Towards Human-in-the-Loop Onset Detection: A Transfer Learning Approach for Maracatu
This addresses the challenge of underrepresented musical traditions in music information retrieval, offering an efficient human-in-the-loop approach with minimal annotation effort.
The paper tackles onset detection in Afro-Brazilian Maracatu music by adapting Temporal Convolutional Networks through transfer learning, achieving F1 scores up to 0.998 and improvements over 50 percentage points in some cases.
We explore transfer learning strategies for musical onset detection in the Afro-Brazilian Maracatu tradition, which features complex rhythmic patterns that challenge conventional models. We adapt two Temporal Convolutional Network architectures: one pre-trained for onset detection (intra-task) and another for beat tracking (inter-task). Using only 5-second annotated snippets per instrument, we fine-tune these models through layer-wise retraining strategies for five traditional percussion instruments. Our results demonstrate significant improvements over baseline performance, with F1 scores reaching up to 0.998 in the intra-task setting and improvements of over 50 percentage points in best-case scenarios. The cross-task adaptation proves particularly effective for time-keeping instruments, where onsets naturally align with beat positions. The optimal fine-tuning configuration varies by instrument, highlighting the importance of instrument-specific adaptation strategies. This approach addresses the challenges of underrepresented musical traditions, offering an efficient human-in-the-loop methodology that minimizes annotation effort while maximizing performance. Our findings contribute to more inclusive music information retrieval tools applicable beyond Western musical contexts.