Can anyone shortly descripe what the difference in using FFT as opposed to MDCT for transforming between time- and frequencydomain? Is there one or two large pros in using one over the other. I.e. the MPEG-1 Psychoacoustic Model-1 uses a FFT (with a complex expontial series) to perform the transformation, but why not use purely MDCT for this?
Difference in transforms in psychoacoustic model
Started by ●January 14, 2009
