To be an effective teacher, one must possess strong learning abilities. Developing lesson planning, pursuing learning objectives, and assessing post-lesson accomplishments all these depend on reflection and ongoing learning. As education is context-specific, the iterative process of preparing, reflecting, and improving is what makes teaching effective. The scarcity of long-term as well as intense professional development is a problem, and educational institutions have indicated that staff-led professional development has to be of a higher caliber. Considering the difficulties involved in conducting classroom observations, instructors barely get any constructive criticism--a critical component of improvement. The goal of this study is to provide teachers with thorough and insightful automated feedback, therefore eliminating a major barrier to their pedagogical efficacy. We propose an Artificially Intelligent Smart Teacher Feedback System (AISTFS) in this study. The proposed framework demonstrates competence in evaluating the delivery of pedagogy by providing teachers with performance ratings that are derived from detailed analysis. Notably, the proposed model shows an astounding 97% recall rate and a high precision score of 95%, indicating that the model is effective at picking up on variations in teaching. A comparison with state-of-the-art algorithms illustrates the prevalent performance of the proposed framework. Its remarkable real-time processing speed and detection accuracy mark it as a significant headway in the field of education technology.