Defines in Content Understanding how input is processed and what fields come out, either through a prebuilt analyzer or through a field schema you write yourself.
Also called analyzer.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Content Understanding analyzer in context, with comparison tables and the common traps.
Terms in this definition
- Azure Content Understanding in Foundry Tools
Foundry Tool whose generative AI analyzers turn documents, images, audio and video into structured fields by extracting, classifying or generating values.
- Prebuilt analyzer
Any Content Understanding analyzer supplied ready to use, for example
prebuilt-read,prebuilt-layout,prebuilt-invoice,prebuilt-audioSearchorprebuilt-callCenter. Additional fields can be added to customise one. - Field schema
The list of fields a Content Understanding analyzer extracts, each defined by name, type, description and method: extract, classify or generate.
Related terms
- Base analyzer
Prebuilt parent analyzer for a given modality (
prebuilt-document,prebuilt-image,prebuilt-audioorprebuilt-video), from which a custom analyzer inherits by settingbaseAnalyzerId. - enableSegment
Content Understanding analyzer option that divides a file into segments, assigns each one to a category from contentCategories and can hand each segment to a different analyzer.
- estimateFieldSourceAndConfidence
Optional Content Understanding setting, set for the whole analyzer or as estimateSourceAndConfidence on single fields, that reports each field's confidence from 0 to 1 and where it came from, so weak results can be sent for human review.
- prebuilt-audio
Base Content Understanding analyzer for audio. Before any fields are extracted, it produces a WebVTT transcript with speakers separated by diarization.
- prebuilt-audioSearch
Ready-made audio analyzer in Content Understanding whose output is a transcript separated by speaker together with a single-paragraph summary of the conversation.
- prebuilt-callCenter
Ready-made analyzer in Content Understanding for post-call analysis. It returns a role-labelled transcript of the call together with its summary, sentiment, topics and categories and the companies and people mentioned.
- prebuilt-documentFieldSchema
Utility analyzer in Content Understanding that looks at a document and suggests a field schema for it.
- prebuilt-documentSearch
Retrieval-oriented Content Understanding analyzer that uses generative models to turn a document into Markdown paragraphs, tables and figure descriptions, plus a summary.