Create a new media description job

Input media URL. Supports URIs of files uploaded to Cloudglue Files endpoint, public YouTube video URLs, public TikTok video URLs, public Loom share URLs, public HTTP URLs (including direct image URLs such as JPEG/PNG/WebP), and files which have been granted access to Cloudglue via data connectors.

Note that YouTube videos are currently limited to speech level understanding only. Public TikTok video URLs will be scraped and stored automatically — subject to charges. For files via our Data connectors, see our documentation on data connectors for setup information.

Whether to generate video-level and segment-level (moment-level) summaries and titles

Whether to generate speech transcript

Whether to generate visual scene description

Whether to generate scene text extraction

Whether to generate audio description

Include narrative chapters in the response when segmentation strategy is 'narrative'. Only affects cached completed results; newly created pending jobs do not include chapters.

Include shot boundaries in the response when segmentation strategy is 'shot-detector'. Only affects cached completed results; newly created pending jobs do not include shots.

When true, the video's search documents will be added to the account's default index, making them searchable via the Response API and Deep Search API with source: 'default'.

Optional free-form guidance for the description. Use it to name the terms, product names, acronyms, or people that matter for this media and how they are spelled, or to say what the description should pay attention to — e.g. "Spell product names as Cloudglue and Aviary; call out any pricing discussed". It steers emphasis and vocabulary across the visual, scene-text, speech, and summary passes; it cannot make the description report content that is not in the media, and it does not change the response shape. To constrain transcript speaker labels to a known cast, use participants instead — a prompt is context, not a constraint. Changing it produces a new describe job rather than reusing a cached one.

Include the file's user-defined metadata and source metadata on the returned file object.

Segmentation job id to use. If not provided will use default to uniform 20s segmentation. Cannot be provided together with segmentation_config.

The URL of the processed video

Duration of the video in seconds

URL of the file-level thumbnail for the video. Only present when include_thumbnails=true.

Unix timestamp in milliseconds when the job was created

Whether this describe job's search documents are included in the default index.

Error message if status is 'failed'

The ID of the segmentation job if the describe was run with a segmentation

Total number of chapters (only present when include_chapters=true and segmentation strategy is 'narrative')

Total number of shots (only present when include_shots=true and segmentation strategy is 'shot-detector')