An Improved Pattern-Matching Method for Piano Multipitch Detection
A previous method presented by the authors carried out multi-pitch piano sound identification by using a pattern-matching process. In that method, the identification required, besides the matching-metric calculation, both a spectral predetection process and a validation step. Predetection allowed to select a subset out of the eighty-eight patterns, whereas the validation verified whether the detected note were actually in the analyzed spectrum. Both highly increased the true-positive detections ratio, but they imposed restrictions to identification of complex real sounds (e.g., two-hands playing). This paper presents an improvement in the method that allows to get rid of both, predetection and validation, by using a modified matching-metric algorithm. This work has been supported by the Spanish National Project TEC2006-13067-C03-01/TCM.
Click to purchase paper or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $20 for non-members, $5 for AES members and is free for E-Library subscribers.