Knowledge Distillation for Singing Voice Detection

TL;DR

This paper explores using knowledge distillation to compress large neural networks for singing voice detection, enabling deployment on resource-limited devices while maintaining high accuracy.

Contribution

It introduces the application of knowledge distillation techniques to compress SVD models, addressing deployment challenges in music information retrieval.

Findings

Knowledge distillation effectively reduces model size.

Compressed models retain high detection accuracy.

Ensemble distillation improves performance further.

Abstract

Singing Voice Detection (SVD) has been an active area of research in music information retrieval (MIR). Currently, two deep neural network-based methods, one based on CNN and the other on RNN, exist in literature that learn optimized features for the voice detection (VD) task and achieve state-of-the-art performance on common datasets. Both these models have a huge number of parameters (1.4M for CNN and 65.7K for RNN) and hence not suitable for deployment on devices like smartphones or embedded sensors with limited capacity in terms of memory and computation power. The most popular method to address this issue is known as knowledge distillation in deep learning literature (in addition to model compression) where a large pre-trained network known as the teacher is used to train a smaller student network. Given the wide applications of SVD in music information retrieval, to the best of our knowledge, model compression for practical deployment has not yet been explored. In this paper, efforts have been made to investigate this issue using both conventional as well as ensemble knowledge distillation techniques.