Learning Frame Similarity using Siamese networks for Audio-to-Score Alignment

TL;DR

This paper introduces a learned frame similarity approach using Siamese networks for offline audio-to-score alignment in piano music, outperforming traditional DTW methods with handcrafted features across various acoustic conditions.

Contribution

The paper presents a novel learned similarity method with Siamese networks that improves alignment accuracy and robustness over traditional DTW-based techniques.

Findings

Higher alignment accuracy than DTW with handcrafted features

Robust performance across different acoustic conditions

Effective adaptation to various domains

Abstract

Audio-to-score alignment aims at generating an accurate mapping between a performance audio and the score of a given piece. Standard alignment methods are based on Dynamic Time Warping (DTW) and employ handcrafted features, which cannot be adapted to different acoustic conditions. We propose a method to overcome this limitation using learned frame similarity for audio-to-score alignment. We focus on offline audio-to-score alignment of piano music. Experiments on music data from different acoustic conditions demonstrate that our method achieves higher alignment accuracy than a standard DTW-based method that uses handcrafted features, and generates robust alignments whilst being adaptable to different domains at the same time.