Hi there,
The requirements look quite clear and straightforward to implement. A new video will be created from the records in the database. Initially, the text will be converted to speech via one of the text 2 speech services, e.g google text to speech, amazon polly. That'll constitute the audio part of the video but what about the visual part? Will that be black screen or maybe rotating images of some sort?
Once the video is ready, other fields like title, description, maybe tags will be fetched from database and used while uploading it via Youtube API.
You want this to be a windows desktop application if I'm not mistaken. Though there's no question about how the data will be fetched from the database I'm not sure how are you planning to fill the database. Will that be a remote database maybe?
Anyway, I am planning to use c# to implement it and I expect it to be ready in a week at most, thanks.