Well, I decided to try out FLAC this week. For those of you who missed it, FLAC (see http://flac.sourceforge.net/) is a lossless encoder (kinda like ZIP, but for WAV files).

My motivation is the fact that I'd like to start keeping an archive, so that I don't have to re-rip my CD collection every time there's a new version of LAME released.

So, I decided to try it out on my collection of Harry Potter audio CDs (read by Stephen Fry). I usually have to re-rip Harry Potter (43 CDs) every couple of months, when Mike spots that there's a glitch in one of the tracks, so this time I'm being thorough.

It's good. These are (obviously) spoken text, so you'd expect good compression, but FLAC gets the files down to about 20% of their original (WAV) size. The compression is flawless.

Random shell scripts which I found useful:

Given a directory structure like this:

/home/music/roger/Books/Harry Potter/

    Philosopher's Stone/Disc 01/

    01 - Chapter 1- The boy who lived.wav

$ cd /home/music/roger/Books/Harry Potter/Philosopher's Stone

$ md5sum */*.wav > MD5SUMS

$ for i in */*.wav ; do \

    ( grep `flac -c -d ${i%.wav}.flac | md5sum` MD5SUMS \

    | grep $i || echo "Failed" ) ; done >VERIFY

$ diff MD5SUMS VERIFY

Generates MD5 checksums for the WAV files, which you can then verify against the decoded FLAC files.

Since they match, I'm going off now to delete the WAV files. Just thought I'd share this with you.