Extracting Patterns from Guitar Accompaniment Data: Some Experimental Results

Ernesto Trajano de Lima

Centro de Informática (CIn) ­ Universidade Federal de Pernambuco
Caixa Postal 7851­ 50732-970 ­ Recife ­ Pernambuco ­ Brazil

Søren Tjagvad Madsen

Austrian Research Institute for Artificial Intelligence (ÖFAI)
Freyung 6/6, A-1010, Vienna, Austria
Department of Computational Perception ­ University of Linz
Austria

Márcio Dahia

Centro de Informática (CIn) ­ Universidade Federal de Pernambuco
Caixa Postal 7851­ 50732-970 ­ Recife ­ Pernambuco ­ Brazil

Gerhard Widmer

Austrian Research Institute for Artificial Intelligence (ÖFAI)
Freyung 6/6, A-1010, Vienna, Austria
Department of Computational Perception ­ University of Linz
Austria

Geber Ramalho



Abstract:

The analysis of expressive performance, an important research topic in Computer Music, is almost exclusively devoted to the study of western classical piano music. Instruments like the acoustic guitar and styles like Bossa Nova and Samba have almost never been studied, despite their harmonic and rhythmic richness. This paper describes some experimental results obtained with the extraction of rhythmic patterns from the guitar accompaniment of Bossa Nova songs. The performances, recorded with a MIDI guitar, were represented as strings and processed by three different string matchers. Results obtained showed that parts of a previously acquired catalogue of patterns could be found in the data set. Surprising results, such as very long patterns, also showed up.

Download paper: [pdf] 110244 bytes.

Download presentation slides: [pdf] 361991 bytes.