- Arvato Systems VPMS customers and Vidispine subscribers will be able to exploit the potential of advanced media analytics, AI and Machine Learning (ML) services
- Vidinet users will execute multiple media analyzers and pre-trained AI models through a single interface
With minimal configuration, Vidinet users can execute multiple media analyzers and pre-trained AI models through a single interface that provides unified, aggregated and normalized results. The flexibility of this unified data model - the Analyzed Data Unit (ADU) - differentiates the cognitive functionality in Vidinet from simple AI aggregators.
Vidinet also enables users to cascade and combine different services to gain further insight into their media. For example, a common need might be to create metadata relating to the context and tone of the content; this can be achieved using use a ‘speech to text’ service to create a transcript of a clip, which could then be passed through a natural language processing algorithm for tone and context. The result could also be combined with object recognition analysis to increase the confidence level of that data.
This information can improve accuracy and relevance in search applications for journalistic and archive workflows. It can also be used to enhance monetization processes and optimize advertising placement through elements such as context recognition.
Later this year, additional functionality will enable results of any analysis to be stored in a graph database, allowing for the real-time combination and extrapolation of metadata to provide additional insight into media - for example, retroactively assessing the genealogy of a media file.
Vidinet will provide a common training interface for supported services, enabling training to be easily integrated into, and become an integral part of, media workflows.
“Vidinet has already dramatically changed the way that media organizations build and implement media supply chains. This addition to Vidinet takes an existing concept of aggregating AI services and extends it in a way that makes the implementation of AI and ML in media workflows far simpler and more maintainable,” said Erik Ahlin, CEO at Vidispine.
Holger Noske, VP Broadcast Solutions at Arvato Systems, added, “Taking a holistic approach to the development of this technology and applying it to our joint portfolio also responds to the rapidly changing demands of media organizations needing to fulfill consumer and advertiser demands with greater efficiency and profitability.”
The new features were very well received at this year's NABShow in Las Vegas (April 8-11, 2019). They are now available to beta users with services from AWS, Microsoft and Valossa. More services from Google, Vocapia and TextRazor will follow later this year.
Since 2009, Vidispine has enabled the media industry to solve their most complex challenges when it comes to building data-driven, cloud-based media supply chain and media asset management solutions. With video becoming a core function or feature for industries outside the media industry, there is a growing need for flexible tools and platforms that can help organizations to adapt content, applications, and processes quickly. Vidispine has commoditized media asset management on a platform allowing companies in any industry to build customized media supply chains and media applications. Vidispine is an Arvato Systems company.