Transfer of knowledge among instruments in automatic music transcription
This work addresses the problem of computational efficiency in automatic music transcription for researchers and practitioners, though it is incremental as it builds on existing transfer learning and synthesized data methods.
The paper tackles the high computational cost of training automatic music transcription models by using synthesized audio data from software synthesizers to train a universal model, achieving results that show this approach can serve as a good base for pretraining general-purpose models adaptable to multiple instruments.
Automatic music transcription (AMT) is one of the most challenging tasks in the music information retrieval domain. It is the process of converting an audio recording of music into a symbolic representation containing information about the notes, chords, and rhythm. Current research in this domain focuses on developing new models based on transformer architecture or using methods to perform semi-supervised training, which gives outstanding results, but the computational cost of training such models is enormous. This work shows how to employ easily generated synthesized audio data produced by software synthesizers to train a universal model. It is a good base for further transfer learning to quickly adapt transcription model for other instruments. Achieved results prove that using synthesized data for training may be a good base for pretraining general-purpose models, where the task of transcription is not focused on one instrument.