Improved Acoustic Modeling for Automatic Piano Music Transcription Using Echo State Networks

Research output: Contribution to book/Conference proceedings/Anthology/ReportChapter in book/Anthology/ReportContributedpeer-review

Contributors

Abstract

Automatic music transcription (AMT) is one of the challenging problems in Music Information Retrieval with the goal of generating a score-like representation of a polyphonic audio signal. Typically, the starting point of AMT is an acoustic model that computes note likelihoods from feature vectors. In this work, we evaluate the capabilities of Echo State Networks (ESNs) in acoustic modeling of piano music. Our experiments show that the ESN-based models outperform state-of-the-art Convolutional Neural Networks (CNNs) by an absolute improvement of 0.5 F1 -score without using an extra language model. We also discuss that a two-layer ESN, which mimics a hybrid acoustic and language model, achieves better results than the best reference approach that combines Invertible Neural Networks (INNs) with a biGRU language model by an absolute improvement of 0.91 F1 -score.

Details

Original languageEnglish
Title of host publicationAdvances in Computational Intelligence
PublisherSpringer Verlag
Number of pages12
Publication statusPublished - 21 Aug 2021
Peer-reviewedYes

External IDs

Scopus 85115199523
ORCID /0000-0003-0167-8123/work/167214850

Keywords

Keywords

  • Acoustic modeling, Automatic piano transcription, Echo state network