Academic
Publications
Automatic Chord Transcription with Concurrent Recognition of Chord Symbols and Boundaries

Automatic Chord Transcription with Concurrent Recognition of Chord Symbols and Boundaries,Takuya Yoshioka,Tetsuro Kitahara,Kazunori Komatani,Tetsuya O

Automatic Chord Transcription with Concurrent Recognition of Chord Symbols and Boundaries   (Citations: 27)
BibTex | RIS | RefWorks Download
This paper describes a method that recognizes musical chords from real-world audio signals in compact-disc recordings. The automatic recognition of musical chords is necessary for music information retrieval (MIR) sys- tems, since the chord sequences of musical pieces cap- ture the characteristics of their accompaniments. None of the previous methods can accurately recognize musi- cal chords from complex audio signals that contain vocal and drum sounds. The main problem is that the chord- boundary-detection and chord-symbol-identification pro- cesses are inseparable because of their mutual depen- dency. In order to solve this mutual dependency problem, our method generates hypotheses about tuples of chord symbols and chord boundaries, and outputs the most plau- sible one as the recognition result. The certainty of a hy- pothesis is evaluated based on three cues: acoustic fea- tures, chord progression patterns, and bass sounds. Ex- perimental results show that our method successfully rec- ognized chords in seven popular music songs; the average accuracy of the results was around 77%.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
Sort by: