This paper provides a computationally efficient noise filtering technique for reducing stepper motor noise energy in a Digital Still Camera application without significantly altering the speech signal energy. The filtering technique uses a cascade of bi-quadratic filters to improve the signal to noise ratio by reducing noise energy. The noise filter described in this paper uses an Automatic Gain Controller stage to maintain signal energy when the noise filter is active. Such a filter achieves about 10 dB PSNR improvement at 2 MHz of ARM9 processor load. Furthermore, the paper briefly explores the filter structure for efficient hardware or software realization.