High Resolution Guitar Transcription via Domain Adaptation

Author:

Xavier Riley, Drew Edwards, Simon Dixon

Keyword:

Electrical Engineering and Systems Science, Audio and Speech Processing, Audio and Speech Processing (eess.AS), Machine Learning (cs.LG), Sound (cs.SD)

journal:

date:

2024-02-23 00:00:00

Abstract

Automatic music transcription (AMT) has achieved high accuracy for piano due to the availability of large, high-quality datasets such as MAESTRO and MAPS, but comparable datasets are not yet available for other instruments. In recent work, however, it has been demonstrated that aligning scores to transcription model activations can produce high quality AMT training data for instruments other than piano. Focusing on the guitar, we refine this approach to training on score data using a dataset of commercially available score-audio pairs. We propose the use of a high-resolution piano transcription model to train a new guitar transcription model. The resulting model obtains state-of-the-art transcription results on GuitarSet in a zero-shot context, improving on previously published methods.

PDF: High Resolution Guitar Transcription via Domain Adaptation.pdf