Jitesh Kumar et al
Birla Institute of Technology, India

The audio available electronically presents considerable challenges for information retrieval. As an indication of the magnitude of such a collection currently the number of audio CD tracks in retail is approximately 3 million. There is a need to annotate audio items with descriptors in order to facilitate faster retrieval. This paper describes a comparative evaluation of two fuzzy-derived techniques for automatic audio genre classification. The first model is implemented using conventional fuzzy-based paradigm (FIS) where human expertise and operator knowledge are used to select the parameters for the system. The second model used an adaptive Neuro-fuzzy inference system (ANFIS), where automatic adjustment of the system parameters is affected by a neural networks based on prior knowledge. Three operating parameters were used as inputs to the model, namely Standard deviation, Mean absolute deviation and Median absolute deviation. The models derived using the two techniques were validated using test data that had not been used during training. The ANFIS model was shown to achieve an improved accuracy compared to a pure fuzzy model, based on conveniently selected parameters.