Abstract: Generating musical audio directly with neural networks is notoriously difficult because it requires coherently modeling structure at many different timescales. Fortunately, most music is also highly structured and can be represented as discrete note events played on musical instruments. Herein, we show that by using notes as an intermediate representation, we can train a suite of models capable of transcribing, composing, and synthesizing audio waveforms with coherent musical structure on timescales spanning six orders of magnitude (~0.1 ms to ~100 s), a process we call Wave2Midi2Wave. This large advance in the state of the art is enabled by our release of the new MAESTRO (MIDI and Audio Edited for Synchronous TRacks and Organization) dataset, composed of over 172 hours of virtuosic piano performances captured with fine alignment (~3 ms) between note labels and audio waveforms. The networks and the dataset together present a promising approach toward creating new expressive and interpretable neural models of music.
Keywords: music, piano transcription, transformer, wavnet, audio synthesis, dataset, midi
TL;DR: We train a suite of models capable of transcribing, composing, and synthesizing audio waveforms with coherent musical structure, enabled by the new MAESTRO dataset.
Code: [![Papers with Code](/images/pwc_icon.svg) 4 community implementations](https://paperswithcode.com/paper/?openreview=r1lYRjC9F7)
Data: [MAESTRO](https://paperswithcode.com/dataset/maestro), [MusicNet](https://paperswithcode.com/dataset/musicnet)
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 3 code implementations](https://www.catalyzex.com/paper/arxiv:1810.12247/code)