Western Musical Instruments Sound Eigenvalue Extraction

In recent years, rapid advances in digital music creation, collection and storage technology have enabled organizations to accumulate vast amounts of musical audio data. The booming of multimedia resources from the Internet brought a tremendous need to provide new, more advanced tools for the ability to query and process vast quantities of musical data, since searching through multimedia data is a highly nontrivial task requiring content based indexing of the data, which are not easy to describe with mere symbols. There are several practical areas where the outcomes of this research can be applied.. Music File Annotation, Music Transcription, Structured-Audio Encoding, Instrument Music Recommendation Engine. Recognition and separation of sounds played by various instruments is very useful in labeling audio files with semantic information. This is a non-trivial task requiring sound analysis, but the results can aid automatic indexing and browsing music data when searching for melodies played by user specified instruments. Numerous approaches on acoustic feature extraction have already been proposed for timbre recognition. Unfortunately, none of those monophonic timbre estimation algorithms can be successfully applied.Trimbre recognition remains an unsolved problem in polyphonic sounds, which are the more usual cases in the real music world.In this dissertation, I develped new temporal acoustic features bases on MPEG-7 instantaneous 13 spectral features to improve the discriminating ability of the classifier for some musical instruments that share the similar pattern in spectral space but unique characteristics in short term temporal feature space. The experimental results including Recall, Precision, F-scror show that the improved features based on MPEG-7 have higher recognition rate. I use J48 and KNN algorithms provide by Weka to conduct experiments.