About this Research Topic
Recent advancements in wildlife monitoring techniques have resulted in significant changes in the field. Among these cutting-edge and non-invasive approaches is passive acoustic monitoring (PAM), which involves deploying sound recorders in the field followed by interpretation of the recordings. Passive acoustic monitoring is predominantly utilized to survey birds, making them the most frequently studied group of animals. Consequently, our existing understanding of how to effectively monitor and automatically identify bird species from sound recordings is high.
Additionally, the development of automated techniques based on machine and deep learning and audio signal processing algorithms has further transformed bird monitoring, with unprecedented space and time resolution. These advancements enable researchers to monitor various aspects of birds, such as species occupancy, population density and distribution, species diversity, etc. Furthermore, combining PAM with these techniques allows for tracking cryptic and rare species, monitoring birds in otherwise inaccessible environments, and uncovering previously unknown patterns of bird behavior and habits.
However, despite the progress made, there are still many challenges and uncertainties that have to be resolved.
With this Research Topic we aim to provide an overview of the recent developments of new tools, methods, and algorithms for bird automated identification and bird monitoring using acoustic recordings.
Our objective is to enhance bird monitoring programs utilizing PAM by evaluating techniques for bird signal processing and automated software for bird detection. Furthermore, we seek to expand our understanding of the required monitoring effort to ensure reliable bird monitoring, and to assess the effectiveness of this technique in estimating bird occupancy, population density, species diversity, and conducting ecological studies on a broad spatiotemporal scale.
For this Research Topic, we especially welcome papers that focus on:
• The assessment of own developed bird recognizers through machine learning approach (e.g. machine and deep learning both supervised and unsupervised models and algorithms).
• Using innovative spectro-temporal representations or feature embedding for acoustic bird monitoring.
• Evaluate the performance of already developed software for bird automated detection (e.g. BirdNET, Arbimon, Kaleidoscope Pro, etc.).
• Methodological studies assessing the ability of passive acoustics for estimating bird density and/or richness.
• Proposing a common framework/protocol for reliable bird monitoring (e.g. survey coverage, assessment of effective detection radius, number of recording days required, and duration of each recording).
Additionally, the development of automated techniques based on machine and deep learning and audio signal processing algorithms has further transformed bird monitoring, with unprecedented space and time resolution. These advancements enable researchers to monitor various aspects of birds, such as species occupancy, population density and distribution, species diversity, etc. Furthermore, combining PAM with these techniques allows for tracking cryptic and rare species, monitoring birds in otherwise inaccessible environments, and uncovering previously unknown patterns of bird behavior and habits.
However, despite the progress made, there are still many challenges and uncertainties that have to be resolved.
With this Research Topic we aim to provide an overview of the recent developments of new tools, methods, and algorithms for bird automated identification and bird monitoring using acoustic recordings.
Our objective is to enhance bird monitoring programs utilizing PAM by evaluating techniques for bird signal processing and automated software for bird detection. Furthermore, we seek to expand our understanding of the required monitoring effort to ensure reliable bird monitoring, and to assess the effectiveness of this technique in estimating bird occupancy, population density, species diversity, and conducting ecological studies on a broad spatiotemporal scale.
For this Research Topic, we especially welcome papers that focus on:
• The assessment of own developed bird recognizers through machine learning approach (e.g. machine and deep learning both supervised and unsupervised models and algorithms).
• Using innovative spectro-temporal representations or feature embedding for acoustic bird monitoring.
• Evaluate the performance of already developed software for bird automated detection (e.g. BirdNET, Arbimon, Kaleidoscope Pro, etc.).
• Methodological studies assessing the ability of passive acoustics for estimating bird density and/or richness.
• Proposing a common framework/protocol for reliable bird monitoring (e.g. survey coverage, assessment of effective detection radius, number of recording days required, and duration of each recording).
Keywords: Audio signal processing, autonomous recording units, automated detection, bioacoustics, bird monitoring, occupancy, passive acoustic monitoring, signal recognition, deep learning, soundscape monitoring, embedding features
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
