wish: vbr header analysis
Posted: Fri Oct 17, 2003 5:43 pm
i encountered this problem:
the biggest part of my mp3 collection was encoded using musicmatch. unfortunately, musicmatch writes a vbr header in VBRI mode which most players and programs are aware of. so i "repaired" these files with mp3-tag-studio which adds a normal XING vbr header to these files. all players like xmms or winamp and tools like mp3-tag-studio, easy-tag and so on understand these vbr headers but mediamonkey. mediamonkey's player understands these headers too, because the mad-plugin understands these headers (the player shows the right bitrate/length), but the mediamonkey library does not recognize the vbr header and assumes cbr files. you obviously get wrong durations and the vbr header seems to be interpreted as normal mp3-frame. so my wish is to improve the vbr header analysis of the library.
here is a snampshot of vbr-headers written by mp3-tag-studio:
MPEG Version 1, Layer III, Not protected, 48kbps, 44100 Hz, frame is not padded, Stereo ...
FF FB 30 04 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00
now follows the standard XING header that is
58 69 6E 67 00 00 00 0F ...
after the XING header follows the VBRI header which can be ignored. the total size of the frame is 0x02A6 bytes.
another wish would concern vbr headers written by lame. these headers contain a lame tag which has replay-gain information stored. would it be possible to use these gain values (so there is no need for volume analysis for this file) and to store newly calculated gain values in that tag additionaly to the library (so the information is present in the file even if it gets removed from the library) ?
with kind regards, dirk
the biggest part of my mp3 collection was encoded using musicmatch. unfortunately, musicmatch writes a vbr header in VBRI mode which most players and programs are aware of. so i "repaired" these files with mp3-tag-studio which adds a normal XING vbr header to these files. all players like xmms or winamp and tools like mp3-tag-studio, easy-tag and so on understand these vbr headers but mediamonkey. mediamonkey's player understands these headers too, because the mad-plugin understands these headers (the player shows the right bitrate/length), but the mediamonkey library does not recognize the vbr header and assumes cbr files. you obviously get wrong durations and the vbr header seems to be interpreted as normal mp3-frame. so my wish is to improve the vbr header analysis of the library.
here is a snampshot of vbr-headers written by mp3-tag-studio:
MPEG Version 1, Layer III, Not protected, 48kbps, 44100 Hz, frame is not padded, Stereo ...
FF FB 30 04 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00
00 00 00 00
now follows the standard XING header that is
58 69 6E 67 00 00 00 0F ...
after the XING header follows the VBRI header which can be ignored. the total size of the frame is 0x02A6 bytes.
another wish would concern vbr headers written by lame. these headers contain a lame tag which has replay-gain information stored. would it be possible to use these gain values (so there is no need for volume analysis for this file) and to store newly calculated gain values in that tag additionaly to the library (so the information is present in the file even if it gets removed from the library) ?
with kind regards, dirk