Abstract:
The large quantities of information in both
audio and video is a great need for efficient ways of
searching and retrieving relevant information. The
goal of an audio indexing is to provide the capability of
searching and browsing through audio content. This
work presents an effort to discriminate Myanmar
ethnic traditional style music based on the large
number of audio features(timbre, rhythm, pitch) are
evaluated for their suitability in such a classification
task, including well-known physical and perceptual
features, audio descriptors. For this purpose, we use
the Multi Band Autocorrelation Peak (MBAP) features,
extracted in multiple bands providing complementary
information and give a better description of the audio
signal and then use the classifier based on Training
and Testing audio data to get class labels .The system
have selected more popular cultural music styles of
Myanmar ethnic music namely: Kachin, Kayin, Shan
and Yakhine ethnic group songs for experiments.
