So having recorded from vinyl directly onto my MP3 player (as WMA files) I wanted to convert to MP3. That done, I loaded them onto MP3 Gain only to get an Error Log for all the tracks, saying 'Errors During Processing' they are not being 'analyzed' or 'normalized'. Nor is there an Error Log containing any details.
It's the latest downloadable version , have tried removing, then uploading again, still the issue remains. I checked with other MP3s and it seems to work, but for some reason is refusing to process these particular files. Any ideas, help to resolve this puzzle would be most helpful
If you don't mind emailing me one of them I'll have a look at it. email@example.com
Hi there, thanks for your kind offer to have a look at this problem. Have uploaded one of the troublesome MP3s
Sorry, but that service wants me to sign up to download the file, and I'd rather not. Be much easier to attach it to the email anyway. ;)
Did try to email but file's too big
Dropbox or something similar that just gives you a public http link?
Strange, it says it's under 6 MB, and gmail's limit is 20.
ah maybe it was mail service, shall try another account…
OK Sent via a Yahoo account, hope you got it :)
Got it. What are you using to encode to mp3? Some of the non-audio data in there is screwing with mp3gain. First it thinks it is actually a layer 2 file, which it isn't, and after disabling the check for layer 2 files it then thinks there aren't enough samples to even work with. However, if I use foobar2000 and rebuild the mp3 stream then mp3gain works fine on the rebuilt file. BTW, foobar2000 is much faster for this job, but isn't as straight forward. Its analyzation algorithm is incredibly fast compared to mp3gain's, and it will also utilize all your CPU's cores if you have more than one. Uses the same ReplayGain rules, but is faster.
Ah..I wonder, had saved the WMA file to 'WavePad Sound Editor' Then upon editing there saved directly as MP3. Maybe that where the issue arose?
Maybe the lame dll that WavePad uses, or the parameters WavePad is passing to it, is the problem. Could try renaming the lame dll it uses to lame_enc.old and download the latest version of lame and see if it makes any difference. But either way, foobar2000 will indeed fix your existing files and then you can either use it to do the ReplayGain stuff, or mp3gain will work then, too. Load your files in foobar2000, right-click, Utilities, Rebuild mp3 stream.
Many thanks for your help on that, really appreciate your insight. I just reconverted the offending MP3s and they worked fin in MP3 Gain, so looks like the issue was indeed the WavePad
Sounds like you got things worked out already, but here's a little more detail on the exact problem with the file:
It looks like WavePad uses a "lazy" ID3v2 tag writer. ID3v2 tags are inserted at the beginning of the mp3 file. A proper tag writer will insert the tag and shift the existing mp3 data to make room for the tag. It's faster and easier, though, to just write over the existing mp3 data. That screws up the first few milliseconds of mp3 data, but not so much that you'd ever notice audibly.
As an analogy, pretend that your mp3 data is just English sentences. Let's say your mp3 file starts like this:
"Four score and seven years ago our fathers brought forth, upon this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal."
A good ID3v2 tag writer will insert its tag data like this:
"I AM AN ID3v2 TAG. Four score and seven years ago our fathers brought forth, upon this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal."
A bad ID3v2 tag writer will do this:
"I AM AN ID3v2 TAG.en years ago our fathers brought forth, upon this continent, a new nation, conceived in Liberty, and dedicated to the proposition that all men are created equal."
Sorry to bump this but I'm having the same issue with this one single file… after going through my 7k mp3 files, MP3Gain couldn't handle this little one here:
I followed your advice and rebuilt the MP3 stream, also did the VBR header fix since it's VBR, just in case of, but it didn't help, MP3Gain still fails to analyse it properly.
So, it's no big deal for me, but I thought maybe there's a very simple fix, and I don't have to remake the MP3s… or otherwise it might be useful for future bug fixes and what not.
Pretty sure that file's just corrupted and that song should be encoded again. Every utility I've checked it with has either returned a corrupted file error or simply crashed.
Oh well, I'll do that then. Thanks for having a look at it anyways.