Predictive Search Unveiled by Expect Labs; SXSW Accelerator Highlights Emerging Technologies
In the future, search engines will be able to read your mind – or at least your past conversations. On Tuesday at SXSW in Austin, Texas, several emerging companies presented their innovations to the crowd and in front of a panel of judges as part of the Interactive Accelerator track at the conference.
Moderator Chris Shipley hosted the session to evaluate companies and their respective technologies, judging them on a 100-point scale according to originality, scalability, functionality, potential for success, as well as the caliber of the teams charged with the challenge of turning a million dollar idea into a multi-million dollar company.
Tim Tuttle, CEO and founder of Expect Labs based in San Francisco unveiled his technology platform that combines predictive algorithms, voice recognition and search to present people with relevant information online in real-time, based on current and past conversations. His company is banking on the fact that people will no longer need to enter a question or query into search engines in order to be presented with relevant information.
In a project funded by Google Ventures, Intel and Greylock Partners, Expect Labs has created a technology platform that continuously analyzes signals from all aspects of your life including: voice, social data streams, location, generates a real-time model for just for you, then proactively searches across the web for relevant content. The company’s Mind Meld app listens to your conversations and then proactively finds information that is relevant.
You can share information with your friends in real-time and instantly, organized and archived to the cloud so you can find it later.
For the past 2 years, Expect Labs has been developing a technology platform to serve as the foundation for new types of intelligent applications. Tuttle calls it its “Anticipatory Computing Engine.”
As part of the judging panel, CEO of Magnify.net Steve Rosenbaum brought up the potential creepy factor. He said, “Google wants signals as an investor, and we see that this platform may be providing just too much information. It’s cool but adds to digital overload program.” Rosenbaum went on to comment that users are already bombarded with information and to have more content pushed to them during a conference call might just be distracting. People might want to have more control over what they search for and see.
In response to this comment, Tuttle mentioned, “A big part of our technology is if you can make a highly accurate understanding of context and language, and this is only possible if you analyze the entire information space that is relevant to that application. It is possible over a long period of time to understand context. These systems will become highly accurate.”
Another adjudicator Scott Case of Startup America added, “This is a great solution for a problem that you have not figured out yet. Last thing we need is more distraction with a stream. Talk to me about why this is a robust innovation.”
Tuttle responded that this technology is about driving voice and touchscreen interaction with people. Their platform has solved the voice recognition problem more than before because its predictive algorithms improve the accuracy over time by also looking at content sources over an extended period of time.
Expect Labs has quickly become ubiquitous— within the past six months the startup debuted Mind Meld at Disrupt SF 2012, scored itself $2.4 million in funding from some major names, and inked a deal with the voice recognition company Nuance. Recently Expect Labs has tapped into big data company Factual Inc. and its huge database of location data in a bid to make its computing engine even smarter.
No comments yet
Leave a Comment
- Ad Tech’s Next Steps: 7 Intriguing Predictions for 2017
- Horizontal or Vertical? How branding value is amplified or compromised using different mobile video ad formats
- Performance Measurement: How you doin’?
- 5 Ways Publishers Can Use Location Data
- Audience Targeting: How Accurate…and Useful…Is Your Data?