There is a distinct feature of Hindu religious and philosophical texts that are written by several readers rather than by one individual. The Upanishads are considered one of the world's oldest philosophical texts that form the foundation of Hindu philosophy. Philosophy begins with the Bhagavad Gita, the core text of Hinduism. The text summarizes the critical philosophies of the Upanishads with a strong focus on ‘Karma’. These texts have been translated into many languages and there exist studies about themes and topics that are prominent; however, there are not many studies of topics modeling using language models which are powered by deep learning. In this study, we have analyzed the distinct and overlapping topics among the readers and compared selected texts of the Upanishads with the Bhagavad Gita. Using deep learning AI methods, we analysed English translations of the Bhagavad Gita. Bhagavat Gita – the song of the Lord is used as the input data. In this study, deep neural networks are used for training, where input data after data processing and analysis is passed to neural network, which then performs self-tuning to improve the model. In this work, a deep neural network is used for training where input data is passed to neural network after data analysis and processing which then performs auto-tuning that helps to make this model better.