: I'm looking for simple way to extract MPEG Audio frames from file
: frame by frame (Layer I). Currently I'm using John Fulmer's
: aencode based on ISO sources. It produces file with frames
: separated by sequences of zero bytes min 8 bytes long.
: Is it guaranteed by any standard (MPEG Audio file format)?
: Frame start detection based on syncword (bin 1111 1111 1111)
: recognition won't do - there are many syncword bit pattern
: occurences in frame body.
I've never really studied the bitcontense of an MPEG-stream, but to my
knowledge the syntax was designed to avoid "illegal" syncwords
(1-sequences are avoided when coding bitalloc, scalefactors and in
subbandsamplesquant.). So, something else is wrong ???
Also, normally when (de)coding MPEG audio (layer 1/layer 2) a fixed
bitrate is used; get the bitrate, check the padding bit and CALC. the next
Why use layer 1, when layer 2 performs (much) better ?
feel free to contact me by e-mail, using THE ABOVE address.
(i wrote a thesis on MPEG-audio parameterevaluation and a (partial)
DSP56002 realtime encoding implementation)