Plan

1.Write a script to download material from BBC
2.Sort out options to covert Audio into Text Use SpeechRecognizer other best alternative
3.Clean transcribed text make sure coversion is near perfect it should be as it is narrated by one of the best narrators
4.NLP : Use Spacy or NLTK to extract entities/subjects and Entity-Event/topic-Relationship
5.Make it a searchable responsitory of text