Classification of Acoustic Signal using Neural Network
Yoon-Ho Kim, Byongmin Yoon · Proceedings of KIIT Summer Conference · 2005
In this paper, a neural-based training and classifing of acoustic signal is presented. This approach is some what a fields of application aspects rather than technical problems such as audio codec, MIDI. In the first, acoustic signal is transformmed using DWT so as to extract a acoustic feature parameters such as loudness, pitch, harmonicity and bandwidth. these accoustic parameters are exploited to the input vector of neural perceptron. In addition, Two kinds of learning rule are involved and compared. Experimental results showed that proposed scheme can be apply for tunning the dissonance chord.