Skip to main content
Topic: Fixed model for LAME psychoacoustic model (Read 2696 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Fixed model for LAME psychoacoustic model

Hi, one more question :-)

Has anybody tried to use a fixed (i.e., non FFT dependent psychoacoustic model) with LAME?

I can see that toolame has one for Layer 2, inspired by Hyen-O Oh et al "Low power mpeg audio
encoders using simplified psychoacoustic model and fast bit allocation" IEEE Trans on Consumer
Electronics v47 n3 August 2001. p613.

107: /* fixed psycho acoustic model. Values of SNR taken from the 'toolame'
108:    project */
109: static const float fixed_smr[SBLIMIT] =  {
110:      30,  17,  16,  10,  3,  12,    8,    2.5,
111:      5,  5,  6,  6,  5,    6,  10,    6,
112:      -4, -10, -21, -30, -42,  -55,  -68,  -75,
113:    -75, -75, -75, -75, -91, -107, -110, -108
114: };

I was wondering if anybody has a similar one ready to be plugged into LAME.

Thanks

 
SimplePortal 1.0.0 RC1 © 2008-2019