Natural Language (NLS) or Visual Search lets you look for assets in your HIVO portal using everyday language. Whether or not images are tagged, NLS interprets both your keywords and the visual content of assets—across 100+ supported languages—to deliver intuitive, precise results far beyond the capabilities of traditional metadata-only searches.
HIVO’s Similarity Search offers a visual, metadata-independent approach that spans your entire asset library—whether images are tagged or not. It instantly ranks either the full library or a selected subset, so you can immediately identify assets with the highest visual resemblance. Multi-target searches let you redefine queries or construct sophisticated searches for abstract themes—such as fun, climate or leadership—by using multiple example images.
With Search by Image, you can determine at a glance whether similar visuals already exist in your HIVO portal. Simply paste an external image URL to locate like assets and avoid purchasing duplicate—or overly similar—stock imagery.
HIVO’s AI-Powered Duplicate Finder employs advanced AI algorithms to detect and manage duplicate files—across images, documents and videos—either before upload or within the waiting room. This ensures complete control over asset duplication and eliminates redundant content.
Text-in-Image Search enables you to find images containing specific text directly from your HIVO portal’s search bar. Whether you need to locate campaign assets or retire those bearing outdated slogans, this feature delivers results with a single keystroke.
Save time by automatically tagging known faces with HIVO’s AI Face Recognition. Search quickly for assets featuring specific individuals—even within vast collections—by leveraging face-mapping technology to streamline identification.
HIVO automatically generates transcripts for all audio and video files in your DAM. Using AI-driven speech-to-text models that support more than 100 languages, it converts spoken content into searchable text, making multimedia assets fully discoverable.