<p>The recent developments in the field of artificial intelligence (AI) have resulted in the popularization of machine learning (ML) applications in all sectors. But, performance and interpretability of the model are usually in a trade-off. This has given birth to the emergence of Explainable AI (XAI), which aims at making AI systems more transparent and understandable. This study presents an overview of the recent XAI approaches that are used in different real-time applications. This survey classifies and evaluates the existing approaches based on their goals, process, benefits, and evaluation parameters. Rather than emphasizing model superiority, the analysis focuses on understanding trade-offs between accuracy, interpretability, and application context. Results are presented comparatively to reflect realistic strengths and constraints. A comparative analysis helps to point out the peculiarities of each model. We also postulate on performance analysis and future research directions to be undertaken to improve the design and applicability of XAI techniques.</p>