From what I can understand of id3v2 SYTL (Synched Lyrics or Text... is that right??) they are directly kind of "between" audio frames? Am I right? Is there any easy way to find which frame the player is? The way I sync with the player now is to watch the /proc/empeg_notify (By the way, could somebody confirm that this file is in the RAM of the player, and not anywhere on a HDD???). The viewer would have to include some sort of mp3 decoding library, which my current knowledge of C prevent using... Way too much complicated. And if the lyrics are "interlaced" in the frames, one would have ot read through the entire mp3 to find all the lyrics.

As Bitt says above, we don't care where the SYLT frames are, they can appear anywhere. id3lib handles most of the mechanics of reading/writing ID3 frames, including the SYLT frame (though you need to format the timestamps and lines of text in a pretty strict way.)

Once you release your program with source, I'll be happy to give ID3v2 SYLT a shot. I made a bit of progress several months ago. Thing is, we need to do both reading and writing, because I've never found a tagger that's implementing SYLT frame writing, except in raw form, where you need to encode the timestamps and headers yourself anyway.

I realize why you chose the tag format you did, and that format is perfectly fine for testing it out and getting it running. But there are a lot of audiophiles around here (obviously!) who aren't going to want gaps in between our songs (which these tags create.) I think ID3v2 SYLT frame support would give this already awesome program a much wider audience.

If you want to make your source available now, I have some free time in the next week or two before my classwork gets really heavy... Don't worry about it being ugly or half-working, you haven't seen ugly, unfinished code until you've seen my emptriv or empwake code. If your program is in any working condition, I think a few hackers around here would love the chance to contribute.