I have an mp3 file with VBRI header which has 128 seek points and the duration of the mp3 file is 14min.18sec
The hexdump of header is attached in the image.
The file has seek issues ( ex. seeking at the end results in playback to continue and progress bar of totem mediaplayer showing timestamp more than the whole duration of the file.)
I want to know if the seek table itself is wrong in this file as I suspect having 128 SEEK POINTS for a 14minute file with 32853 frames seems less.
Even if there are 128 correct seek points , it should have relatively seeked to some position which is not happening.
For example -
I tried to seek to 14:08 sec. relatively mpegaudioparse is taking 125th position ( 14:08/14:18*128 ≈ 125) from seek table and getting the byteposition to seek as 1572626 ( ≈ 1.5 MB) . The whole file size is around 18MB.
Hence I think seek table is wrong in the mp3 file.