Sleep apnea is a serious sleep condition characterized by frequent pauses in breathing while sleeping, causing loud snoring and extreme tiredness during the day. The brain causes brief awakenings to restore normal breathing during sleep, resulting in interruptions that disrupt restful and healthy sleep. Unchecked sleep apnea can worsen heart diseases and increase blood pressure, thus, underscoring the importance of early identification. This study utilizes Electrocardiogram (ECG) signals to detect sleep apnea. A 1D Convolutional Neural Network (CNN) is introduced for the task in hand. The pre-processing consists of eliminating particular frequencies and modifying the signals for uniformity and guaranteeing precise analysis. Nine different CNN architectures have been constructed and examined with the goal of improving feature extraction and classification. The ninth CNN architecture has showed how well it can differentiate between cases of apnea and non-apnea by achieving a test accuracy of 90.16% and a loss of 0.24. All experiments were carried out using the Apnea-ECG Database provided by PhysioNet.