Data representations for audio-to-score monophonic music transcription

作者：

Highlights：

• End-to-end method to perform monophonic music transcription using neural networks.

• Input is an audio waveform, output is a music score.

• Thorough evaluation of different input and output representations to solve the task.

摘要

•End-to-end method to perform monophonic music transcription using neural networks.•Input is an audio waveform, output is a music score.•Thorough evaluation of different input and output representations to solve the task.

论文关键词：Automatic music transcription,Audio processing,Neural networks,Audio to score,Monophonic music

论文评审过程：Received 7 April 2020, Revised 24 June 2020, Accepted 15 July 2020, Available online 26 July 2020, Version of Record 2 August 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113769