Highland piping ornament recognition using dynamic time warping

This work uses a custom-built digital bagpipe chanter interface to assist in the process of learning the Great Highland Bagpipe (GHB). In this paper, a new algorithm is presented for the automatic recognition and evaluation of the various ornamentation techniques that are a central aspect of traditional Highland bagpipe music. The algorithm is evaluated alongside a previously published approach, and is shown to provide a significant improvement in performance. The ornament detection facility forms part of a complete hardware and software system for use in both tuition and solo practice situations, allowing details of ornamentation errors made by the player to be provided as visual and textual feedback. The system also incorporates new functionality for the identification and description of GHB fingering errors.