The purpose of this survey study is to present a thorough and unbiased overview of several deep learning models for the conversion of Indian Sign Language (ISL) to text. We have examined a few research papers that have put forth some deep Learning and Machine Learning models, such as support vector machines (SVMs), recurrent neural networks (RNNs), and convolutional neural networks (CNNs). In addition, we have talked about a number of issues that must be resolved in order to construct ISL-to-text conversion models, including the difficulty of ISL grammar and the dearth of extensive publicly available ISL datasets. Finally, we have contrasted and compared how well these models perform on ISL and text conversion. We’ve observed that with some great accuracies on a few datasets, CNNs and ResNets have produced the best results yet, whereas models like InceptionV3 are not that plausible in this domain. Though we have come across several intriguing possibilities for more research, there is yet an opportunity for improvement.