| |
Abstract:
The performance of dedicated VLSI neural processing hardware
depends critically on the design of the implemented algorithms. We
have previously proposed an algorithm for acoustic transient
classification. Having implemented and demonstrated this algorithm
in a mixed-mode architecture, we now investigate variants on the
algorithm, using time and frequency channel differencing, input and
output normalization, and schemes to binarize and train the
template values, with the goal of achieving optimal classification
performance for the chosen hardware.
|
