Automated extraction of technical and contextual metadata including speech-to-text, on-screen character recognition, and image recognition.