Skip to main content

Topic: Reverse engineering the best AAC psycoacoustic models (Read 4011 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
  • enry2k
  • [*][*]
Reverse engineering the best AAC psycoacoustic models
There is a lack of good quality open-source AAC encoders and the good ones rely on effective psycoacoustic models. I was wondering whether it was possible to emulate the successful psycoacoustic models by comparing the spectrum of original wave with different decoded signals of specific difficult tracks.
Maybe one could find out what makes Apple model so good compared to Faac in order to use similar strategies.

Is this idea pure nonsense?
Regards
Enrico

  • saratoga
  • [*][*][*][*][*]
Reverse engineering the best AAC psycoacoustic models
Reply #1
It would be easier to just develop your own encoder. The principles behind these models are well understood. I think there is no good open source aac encoder because there is not much interest in working on one.

  • eahm
  • [*][*][*][*][*]
Reverse engineering the best AAC psycoacoustic models
Reply #2
  • Last Edit: 20 August, 2013, 12:10:56 PM by eahm