Onsets and Frames: Dual-Objective Piano Transcription

TL;DR

This paper introduces a dual-objective neural network for polyphonic piano transcription that jointly predicts onsets and frames, significantly improving accuracy and producing more natural transcriptions.

Contribution

The novel model jointly predicts onsets and frames, using onset predictions to condition framewise pitch detection, leading to substantial accuracy improvements.

Findings

Over 100% relative improvement in note F1 score on MAPS dataset

Enhanced transcription naturalness through velocity prediction

Effective joint modeling of onsets and frames

Abstract

We advance the state of the art in polyphonic piano music transcription by using a deep convolutional and recurrent neural network which is trained to jointly predict onsets and frames. Our model predicts pitch onset events and then uses those predictions to condition framewise pitch predictions. During inference, we restrict the predictions from the framewise detector by not allowing a new note to start unless the onset detector also agrees that an onset for that pitch is present in the frame. We focus on improving onsets and offsets together instead of either in isolation as we believe this correlates better with human musical perception. Our approach results in over a 100% relative improvement in note F1 score (with offsets) on the MAPS dataset. Furthermore, we extend the model to predict relative velocities of normalized audio which results in more natural-sounding transcriptions.