This paper describes how tools provided by Natural Language Processing and Information Retrieval can be applied to music. A method of converting complex musical structure to features (n-grams) corresponding with words of text was introduced. Mutual correspondence between both representations was shown by demonstrating certain important regularities known from text processing, which may also be found in music. The problem of automatic composer attribution to which statistical analysis of n-gram profiles known from statistical NLP was applied served as a case study. A MIDI files corpus of piano pieces was chosen as the source of data.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
This study presents the algorithm creating rhythmic hypotheses worked out by the authors, and then addresses the problem of determining its computational complexity. A short review of rhythm extraction methods is presented, first. Then, three phases of the algorithm engineered by the authors, namely creating periods, creating simplified hypotheses and creating full hypotheses are examined. The analyses of computational complexity of the method proposed assume that the engineered method is expected to rank rhythmic hypotheses formed of three rhythmic levels above meter. This proved to be sufficient for providing automatic drum accompaniment for a given melody without delay.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.