Microsoft Foundry IQ Now Supports Audio and Video as Indexed Knowledge Sources

Microsoft has integrated Azure Content Understanding into Foundry IQ standard mode, enabling organizations to extract and index content from audio, video, documents, and images for use in AI agent workflows. The integration, announced at Build 2026 and available from the 2026-05-01-preview API release, addresses a longstanding gap where recorded meetings and calls remained unsearchable despite containing valuable institutional knowledge. Developers have two ingestion paths: a native pipeline mode activated by a single property flag on file-based sources, and an explicit approach where Content Understanding is run independently before indexing the output as text. Microsoft's current documentation emphasizes document understanding over audio and video, so developers are advised to verify media-type support in their specific region and API version before building workflows around it. The explicit path is recommended as a reliable fallback that also makes adopting the native path straightforward once full media support is confirmed.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in