I think this message bounced or something, I'm mailing it to the
mailinglist..
/jp
---------- Forwarded message ----------
From: Mark Taylor <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Subject: help with short_block switching formulas ???
Reply-to: [EMAIL PROTECTED]
Hi Everyone,
Here's something where it would be great to get some new ideas.
I know some pretty knowledgeable people read this mailing list,
so how giving us some hints on what the commercial encoders do???
Probably the biggest problem with LAME is the short block switching
formula. Several people have sent me samples where LAME misses some
noticable pre-echo events that FhG picks up and sounds great.
A frame by frame analysis shows that most of this is due to FhG
switching to short blocks at the correct frame, and LAME just
uses long blocks.
LAME picks up some of these pre-echo events, but misses the more
subtle or masked ones. (The ISO encoder basically misses *all*
pre-echo events because of bugs.)
The short_block switching criterion used in LAME is from the ISO code/docs:
pe (perceptial entropy) > 1800.
I'm pretty sure FhG does not use this formula, since they sometimes
switch to short_blocks at frames where the pe is quite low. so the
questions are:
1. is there a more accurate pe formula than the one in the ISO docs?
2. is there a better short_block switching criterion?
Some MPEG1 and MPEG2 papers claim the pre-echo detection is from
B. Edler, "Coding of audio signals with overlapping block
transform and adaptive window functions" (in German) Frequenz,
vol 43 1989 pp 252-256.
Maybe someone in Germany could check this out?
If this paper just gives the pe>1800 formula, it's not that usefull,
but maybe it contains something better?
Mark
