Low cost speech detection using Haar-like filtering for sensornet

Abstract

Haar-like filtering based speech detection is proposed as a new and very low calculation cost method for sensornet applications. The simple Haar-like filters having variable filter width and shift width are trained to learn appropriate filter parameters from the training samples to detect speech. Our method yielded speech/nonspeech classification accuracy of 96.93% for the input length of 0.1s. Compared with high performance feature extraction method MFCC (Mel-frequency cepstrum coefficient), the proposed Haar-like filtering can be approximately 85.77% efficient in terms of the amount of add and multiply calculations while capable of achieving the error rate of only 3.03% relative to MFCC.