So what is a Cognitive Media Platform and why is it important/disruptive?
The answer to this question comes from Newport Beach, California-based Veritone which has quietly been working under the radar for over two years to change the way the processing of unstructured audio and video streams can be easily turned into near real-time actionable business intelligence.
As noted in the unveiling of the Veritone platform, this is the result of brothers Chad and Ryan Steelberg, two entrepreneurs recognized for their innovations in the programmatic and efficient delivery of online/offline advertising. In working with audio and video data they realized that equipping machines with the power to organize, understand and respond to such content regardless of where it came from was the next logical progression in maximizing its value as a business tool.
What is a Cognitive Media Platform?
Without going into all of the technical details, a good way to describe a Cognitive Media Platform (CMP) is as a cloud-based ecosystem that ingests, analyzes and activates private and public media. As the company says, its goal is to transform data to understanding. This is achieved through the intelligent interrogation of unstructured media streams—television, YouTube (News - Alert), radio, podcasts and daily spoken-word communication such as meetings, speeches, presentations, voicemail and telephone calls— through a set of cognitive algorithms acting in concert with one another.
The Veritone platform figuratively and literally functions as the conductor controlling each cognitive engine such as transcription, facial recognition, speaker and object identification, tonal analysis and audience measurement.
What this looks like at a high level is shown below.
The functionality rests on a series of developer module/APIs:
Cognitive Processing : Transcribe, transcode, arbitration, CRM data from client, speaker ID, finger printing, Web analytics from client, audience/research, geo location, sentiment extraction and trends.
Mine & Search: Index, search and watch lists.
Actions: Reporting, alerts, playback, collections, Web hooks, ad insertion, widgets.
In answer to the question as to why this is important/disruptive, Chad Steelberg, CEO, Veritone, said: “Humans have finite capacity to store and process information. We have become the limiting factor. For the past 17-plus years, companies have solved the storage problem by organizing and making accessible the world’s information. The time is now for a platform that can understand media, thus impacting human interaction more than any other technology to date. Veritone was founded on a single idea: to transform the world’s media into actionable intelligence.”
In discussing the platform with me, Steelberg explained a few interesting highlights of it. “Everything is tied to a timeline and we can act as a conductor phrase-by-phrase and tone-by-tone for 45 transcription engines and do so at an unprecedented level of accuracy and at very low cost.”
Steelberg added that: “There is no distinction between human and machine. Google (News - Alert) created the world’s biggest filing cabinet and we are taking it a step further by providing a way to extract understanding in a way that has been unimaginable. We are the next generation of analysts.” He noted that, for example, the time for a TV station (a good way to think about the term programmatic) to find the precise time and nature of all clips related to a given subject is reduced exponentially and at a very low cost.
A few other things that are noteworthy about the announcement are that Veritone, through its open architecture is enabling third-party cognitive engine developers to participate in its CMP. Its Restful APIs allow developers to embed the platform-derived intelligence into their applications and to develop and monetize their own cognitive engines.
From a monetization perspective as a cognitive capability the CMP is a learning platform that becomes smarter, faster and more comprehensive over time. And, as part of the coming out party, Veritone has a series of use cases that are worth a view.
In an age when not just content but increasingly context will be the engine for next generation value creation, call it disruptive or a jump in the evolution of extracting data to gain deeper insights, cognitive computing is a major trend and capabilities such as the Veritone CMP are examples, like IBM’s (News - Alert) Watson, of the power of having searchable and extractable access to a wealth of valuable information which has historically been too expensive, time-consuming or even impossible to obtain.
Edited by Dominick Sorrentino