Piano Transcription in the Studio Using an Extensible Alternating Directions Framework

TL;DR

This paper presents a novel extensible ADMM-based framework for piano transcription that leverages instrument-specific models and additional recording information to significantly improve accuracy beyond previous limits.

Contribution

It introduces a new signal model for pitched percussive instruments and develops a global parameter estimation method within the ADMM framework, enhancing transcription accuracy.

Findings

Achieves 93-95% f-measure for note onsets on Yamaha Disklavier recordings.

Effectively mitigates the 'glass ceiling' effect in piano transcription.

Utilizes instrument-specific data and regularizers to improve transcription performance.

Abstract

Given a musical audio recording, the goal of automatic music transcription is to determine a score-like representation of the piece underlying the recording. Despite significant interest within the research community, several studies have reported on a 'glass ceiling' effect, an apparent limit on the transcription accuracy that current methods seem incapable of overcoming. In this paper, we explore how much this effect can be mitigated by focusing on a specific instrument class and making use of additional information on the recording conditions available in studio or home recording scenarios. In particular, exploiting the availability of single note recordings for the instrument in use we develop a novel signal model employing variable-length spectro-temporal patterns as its central building blocks - tailored for pitched percussive instruments such as the piano. Temporal dependencies between spectral templates are modeled, resembling characteristics of factorial scaled hidden Markov models (FS-HMM) and other methods combining Non-Negative Matrix Factorization with Markov processes. In contrast to FS-HMMs, our parameter estimation is developed in a global, relaxed form within the extensible alternating direction method of multipliers (ADMM) framework, which enables the systematic combination of basic regularizers propagating sparsity and local stationarity in note activity with more complex regularizers imposing temporal semantics. The proposed method achieves an f-measure of 93-95% for note onsets on pieces recorded on a Yamaha Disklavier (MAPS DB).