A new definition of the distortion matrix for an audio-to-score alignment system

TL;DR

This paper introduces a novel distortion matrix definition for an audio-to-score alignment system based on DTW, utilizing note combinations and spectral pattern learning to improve alignment accuracy.

Contribution

It proposes a new way to define the distortion matrix by arranging score info in note combinations and learning spectral patterns, enhancing DTW-based alignment.

Findings

Improved alignment accuracy with the new distortion matrix.

Effective spectral pattern learning for note combinations.

Enhanced robustness of score following system.

Abstract

In this paper we present a new definition of the distortion matrix for a score following framework based on DTW. The proposal consists of arranging the score information in a sequence of note combinations and learning a spectral pattern for each combination using instrument models. Then, the distortion matrix is computed using these spectral patterns and a novel decomposition of the input signal.