Applying machine learning to primate bioacoustics: Review and perspectives.

Area of Science:

• Bioacoustics
• Computational Linguistics
• Machine Learning

Background:

• Primate vocal communication analysis relies on signal processing.
• Emerging passive acoustic monitoring generates large datasets.
• Machine learning offers advanced analytical capabilities.

Purpose of the Study:

• To review computational bioacoustics and signal processing in primate vocal communication.
• To explore machine learning (ML) and deep learning (DL) applications.
• To discuss challenges and future directions in automated analysis.

Main Methods:

• Review of signal and speech processing techniques.
• Exploration of supervised and self-supervised ML models.
• Analysis of passive acoustic monitoring data.

Main Results:

• ML and DL methods show significant potential for analyzing large-scale primate vocal data.
• Automated analysis aids in understanding animal communication and comparative linguistics.
• Identified challenges in data collection and annotation.

Conclusions:

• Computational bioacoustics and ML are crucial for primate communication research.
• Addressing data challenges is key to unlocking further insights.
• The field offers numerous opportunities for future research and innovation.