Expect Labs Taps Nuance

Expect Labs, a San Francisco–based start-up, announced plans to integrate Nuance Communications' voice technology to expand the voice recognition capabilities of its Anticipatory Computing Engine.

The Expect Labs Anticipatory Computing Engine is a technology platform designed to analyze and understand conversations in real time and proactively find related information. It is the underlying platform that powers MindMeld, the intelligent voice and video calling app for the iPad. Leveraging the Dragon Mobile SDK as part of Nuance's NDEV mobile developer program, Expect Labs will be able to extend the voice capabilities of MindMeld across a variety of mobile device platforms.

By continuously analyzing signals such as location, voice, and online activity, the Expect Labs platform creates a real-time context model, which can be used to anticipate and filter any information that a user may need. As an initial demonstration of the capabilities of this platform, Expect Labs has built MindMeld, the first mobile voice and video calling app that can understand conversations in real-time.

"With the growing adoption of smartphones, the past year has witnessed the emergence of a first generation of intelligent mobile applications," Expect Labs said in a statement. "With the launch of the Expect Labs Anticipatory Computing Engine, which deploys Nuance's voice capabilities, application intelligence can now be extended well beyond answering questions and simple voice search. Since it is designed to pay attention continuously as you use your device, Expect Labs' technology can bring context-aware intelligence to a wide range of applications related to real-time communication and collaboration."

Expect Labs has filed several patents covering its core technology and has pioneered advances in three key areas:

  • Real-Time, Multiparty Conversation Analysis: The Expect Labs technology platform is capable of concurrently analyzing audio signals for each participant in a group conversation in order to understand the key topics and concepts of the conversation in real time.
  • Continuous Context Modeling: Based on this understanding of the conversation, the Expect Labs technology constructs a continuous model that captures the overall meaning of any conversation as it changes from second to second.
  • Proactive Information Discovery: Rather than wait for a user to make a request, the Expect Labs platform leverages its real-time context model to proactively analyze thousands of pieces of content and data sources that may potentially contain relevant information. With this approach, Expect Labs can identify highly relevant pieces of information before a user needs to make an explicit request.

"We launched NDEV Mobile Platform to inspire mobile developers like Expect Labs to create breakthrough applications with our industry-leading speech recognition platform," said Matt Revis, vice president and general manager, Dragon Devices, Nuance Mobile, in a statement.

"MindMeld is a great example of the types of innovative applications that we're seeing across the NDEV program."

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

Speech Developer Programs Have Hit Prime Time

From the novice to the experienced, it's boom time for developers.