You are here:  Home  >  Data Education  >  Current Article

Speech And Smarts Take To TV Search

By   /  September 25, 2013  /  No Comments

rsz_vev1The story below features an interview with Sam Vasisht, CMO of Veveo, who is speaking next week at the Semantic Technology And Business Conference in NYC. You can save $200 when you register for the event before October 2.

A recent focus group report from Veveo, whose semantic technology powers conversational interfaces that enable search on connected devices from TVs to tablets and smart phones to set-top boxes, reveals that three out of four participants are dissatisfied with their existing pay-TV content discovery experience. Reasons include that they are unable to search using keywords; they don’t know how to spell what they were looking for or remember the name of the show they want when entering search terms; it takes too long to scroll through the electronic programming guide; and they don’t see any recommendations that seemed relevant to them.

In an online survey the vendor conducted, it also found that, when users were asked if they would like to use voice or if they’d use if they had it to find content, more than 60 percent said yes, according to CMO Sam Vasisht. “People think there has to be a better way,” he says. “The level of interest and the sense of urgency that companies have about making voice-enabled feature a part of TV is becoming very strong. But just voice commands won’t get you there. You need something above and beyond.”

Veveo has at its roots a semantic Knowledge Graph for aiding with semantic interpretation of users’ spoken queries about what they’re hoping to watch. It provides the natural language processing technology and the ability to learn users’ behavior to better predict what the user wants out of the search (see our story for more detail about how it works here.) At the upcoming Semantic Technology and Business Conference in NYC, Vasisht in this session will explain why a conversational interface needs a semantic underpinning. Think, Vasisht says, of how people think about movies: They may be in the mood for an action movie but they don’t necessarily remember the actual title of a specific film, just that it had to do with a terrorist plot and maybe starred someone whose name would be very troublesome to have to type in, like Schwarzenegger.

semtechnyclogoIn that light, you can see why “semantic technology makes a huge difference in TV viewing,” he says. Veveo’s Knowledge Graph continues to grow, adding new movies each time they are released or new actors each time another one comes to the fore, for instance. It’s also building out its Knowledge Graph to support multiple languages. “It’s not just doing NLP for the different languages but also the content and associations for those,” he says. Hollywood movies, for instance, often have different  has different titles in other languages that must be mapped, and accommodations also need to be made for more obscure foreign films or local talent, like anchor staff on German news shows.

Vasisht says that Veveo currently is in discussion with large companies, including consumer electronics companies and smart TV manufacturers that supply to global markets, to define proofs-of-concept, and by the time of the Consumer Electronics Show next year, he thinks there could be some announcements of products that exploit its conversational interface. At the same time, Veveo is looking at how its technology can be leveraged in the enterprise world.

“The challenge there is more acute than even in media,” he says, pointing out things such as the presence of huge and siloed ERP systems through which users often must tediously conduct multi-factor searches. “What happens if the information is across multiple databases? The Graph lets you semantically search and pull that together, so it’s an overlay on top of distributed databases,” he says.

“I think the Knowledge Graph itself adds a lot of value, but going from that to a conversational interface is a logical next step,” Vashist says. A request to search HR databases for a Java developer that speaks Cantonese and lives in Hong Kong is helped by semantic tech that understands that Cantonese is a language and Hong Kong is a location and that Java in this case means a programming language and not a place. But how much better is it not to have to type such a request in but just speak it into a mike? “Plus,” he notes, “with so many enterprises going mobile, that’s a much easier interface on a smartphone.”

To hear more on this topic, register for SemTech here.




About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Tech Primer: Serverless Computing Converging FaaS

Read More →