A claims department needs structured fields pulled from dashcam videos, scanned forms, call recordings, and damage photos, all mapped to one shared business schema. Which service fits best?
Show answer and explanation
Answer: D
Azure Content Understanding is designed to extract structured results from multimodal content, including documents, images, audio, and video, using a schema you define. Because the claims team wants one business schema across varied inputs, a multimodal extraction service is a better fit than assembling separate single-modality services and reconciling their outputs. The result is a consistent structure regardless of the original file type.
Why the other options are wrong
- A. Entity recognition works only on text. It cannot read video, scanned forms, or photographs, and would miss most of the claims evidence.
- B. Speech transcription yields text from audio only. It leaves forms, videos, and photos unprocessed and does not map results to a shared schema.
- C. Image Analysis tags and describes pictures. It does not handle audio, video, or scanned forms, and it does not produce a unified business schema.