Microsoft Video Indexer is a cloud service that enables you to extract visual and speech metadata from your videos, which can be used to build enhanced search experiences in your existing apps. At Build developer conference earlier this year, Microsoft first announced the public preview of the Video Indexer service. Yesterday, Microsoft announced a significant feature update to the Video Indexer service. Read about the new features and improvements below.
- Speech-to-text – Video Indexer now supports Egyptian Arabic for speech-to-text.
- With the addition of Arabic, Video Indexer portal has been updated to support RTL (right to left) text in the transcript tab as well as the search page.
- Captions – You can now start playing captions automatically in the Video Indexer player by adding showCaptions=true to the iframe embed player URL.
- Download insights from portal – You can now download the JSON insights extracted by Video Indexer via the portal.
- Playback speeds – The Video Indexer player was updated to support different playback speeds.
- Inline transcript editing – The transcript panel has been updated to support inline editing.
- Editing in insights widget – The insights widget has been updated to support editing in embedded mode (based on an access token).
- Keywords editing – The insights widget has been updated to enable the addition and removal of keywords.
- Annotations – The insights widget now shows the annotations extracted by the Video Indexer engine.
You can check out the Video Indexer API reference page for API updates. Apart from this, Microsoft has also updated the Video Indexer backend services to make it more efficient. They have added more video AI algorithms to extract additional metadata, improved accuracy of video AI technologies, more intelligent correlation among the extracted metadata to provide human friendly insights, user interface enhancement to showcase the extracted metadata and insights and more.
Check out Microsoft Video Indexer here.