Exploring Procedural Data Generation for Automatic Acoustic Guitar Fingerpicking Transcription
This addresses data scarcity for researchers and practitioners in music information retrieval, but it is incremental as it builds on existing methods like CRNN and physical modeling.
The paper tackles the challenge of automatic acoustic guitar fingerpicking transcription by proposing a procedural data generation pipeline to overcome data scarcity, achieving reasonable note-tracking results and showing that finetuning with a small amount of real data improves accuracy over models trained only on real recordings.
Automatic transcription of acoustic guitar fingerpicking performances remains a challenging task due to the scarcity of labeled training data and legal constraints connected with musical recordings. This work investigates a procedural data generation pipeline as an alternative to real audio recordings for training transcription models. Our approach synthesizes training data through four stages: knowledge-based fingerpicking tablature composition, MIDI performance rendering, physical modeling using an extended Karplus-Strong algorithm, and audio augmentation including reverb and distortion. We train and evaluate a CRNN-based note-tracking model on both real and synthetic datasets, demonstrating that procedural data can be used to achieve reasonable note-tracking results. Finetuning with a small amount of real data further enhances transcription accuracy, improving over models trained exclusively on real recordings. These results highlight the potential of procedurally generated audio for data-scarce music information retrieval tasks.