High Resolution Guitar Transcription via Domain Adaptation
This work addresses the problem of automatic music transcription for guitar, which is incremental as it refines existing domain adaptation methods for a specific instrument.
The paper tackled the lack of large datasets for guitar transcription by adapting a high-resolution piano transcription model using score-audio pairs, achieving state-of-the-art results on GuitarSet in a zero-shot setting.
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.