<![CDATA[Almost allmusic genreuse guitaras its instrument.Toproducea harmonicguitarvoice needs guitar chords mastery. However, only few peopleareable todistinguish guitar chords. This paper is addressed to develop a computational model to convert guitar voice into appropriate cord. In this research, we use Mel Frequency Cepstrum Coefficient (MFCC) as feature extraction because thistechniqueis oftenusedfor voice processing and good enough in presenting thecharacteristics ofasignal voice. Probabilistic Neural Network (PNN) is implemented to classify the feature into one out of 24 classes of cord. We record 345 for each cord (totally we have 8640 recording data with WAV format). Experimenst are conducted for some number of cepstral coefficients (13, 26, 39 and 52), with 100 millisecond as time frame and 40% overlapping between successive frame. According to the experiment, the maximum accuracy is 94.31% for 52 number of cepstral coefficients.]]>
Copyrights © 2013
