Meta-learning-based percussion transcription and $t\bar{a}la$ identification from low-resource audio

TL;DR

This paper presents a meta-learning approach using MAML for low-resource percussion transcription and tabla identification, enabling rapid adaptation and improved accuracy in polyphonic and limited-data scenarios.

Contribution

It introduces a novel meta-learning framework for percussion transcription and $tar{a}la$ identification, addressing data scarcity and heterogeneity challenges.

Findings

Outperforms existing methods in low-resource settings

Effective for both Indian and Western percussion transcription

Demonstrates robustness across various datasets

Abstract

This study introduces a meta-learning-based approach for low-resource Tabla Stroke Transcription (TST) and $t\bar{a}la$ identification in Hindustani classical music. Using Model-Agnostic Meta-Learning (MAML), we address the challenges of limited annotated datasets and label heterogeneity, enabling rapid adaptation to new tasks with minimal data. The method is validated across various datasets, including tabla solo and concert recordings, demonstrating robustness in polyphonic audio scenarios. We propose two novel $t\bar{a}la$ identification techniques based on stroke sequences and rhythmic patterns. Additionally, the approach proves effective for Automatic Drum Transcription (ADT), showcasing its flexibility for Indian and Western percussion music. Experimental results show that the proposed method outperforms existing techniques in low-resource settings, significantly contributing to music transcription and studying musical traditions through computational tools.