This project involves bird sound classification using audio features extracted from audio files. The audio files were downloaded from a link, and feature extraction was done using librosa and other libraries. The main focus was on extracting MFCC (Mel-frequency cepstral coefficients) and training a Decision Tree model for classification.
The goal of this project is to classify bird species based on audio recordings. The workflow involves downloading audio files, extracting features using librosa, and then using these features to train a Decision Tree model. The features used for training include MFCCs, which are commonly used for audio processing and classification tasks.
-
Python
-
Librosa (for audio processing)
-
NumPy (for numerical operations)
-
Scikit-learn (for machine learning, especially the Decision Tree classifier)
-
Pandas (for data manipulation)
- https://www.kaggle.com/c/birdclef-2022