Leveraging Hierarchical Structures for Few-Shot Musical Instrument Recognition

Abstract

Deep learning work on musical instrument recognition has generally focused on instrument classes for which we have abundant data. In this work, we exploit hierarchical relationships between instruments in a few-shot learning setup to enable classification of a wider set of musical instruments, given a few examples at inference. We apply a hierarchical loss function to the training of prototypical networks, combined with a method to aggregate prototypes hierarchically, mirroring the structure of a predefined musical instrument hierarchy. These extensions require no changes to the network architecture and new levels can be easily added or removed. Compared to a non-hierarchical few-shot baseline, our method leads to a significant increase in classification accuracy and significant decrease mistake severity on instrument classes unseen in training.

References in corpus (6)

• Prototypical Networks for Few-shot Learning
• An Analysis of Hierarchical Text Classification Using Word Embeddings
• Hierarchical Multitask Learning for CTC-based Speech Recognition
• Augmentation Methods on Monophonic Audio for Instrument Classification in Polyphonic Music
• Frame-level Instrument Recognition by Timbre and Pitch
• Hierarchical learning for DNN-based acoustic scene classification

Cited by in corpus (3)

• The Sound Demixing Challenge 2023 $\unicode{x2013}$ Music Demixing Track
• Rank-based loss for learning hierarchical representations
• LC-Protonets: Multi-Label Few-Shot Learning for World Music Audio Tagging