Improved Architecture for High-resolution Piano Transcription to Efficiently Capture Acoustic Characteristics of Music Signals

TL;DR

This paper introduces an improved high-resolution piano transcription model that employs advanced neural architectures and input representations to better capture acoustic features, resulting in higher accuracy and smaller models.

Contribution

The paper proposes novel architectures combining CRNN with dilated convolutions and Transformer decoders, enhancing transcription accuracy and reducing model size.

Findings

Achieved consistent improvement in note-level metrics.

Developed smaller models with comparable or better performance.

Demonstrated effectiveness of Constant-Q Transform input representation.

Abstract

Automatic music transcription (AMT), aiming to convert musical signals into musical notation, is one of the important tasks in music information retrieval. Recently, previous works have applied high-resolution labels, i.e., the continuous onset and offset times of piano notes, as training targets, achieving substantial improvements in transcription performance. However, there still remain some issues to be addressed, e.g., the harmonics of notes are sometimes recognized as false positive notes, and the size of AMT model tends to be larger to improve the transcription performance. To address these issues, we propose an improved high-resolution piano transcription model to well capture specific acoustic characteristics of music signals. First, we employ the Constant-Q Transform as the input representation to better adapt to musical signals. Moreover, we have designed two architectures: the first is based on a convolutional recurrent neural network (CRNN) with dilated convolution, and the second is an encoder-decoder architecture that combines CRNN with a non-autoregressive Transformer decoder. We conduct systematic experiments for our models. Compared to the high-resolution AMT system used as a baseline, our models effectively achieve 1) consistent improvement in note-level metrics, and 2) the significant smaller model size, which shed lights on future work.