Title: Real-Time Voice Cloning System Using Deep Learning, an emerging field in artificial intelligence, has witnessed significant advancements in recent years owing to the rapid progress of deep learning techniques. This survey paper delves into the realm of real-time voice cloning systems that employ deep learning methodologies. The ability to generate highly realistic and natural- sounding speech from limited audio samples has garnered attention due to its potential applications in entertainment, assistive technology, virtual assistants, and more. This survey provides an in-depth analysis of the key components and techniques employed in real-time voice cloning systems. We explore various neural network architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs) that have been utilized for voice cloning tasks. Additionally, we investigate the role of different training paradigms, including supervised, semi-supervised, and unsupervised learning, and discuss their implications on cloning accuracy and efficiency. Furthermore, the paper examines datasets used for training and evaluation, ranging from large-scale multilingual corpora to more specialized speech datasets. Framework has the capability to duplicate voices not encountered during training as well as generate speech from previously unseen text.