TODO: Perform EDA on already scraped TED talks dataset Pick a model to generate embeddings from long-form text. Generate word embeddings from TED talks dataset. Setup Pinecone DB and store word embeddings