From doing some experiments I've found that only using the FFT bins close to the harmonic frequencies instead of all bins between frequencies, and then increasing the energy to the power of 1.2 or 1.3, to artificially increase the dynamic range, the resulting output sounds a bit more natural (to me at least),with less background hiss evident. Anyway, here is the small diff on quantize.c:
1035c1035 < int am,bm; /* limits of current band */ --- > int am,bm,b; /* limits of current band */ 1076a1077,1078 > b = floor((m )*model->Wo/r + 0.5); > am=b-1; bm=b+2; 1081c1083 < Am = sqrt(Em); --- > Am = sqrt(Em); Am=powf(Am,1.3)/8.0; Just thought I'd share. ------------------------------------------------------------------------------ LogMeIn Central: Instant, anywhere, Remote PC access and management. Stay in control, update software, and manage PCs from one command center Diagnose problems and improve visibility into emerging IT issues Automate, monitor and manage. Do more in less time with Central http://p.sf.net/sfu/logmein12331_d2d _______________________________________________ Freetel-codec2 mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freetel-codec2
