YuMe uses humans and machines for video ads
YuMe Networks is new start-up using a machine and human approach to help advertisers decide when to run ads by videos.
YuMe reminds us of Yahoo’s old approach to search engine technology: Helping along machines with a little bit of human intelligence. Let’s face it, machines still aren’t good enough to classify video accurately enough for cautious advertisers. No one wants to repeat the blunders committed by Google a few years back, when its machines inserted ads for suitcases besides gruesome articles about suitcases filled with body parts.
YuMe searches for information supplied by video creators and other data sources and classifies video clips into categories, such as “automotive.” YuMe uses speech-recognition technology to confirm that a video’s audio track is related. But then it also employs people in India to make sure that videos are about what they say they are. (See WSJ story; sub required) The company launched today.
The company has received more than $7 million in funding from Khosla Ventures, Accel Partners, and BV Capital and others.
Next Story: EMax, another way to make you rich on the Web, or just a prank?
Previous Story: Vector leads purchase of SafeNet for $635M
About the Author, Matt Marshall
Matt launched VentureBeat in September of 2006, with the realization that no one else was covering the entrepreneurial and tech innovation scene with the velocity or depth that he was. Prior to founding VentureBeat, he covered venture capital for the San Jose Mercury News from 2001 to 2006. In 2002, Matt was awarded "Journalist of the Year" by the Northern California Society of Professional Journalists. Prior to working at the Merc, he was a correspondent for the Wall Street Journal in Bonn, Germany from 1995 to 1998, and a writer for the Washington Post in 1994. Matt holds a PhD in Government and an MA in German and European Studies from Georgetown University. In addition to VentureBeat, Matt is also the Executive Producer of DEMO, the leading launchpad event for emerging technologies.
