Search down to single frames. Fast. Smart.

You don’t just want to find the right file… you need to find the clips, sub-clips, frames and audio-tracks. You want to perform complex searches based on any kind of metadata model. You want to limit search results based on who is searching, and what is in your audio-visual repository. You want to be able to find your content wherever it is. You want Vidispine.

One of the most complex issues with rich media and video is to find exactly what you are looking for among large amounts of content. Your solution, service or application will never be better than the underlying platform, and how well your data is structured. Even worse, the platform can only be as good as its data model, metadata structure and the tools provided for indexing its content. We support all common ways of searching but adds additional capability in searching within rich media assets across multiple sites and extremely large data sets. The API for searching is straightforward but still powerful. Being able to find the right rich media asset is critical to increase the value of your content. Vidispine is designed to support any metadata model. There are no limitations in number of fields, hierarchies, supported languages or even what you index. Through the Vidispine APIs, developers can easily index anything from complete rich media items to parts of individual audio tracks or anything else in the rich media repository. The indexes are time code based, and have native support for audit trails, multi-language, inter and intra workflows, versioning, multi-tenancy and multi-site, including configurable conflict resolution.



Technology making it possible.

Metadata models and structures can be anything and the definition of the same is extremely flexible. Vidispine support for example hierarchies with infinite depth both on model and fields, weak references, groups, true multi-lingual suppor, tiered metadata definitions to support multiple applications in the same instance, taxonomies, keywords, patterns, ACLs on individual fields and much more. There is also support for specialized plug-ins like face recognition, voice-to-text, colour recognition etc in order to auto-populate the metadata. This is the foundation for high-performance search with great options of mass-populating your content repository with metadata. Finally, it is all about the data model. How data is represented is absolutely crucial to all of the above. Vidispine data model is designed to handle the most advanced cases, with multi-hierarchical, time-based, multi-language, versioned, metadata. But through the API, it is also possible to implement a system with “simple” key-value pairs.