Advancements in computer vision and deep learning have brought about a revolution in object detection applications in recent years. Identifying and locating objects in images or video frames is a critical task in computer vision, known as object detection. This paper introduces a novel method for detecting live objects in real-time by integrating the (YOLO) algorithm with Convolutional Neural Networks (CNN) and audio analysis. Combining visual and auditory cues improves the efficiency and robustness of object detection systems, making them more effective in real-world scenarios where visual information alone may not be enough. The YOLO algorithm, famous for its real-time object detection capabilities, serves as the primary visual detection component. It is complemented by a CNN architecture, which further enhances object recognition, improves accuracy, and reduces false positives. Additionally, audio data is incorporated into the system to provide supplementary information for detecting and classifying objects. This multimodal system is effective in scenarios where visual conditions may be challenging, such as low-light environments or occluded objects. By combining visual and audio modalities, the system overcomes limitations inherent in single-modal approaches, ensuring a more comprehensive and reliable live object detection solution. The system's superior performance in terms of accuracy, speed, and robustness is demonstrated by experimental results on benchmark datasets. A confidence score of 0.7 indicates a high level of confidence in model recognition and indicates that there is strong confidence in the accuracy of detected objects.