Probabalistic Neural Network on Higher Order Spectrum Classification

Probabalistic Neural Network on Higher Order Spectrum Classification
Ivan Fanany1, Adi Triyanto, Nurul Hidayat, Benyamin Kusumoputro
Faculty of Computer Science, University of Indonesia
Many tasks in classification and recognition system need to be performed in real time. While there are difficulties due the complexity of considered pattern, i.e., as the pattern becomes more complex the training set of pattern is insufficient to represent what will arise during tests. In our previous study, we characterized the Higher order spectrum (HOS) pattern property and its classification strategy. The HOS pattern is prefered to be
used for some classification tasks, as for their interesting characteristic, i.e, higher dimensionality, Gaussian noise removal, and phase relation quantification. An appropriate feature extraction mechanism is certainly needed in order to quantize this HOS input pattern before classification. However, for a real time classification task, this well-developed feature extraction process should be adhered by a robust and fast classifier. In this study, we present the use of Probabilistic Neural Network(PNN) as Neural Classifier and compared it with conventional Back Propagation (BP) learning scheme. Both memorization and generalization comparison of those network are discussed as for developing a robust and fast speaker identification in noisy environment. It was shown that even using only a small size codebook, PNN can classify bispectrum pattern above 88% success rate and 36 times faster than BP.
Keywords: higher order spectrum, probabilistic neural net, fast speaker identification

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s