Feature extraction of music signals are often required in applications including; identification and classification of music files over the web or a database. The feature observations for different classes are expected to be distict, and also their computationally efficient implementations to be available. This is necessary both for offline and real time data processing. It is currently an active research field to improve available methods. The need for accurate analysis of a complete music tune conflicts with the time and computation efficiency requirements. Music theory offers new opportunities for feature extraction, especially the intervals systems which reflect the music's characteristics; historical background, geography and culture. Assuming that a standard database would consists of various type of music tunes, features related to the interval systems provide vast amount of opportunities in signal processing. In this work, both the interval systems using just tuning and equal temperaments are analyzed. The computed frequency ratios are tabulated, and proposed to be used for developing feature extraction problems.
展开▼
