Hi S, you would need to compensate by using the Fletcher-Munson equal-loudness contours – these could be put into tables or calculated by functions? I believe a lot of this stuff was done already – I think Alex Porres has translated the Slaney Toolbox into Pd? There's also the roughness stuff done by Pressnitzer, and a bunch of externals associated with that – probably a bit of loudness stuff in there. The main thing to appreciate is that the ear is remarkably non-linear (the price we pay for having such a wide dynamic range!), and that the Fletcher-Munson equal-loudness contours change as average intensity increases–we hear a much different spectrum in a noisy environment than we do in a quiet environment. Of course, the F-M e-l contours are a bit simplistic as later research into the role of the the outer hair cells proves. Still, they're a good starting point. Sounds like a cool project, I'd like to hear a sine sweep track at equal perceptual intensity from 20 Hz to 20 kHz :) Cheers, Rich