Ketos—A deep learning package for creating acoustic detectors and classifiers

Abstract

Passive acoustic monitoring is a useful technique for studying aquatic animals, but sustained observing systems require automated algorithms for detecting and classifying sounds of interest. In the last decade, deep neural networks have proven highly successful at solving a wide range of pattern recognition tasks, and recently, we have seen the first promising applications of deep neural networks to detection and classification tasks in marine bioacoustics. Deep neural networks exhibit a high degree of versatility and adaptability: the same network architecture can be trained to accomplish a multitude of tasks by feeding appropriate training data to the network without the need to modify the underlying algorithm. Thus, neural networks have the potential to transform our approach to developing acoustic detection and classification programs, enabling researchers in the field to develop or re-purpose their own programs. MERIDIAN is contributing towards this goal through the development of the open-source Python package Ketos, which provides a high-level programming interface for building training datasets and developing neural network based detectors and classifiers for analyzing underwater acoustics data. In this contribution, an overview of the software package will be given and its functionalities will be demonstrated through case studies.

OpenSoundscape: An open‐source bioacoustics analysis package for Python

“…The most powerful and widely adopted machine learning tools for deep learning are TensorFlow and PyTorch, both of which require advanced knowledge of Python and machine learning to use effectively. While there are several bioacoustics software tools that aim to provide high‐level APIs that simplify the process of training deep learning algorithms for bioacoustic detection (Ketos, Kirsebom et al, 2021; Koogu, Madhusudhana, 2022; aviaNZ, Marsland et al, 2019; ANIMAL‐SPOT, Bergler et al, 2022; gibbonfindR, Clink & Klinck, 2019; soundClass, Silva et al, 2022), we believe that the flexibility provided by OpenSoundscape will allow the package to be applied to a wider range of bioacoustics problems than is possible with existing software. For example, in the two Python packages in the list above (Ketos and Koogu), customizing data augmentation or machine learning model architecture requires the user to understand the underlying package TensorFlow.…”