In the contemporary digital landscape, there has been a marked increase in the demand for captivating and easily accessible video content. This paper outlines the creation and deployment of an AI-driven commentary bot that is capable of producing natural-sounding spoken commentary for a diverse array of video formats, such as sports highlights, instructional videos, and vlogs. The bot utilizes sophisticated machine learning methodologies to analyze video frames and generate customized scripts that deliver informative and engaging commentary. These scripts are subsequently transformed into audio using cutting-edge Text-to-Speech (TTS) technology. Our methodology commences with the extraction of video frames, which are encoded in base64 format to facilitate efficient processing. By employing Google’s Generative AI, specifically tailored to ensure the generation of safe and relevant content, the bot creates scripts by interpreting visual information from selected frames. The incorporation of gTTS enables these scripts to be converted into high-quality spoken commentary. This system not only boosts viewer engagement but also offers comprehensive audio descriptions of visual components, thereby significantly enhancing accessibility for individuals with visual impairments. The efficacy of the commentary bot is illustrated through a variety of examples and assessed based on user feedback and qualitative metrics. This research underscores the potential of AI-driven innovations in automating content creation and improving media accessibility, paving the way for future developments in this domain.