April 2026

Hybrid Content Support for Entity Extraction

Multimedia content, that contains both images and text, is now identified as Hybrid in the entity extraction process. Specto now runs both the image extraction (for faces and objects) and the document extraction (for full-text extraction and identifying people, places, and organizations) when processing hybrid resources.

Highlights

  • Greater detail for more complex titles that contain both images and text

  • A comprehensive extraction with more context

Collection-level Metadata Included for Title-level Enrichment

When processing the AI Metadata Enrichment, in addition to providing the digital content and the existing metadata record for the title, we are including also the metadata from the collection level to provide additional context for the AI. This upgrade enables our users to have more context and better metadata within both the title and the collection.

Highlights

  • More context for each title, and more context within the collection

  • Enhanced collection-level metadata

Collection Subsets Enable Exhibition on Multi-page Sites

For each exhibition which contains content from a specific Alma Specto collection, users can now display a subset of the collection on specific pages.

Use the following steps to create multi-page subset exhibitions:

  • Using the Dataset column in the CMS, give a name to the subset (e.g. "early years", "solo career", "grunge era", "later years").

  • Create separate pages within the exhibition that represent each of these datasets/subsets, to tell a fuller story about each, within the context of the bigger exhibition.

  • Customize your pages using the Wix site builder.

All these titles are cataloged in Alma Specto as one collection, but the exhibition can tell a fuller story on mutiple pages.

Highlights

  • Users can create a multi-page journey on their sites
  • Users can divide their collection into meaningful subsets to better present their exhibition

  • Enhanced collection-level metadata