The telecommunications industry uses a signaling technique known as dual-tone multifrequency (DTMF) dialing to indicate the depressed pushbuttons on a telephone keypad. The two primary categories for decoding DTMF waveforms include both hardware and software-based solutions. Hardware-based solutions use filter banks to separate frequencies into low-band and high-band channels, and utilize detection schemes in order to reconstruct the pushbutton. These techniques often suffer from high parts count, large circuit board space, and increased costs. Software-based solutions utilize digital signal processors (DSP) with algorithms for implementing Fast Fourier Transforms (FFT) or filter banks. These designs suffer from complex algorithms with a large computational overhead. The new work presented in this paper includes an extension of the simulation results for the design of a neural network based DTMF decoder already developed by the authors.