<p>An irregular pulse, or arrhythmia, can result in serious health problems like heart failure and stroke. It is essential for efficient treatment and management to identify arrhythmias as soon as possible using electrocardiogram (ECG) signals. Manual analysis, which can be laborious and error-prone, is a common component of traditional arrhythmia detection methods. Accurate and dependable automated techniques for diagnosing arrhythmias are therefore becoming increasingly necessary. The goal of this work is to create an effective model for ECG signal-based arrhythmia detection. The main goal is to create and assess a deep learning model that can accurately distinguish between normal and arrhythmic heartbeats. A sizable ECG dataset comprising samples was pre-processed and divided into training and testing sets in order to accomplish this. Because a stacked vector convolutional network (SVCN) can recognize patterns and spatial hierarchies in sequential data, it was used. The multiple convolutional layers, max-pooling, and dense layers that make up the model architecture enable it to extract complex features from the ECG signals. The efficacy of the model was assessed using the python tool (Anaconda (Jupyter Notebook)) performance metrics such as Accuracy, precision, recall, F1-score, confusion matrix, root mean square error (RMSE), mean square error (MSE) and a mean absolute error (MAE) with an overall accuracy of 99%, the results show high accuracy in arrhythmia detection. This work highlights how deep learning models can enhance the precision and efficacy of arrhythmia diagnosis.</p>