Throughout the years, the Internet and social networking sites have experienced a significant expansion in multimedia, notably video material. The amount of video data created has expanded at an exponential pace over time. Yet, many people are unable to read the entire information owing to time limits. The goal of video summarization is to create video summaries that incorporate the most significant and relevant material from a video stream in a concise manner. The summarising technique can remove a significant quantity of information while maintaining the most important information in the article. Our model has received the video, and the speech will be retrieved. The model then translates audio to text by chunking an audio file, which is handy for processing big files. The voice is then automatically transformed into text using automated speech recognition, which is a quick and simple approach to work with audio files. It transforms audio clips to text. Lastly, the BERT algorithm is applied to text summarization. Transformer, an attention mechanism that learns the contextual relationships between words in a summarised text, is used in the summarising process.