Our low-power analog integrated circuit implements a biologically insp
ired algorithm for the spectral analysis of sound. The chip features a
n efficient interface to digital systems; preserving analog processing
's low-power, high-density advantages requires careful attention to in
terface issues. To send the spectral representation off chip, we gener
ate a sparse coding of the output spectrum, and communicate the code a
s an asynchronous stream of events. We store parameters for the spectr
al analysis algorithm as charge on floating nodes, and support the mod
ification of these parameters via Fowler-Nordheim tunneling, under the
control of a digital interface. A prototype system we have developed
uses this chip as a preprocessor.