Abstract :

The Increasing Focus On Biodiversity Preservation And Ecological Monitoring Has Created A Demand For Intelligent, Non-intrusive Systems Capable Of Accurately Recognizing Animal Species. Bioacoustic Analysis, Which Relies On Animal Vocalizations, Serves As An Efficient Alternative To Conventional Visual Monitoring Methods, Especially In Environments With Dense Vegetation Or Limited Visibility. In This Regard, This Study Presents An Advanced Machine Learning Framework For Automatic Animal Species Classification Using Bioacoustic Data. The System Processes Raw Audio Inputs And Extracts A Diverse Range Of Acoustic Features, Including Mel-Frequency Cepstral Coefficients (MFCC), Chroma Features, Mel Spectrogram, Spectral Contrast, Zero-crossing Rate, And Root Mean Square Energy. These Features Effectively Represent The Temporal, Spectral, And Harmonic Characteristics Of Animal Sounds, Allowing Reliable Differentiation Between Species. Several Classification Algorithms, Such As Decision Tree Classifier (DT), Gradient Boosting Classifier (GB), And Nearest Centroid Classifier (NCC), Are Employed To Assess Baseline Performance. To Improve Classification Results And Address The Shortcomings Of Individual Models, A Hybrid Soft Voting Technique Is Proposed By Integrating Support Vector Machine (SVM) And Light Gradient Boosting Machine (LGBM) Through A Probabilistic Voting Mechanism. This Approach Combines The Output Probabilities Of Both Models To Enhance Prediction Reliability And Generalization Capability. Additionally, The System Includes A Graphical User Interface (GUI) That Supports Dataset Handling, Feature Extraction, Model Training, And Real-time Prediction. Experimental Findings Indicate That The Proposed Hybrid Model Delivers Superior Performance In Terms Of Accuracy, Precision, Recall, And F1-score (Harmonic Mean Of Precision And Recall). Overall, The Study Demonstrates The Effectiveness Of Bioacoustic Intelligence As A Scalable And Efficient Solution For Wildlife Monitoring Applications.

Published:

24-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2979-2990

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite