The normal way of reducing the level of specific frequencies in a sound is to use filters. But the ear sometimes acts like a filter on its own. By adding simple sustained tones to a signal, I can get achieve a kind of psychoacoustic filtering since neighbouring frequencies will be masked. The experience for the listener is completely different, but in a sort of cynical way I suppose the effect is the same…
In knock out I start with a recording made on the beach at Scheveningen in the mid-afternoon on the 16th of November, 2008. I add two sine tones, f1 and f2, such that f1-f2=10.8hz and (f1+f2)/2=~60hz. Then I add the 7th, 8th, and 9th harmonics of the mean tone, and a final tone somewhere in the range between 3000-5000hz depending on the characteristics of the diffusion system and the room. The levels should be adjusted live to maximize the beating in the low range and to keep the middle and higher frequencies around the threshold of audibility depending on the level of the recording. In any case, the sustained tones will tend to sound louder simply because they remain constant in relation the extreme changes in the recording.
excerpt [4m53s]
