one is time consuming and the other expensive.

I use Exact Audio Copy to rip my CDs. It has a normalize option but what is nice is that you can set a range. For example, I have it set to normalize to 98% but it will only normalize if peaks are below say 90% or over 100%. There is no compression applied so the dynamics remain the same. Replay gain info is added to the file but the playback software must support it. Downside is that you have to rip the low volume CDs again.

Other option is using Roon for managing and playing back music, which requires an annual subscription or an expensive lifetime membership. Roon has digital processing options and one is volume leveling.

 

[rjharle]

If you take an uncompressed Wav. file and convert it to M4A it is still a Wav. You give it a different file extension in the folder. Both are Wav. files. If you send the original to the trash bin, you can not get back your original Wav.file as it will not convert back as there is missing data, which is the point of lossy codecs.

Source Wikipedia:
Audio CDs
Audio CDs do not use the WAV file format, using instead Red Book audio. The commonality is that audio CDs are encoded as uncompressed PCM, which is one of the formats supported by WAV. WAV is a file format for a computer to use that cannot be understood by most CD players directly. To record WAV files to an Audio CD the file headers must be stripped, the contents must be transcoded if not already stored as PCM, and the PCM data written directly to the disc as individual tracks with zero-padding added to match the CD's sector size

DSD conversion
Convert DSD audio files to/from FLAC, MP3, M4A, AAC, PCM, DXD, Apple Lossless, Opus, Vorbis, and more audio file formats. DSD conversion between DSF, DFF, WavPack DSD formats in bit-exact DSD mode. Convert audio from 60+ file formats. High resolution audio up to 64-bit 384kHz is supported. Rip audio CDs with bit-perfect audio CD ripper. Edit all metadata of audio files and transfer all metadata in format conversions.

Are they both still WAV files?

If I’m looking at a M4a file how would I know the origin of the file?

 

[rjharle]

I only have two solutions but one is time consuming and the other expensive.

I use Exact Audio Copy to rip my CDs. It has a normalize option but what is nice is that you can set a range. For example, I have it set to normalize to 98% but it will only normalize if peaks are below say 90% or over 100%. There is no compression applied so the dynamics remain the same. Replay gain info is added to the file but the playback software must support it. Downside is that you have to rip the low volume CDs again.

Other option is using Roon for managing and playing back music, which requires an annual subscription or an expensive lifetime membership. Roon has digital processing options and one is volume leveling.

Thank you I'll try Exact Audio Copy. Is the normalized function for individual files only, or will it normalize all files disc after disc making all the files have the same loudness?

 

[TLS Guy]

Source Wikipedia:
Audio CDs
Audio CDs do not use the WAV file format, using instead Red Book audio. The commonality is that audio CDs are encoded as uncompressed PCM, which is one of the formats supported by WAV. WAV is a file format for a computer to use that cannot be understood by most CD players directly. To record WAV files to an Audio CD the file headers must be stripped, the contents must be transcoded if not already stored as PCM, and the PCM data written directly to the disc as individual tracks with zero-padding added to match the CD's sector size

DSD conversion
Convert DSD audio files to/from FLAC, MP3, M4A, AAC, PCM, DXD, Apple Lossless, Opus, Vorbis, and more audio file formats. DSD conversion between DSF, DFF, WavPack DSD formats in bit-exact DSD mode. Convert audio from 60+ file formats. High resolution audio up to 64-bit 384kHz is supported. Rip audio CDs with bit-perfect audio CD ripper. Edit all metadata of audio files and transfer all metadata in format conversions.

Are they both still WAV files?

If I’m looking at a M4a file how would I know the origin of the file?

Red Book is a specification and NOT a format. I should know as I master CDs professionally now and again. My DAW has Wave lab mastering program. I can assure you that CDs are made of Wav. files. You line up all the tracks after optimizing them and make sure they meet Red Book spec. Then you make sure they are in the correct order, and set the correct gap between each Wav.file or make files run seamlessly. I can put any CD in the disc tray and see all the Wav.files. All the Wav. files on the CD must have a 44.1 kHz sample rate. The bit rate is 1411.2 kbit/s. If Wikipedia says anything different then it is incorrect. I have mastered quite a few CDs in my time.
Here is a track of a CD I mastered, and you can watch and hear the Wav. file playing on my DAW.

 

[Eppie]

Thank you I'll try Exact Audio Copy. Is the normalized function for individual files only, or will it normalize all files disc after disc making all the files have the same loudness?

It is enabled under EAC/EAC Options... under the Normalization tab. Just check the box and set the levels you want and it applies to every CD until you disable it. For example, I do not use it for classical music due to the inherent wide dynamic range, but it is good for rock albums recorded at lower levels. Your player needs to support regain play info so try it on 2 or 3 CDs before you commit to doing too many.

 

[rjharle]

It is enabled under EAC/EAC Options... under the Normalization tab. Just check the box and set the levels you want and it applies to every CD until you disable it. For example, I do not use it for classical music due to the inherent wide dynamic range, but it is good for rock albums recorded at lower levels. Your player needs to support regain play info so try it on 2 or 3 CDs before you commit to doing too many.

Thank you

 

[rjharle]

Red Book is a specification and NOT a format. I should know as I master CDs professionally now and again. My DAW has Wave lab mastering program. I can assure you that CDs are made of Wav. files. You line up all the tracks after optimizing them and make sure they meet Red Book spec. Then you make sure they are in the correct order, and set the correct gap between each Wav.file or make files run seamlessly. I can put any CD in the disc tray and see all the Wav.files. All the Wav. files on the CD must have a 44.1 kHz sample rate. The bit rate is 1411.2 kbit/s. If Wikipedia says anything different then it is incorrect. I have mastered quite a few CDs in my time.
Here is a track of a CD I mastered, and you can watch and hear the Wav. file playing on my DAW.

Your recording sound very nice

There isn't any question that the files start out as a WAV but, after it goes through its transformation can we still call it a WAV? It looks like a WAV is only a means, a package, on its way to became something else.

Also, since DSD can also be made into M4a files would you then call the M4a a DSD?

 

[TLS Guy]

Your recording sound very nice

There isn't any question that the files start out as a WAV but, after it goes through its transformation can we still call it a WAV? It looks like a WAV is only a means, a package, on its way to became something else.

Also, since DSD can also be made into M4a files would you then call the M4a a DSD?

The Wav. files, like the one you see, are on the CD and the very core and essence of what a CD is. It is not converted to something else. If I open up a CD on my workstation I can view the Wavs just like the one I showed you. It is in exactly in the same format on the CD as my hard drive. They are identical.

Before any DSD is manipulated it has to be converted to PCM. DSD is absolute nonsense. It is a format that is virtually impossible to edit or manipulate. No one bothers any more. BIS of Sweden were the last to try and not brake the DSD chain. They have long given up. So ALL DSD recordings are now converted to PCM for all post recording production work. It is then converted after all editing, balancing, mixing etcetera, back to DSD to hoodwink and satisfy the Audiophools. They are far too stupid to understand and continue to believe DSD discs are superior, even though they have undergone DSD-PCM-DSD conversion. It is all total fakery. DSD needs to go and the sooner the better. As an audio only multichannel format, it should have been superseded by audio only Blu Ray discs a long time ago. It is not true in any sense, that PCM is inferior to DSD.

When you make a M4a file from a DSD source it HAS to be converted to PCM when converted to any other type of file or container. So once you have made your M4a it is PCM and NOT DSD and that is the way it will remain. So there is zero benefit from its transient life as DSD.