Pipeline 3: The Video Expert
The role of the OpenAI GPT-4o Video Expert is to analyze the comment made by the Commentator OpenAI LLM agent, point out the cognitive dissonances (things that don’t seem to fit together in the description), rewrite the comment, and provide a label. The workflow of the Video Expert, as illustrated in the following figure, also includes the code of the Metrics calculations and display section of Chapter 7, Building Scalable Knowledge-Graph-Based RAG with Wikipedia API and LlamaIndex.
The Commentator’s role was only to describe what it saw. The Video Expert is there to make sure it makes sense and also label the videos so they can be classified in the dataset for further use.
Figure 10.10: Workflow of the Video Expert for automated dynamics descriptions and labeling
- The Pinecone index will connect to the Pinecone index as described in the Pipeline 2. The Vector Store Administrator section of this chapter. This time,...
