An empirical investigation into audio pipeline approaches for classifying bird species

TL;DR

This paper empirically compares traditional DNN and convolutional neural network approaches for bird species audio classification, focusing on transfer learning, data augmentation, and model optimization for edge device deployment.

Contribution

It provides empirical insights into the effectiveness of DNN versus CNN models for bird audio classification on edge devices.

Findings

CNN outperforms DNN in accuracy

Data augmentation improves model robustness

Transfer learning reduces training time

Abstract

This paper is an investigation into aspects of an audio classification pipeline that will be appropriate for the monitoring of bird species on edges devices. These aspects include transfer learning, data augmentation and model optimization. The hope is that the resulting models will be good candidates to deploy on edge devices to monitor bird populations. Two classification approaches will be taken into consideration, one which explores the effectiveness of a traditional Deep Neural Network(DNN) and another that makes use of Convolutional layers.This study aims to contribute empirical evidence of the merits and demerits of each approach.