AI to clip videos based on transcript of video and pasted text search.

Trying to automate the process of taking a video script (adobe premier text AI does this) and allowing me to paste in a paragraph (say of a song) and the AI automatically finds those words (or skips them if not available) in the video script and cuts the video and pastes those sections in sequence.

The end goal is to simplify the editing process (takes days) of taking any video from Youtube or otherwise, and grabbing words from the video to form a song, so that we can then take that edited video into audio program to enhance and make anyone sing any song with their own words.

Example end goal:
https://youtu.be/74gx0v5ZJzw?si=r2CwFtDgw34__cOF

Trying to make this an easier process using AI. I'm using AI to get the video to text. That's easy. But to search text in the video and have it automatically split the video in sequence of the searched text (ie. song paragraph) to make it easier, doesn't exist.

Example doing it manually:
https://youtu.be/AyZvJIQaQcs?si=DxnVS72LCDnjtSkk