Decision-Based Transcription of Jazz Guitar Solos Using a Harmonic Bident Analysis Filter Bank and Spectral Distribution Weighting

TL;DR

This paper introduces a simple, fast, and instrument-agnostic decision-tree based transcriber for jazz guitar solos that significantly improves pitch detection accuracy over existing systems.

Contribution

It presents a novel decision-tree transcriber utilizing a harmonic bident analysis filter bank and spectral distribution weighting, with no instrument-specific modeling or prior note profile learning.

Findings

Achieved 34% improvement in F-measure over the reference system.

Reduced erroneous pitch detections by over 50%.

Performed well on jazz guitar solos, with potential for other instruments.

Abstract

Jazz guitar solos are improvised melody lines played on one instrument on top of a chordal accompaniment (comping). As the improvisation happens spontaneously, a reference score is non-existent, only a lead sheet. There are situations, however, when one would like to have the original melody lines in the form of notated music, see the Real Book. The motivation is either for the purpose of practice and imitation or for musical analysis. In this work, an automatic transcriber for jazz guitar solos is developed. It resorts to a very intuitive representation of tonal music signals: the pitchgram. No instrument-specific modeling is involved, so the transcriber should be applicable to other pitched instruments as well. Neither is there the need to learn any note profiles prior to or during the transcription. Essentially, the proposed transcriber is a decision tree, thus a classifier, with a depth of 3. It has a (very) low computational complexity and can be run on-line. The decision rules can be refined or extended with no or little musical education. The transcriber's performance is evaluated on a set of ten jazz solo excerpts and compared with a state-of-the-art transcription system for the guitar plus PYIN. We achieve an improvement of 34% w.r.t. the reference system and 19% w.r.t. PYIN in terms of the F-measure. Another measure of accuracy, the error score, attests that the number of erroneous pitch detections is reduced by more than 50% w.r.t. the reference system and by 45% w.r.t. PYIN.