Overview
This project involves analyzing and classifying noisy audio signals. The analysis focuses on the heartbeat dataset, which contains normal and abnormal heartbeats recordings. The goal is to explore various signal-processing techniques to extract meaningful insights and determine suitable features for classification.
Dataset Information
The dataset consists of two folders:
- Train: Contains labelled samples of normal and abnormal heartbeats.
- Test: Contains unseen samples for validation.
Sampling Rate
Heartbeats dataset: 44.1 kHz
Signal Processing Techniques Used
The following signal-processing methods were applied to analyze the dataset:
- Fast Fourier Transform (FFT)
- Short-Time Fourier Transform (STFT)
- Continuous Wavelet Transform (CWT)