I am making available some spoken word audios on a website that were captured in mp3 from cassettes in the 80s and 90s. I need these mp3s to be available ad infinitum without further loss of quality. As mp3 is lossy, what is the best way to store them for longevity and use as masters for creating new copies to put on the website? I appreciate the wav won’t improve the quality already lost, but would the wav format be beneficial to stop the degradation I’ll get from keeping them as mp3?
As you note, putting WAV in the process doesn’t cure anything, but that’s a good format for archives.
MP3 is a time-delay land mine.
You know if you make an MP3 from an MP3, the quality gets worse. The damage doesn’t quite double, but if you’re close to the limit of hearing, you will be able to tell which is the original. That limit would be 32 for mono and 64 for stereo. Most people can just start to tell there’s something wrong.
So the solution is working in WAV which has no compression distortion, or work in MP3 at a nose-bleed high compression value such as 320. That will work for a while—but you can never come back down.
This is why Audiobooks want MP3 delivery at 192. If you present your book in mono, that’s stupid-high quality, unless you’re a company that plans on making many different audio products from the work. The quality will tick down every time they do.
If you make a WAV and then make another MP3, the effect will be that of not having the WAV in the middle at all.
And do not cross compressions, either. No fair making an M4A from an MP3. The damage will be undetermined.
MP3 is lossy compression but the “loss” happens during compression when data is thrown-away to make the file smaller. No further damage happens when you decompress. The file is automatically decompressed when you play it and of course it’s decompressed if you convert it to WAV.
Then, it’s just like any other digital file and it won’t degrade by itself unless you get some kind of file corruption or the the hard drive (or other medium fails, etc.).
If you open it for editing and then re-export as MP3, damage accumulates with each generation of lossy compression.
…MP3 isn’t always “terrible”. MP3 is “smart” and it tries to throw-away stuff you can’t hear anyway and at high-quality settings (higher bitrates = bigger files = less data thrown-away) it can often sound identical to the original. (It also uses other techniques that are not simply “throwing-away” data.)
AAC (MP4/M4A) is also lossy compression but damage doesn’t accumulate when it’s decompressed and then re-compressed. (That’s only if you start with AAC, not if you make a lossy-to-lossy MP3-to-AAC conversion.)
If you upload to YouTube or other streaming service, they will compress (or re-compress) to their own format and that’s another generation of lossy compression.
WAV (or FLAC, which is lossless compression) is good for archiving because it can be converted to any lossless or lossy format as-needed, now or any time in the future. But that won’t help if you don’t have a lossless original. FLAC files are not less-prone to damage/corruption but they contain a checksum so you can know if they are corrupted or not. Embedded metadata (“tagging”) is more-standardized and better-supported on WAV than FLAC. And FLACs are about half the size of the uncompressed original.
So if my original mp3s are stereo 32 bit 192kbps VBR. What settings would you suggest in wav? These are all spoken word so no music. Also, will I lose anything if I mono the audio?
16-bits is fine, and you can keep the same sample rate (kHz) as the MP3. The sample rate is shown in the lower-left of the Audacity window, and it will match the 1st file you open. (So you don’t have to do anything with the sample rate.)
Also, will I lose anything if I mono the audio?
Is it stereo now? If it’s currently truly-stereo (different sounds in the left & right channels) of course, you’ll loose the stereo. If left & right are identical it won’t make any difference. If you’re not sure you can try it, or leave it stereo.
An uncompressed (WAV) mono file will be half the size of the stereo. With MP3 the file size is determined by the bitrate (kbps) and mono or stereo doesn’t affect size.
So, DVDdoug and koz have much more experience in this area than I have. But I thought I would add my thoughts to the mix. Personally, as the damage is already done, unless you have a compelling reason otherwise, I would just leave it as is. Of course that is the easiest, and it is easier to archive and download, etc. The downside of converting the files to .WAV is that sometime in the future, someone may see the WAV file and for whatever reason do their own conversion to MP3, thinking they are being smart but just unknowingly just creating further damage to the file.
If left & right are identical it won’t make any difference. If you’re not sure you can try it, or leave it stereo.