Recent advances in AI has made it possible to generate images and videos from text prompts (for example, see https://github.com/yzhang2016/video-generation-survey/blob/main/video-generation.md for a list of video generation models). The aim of this project is to develop a pipeline that automatically generates a music video for songs. The video can consist of a sequence of images or videos segments generated from the lyrics, potentially with the help of other user inputs. Assuming we know the lyrics and their timing within the song, one potential idea is to first split the lyrics into different segments, then convert each lyric segment into a suitable text prompt, which can be used to generate a video clip that is consistent with the meaning of that segment. These generated video clips can then be merged into a single video. It is also possible to explore other approaches for the generation.