Detection of blue whale vocalisations using a temporal-domain convolutional neural network
This work addresses the detection of whale vocalizations for marine biology and conservation, but it is incremental as it applies a known CNN method to a specific domain with some preprocessing and label-propagation enhancements.
The paper tackles the problem of detecting blue whale vocalizations from acoustic submarine recordings using a temporal-domain convolutional neural network, achieving an accuracy of 85.4% and recall of 93.5% on a dataset of 34 real-world recordings.
We present a framework for detecting blue whale vocalisations from acoustic submarine recordings. The proposed methodology comprises three stages: i) a preprocessing step where the audio recordings are conditioned through normalisation, filtering, and denoising; ii) a label-propagation mechanism to ensure the consistency of the annotations of the whale vocalisations, and iii) a convolutional neural network that receives audio samples. Based on 34 real-world submarine recordings (28 for training and 6 for testing) we obtained promising performance indicators including an Accuracy of 85.4\% and a Recall of 93.5\%. Furthermore, even for the cases where our detector did not match the ground-truth labels, a visual inspection validates the ability of our approach to detect possible parts of whale calls unlabelled as such due to not being complete calls.