NASA, Xerox, Team On Voice Recognition In Space

With thousands of tasks on hand for astronauts, a voice recognitions system can be of tremendous help.

The space agency has a few bugs to work out of the system first. In testing, the voice recognition system had an error rate of ten percent, far too high for potential use in space. After collaborating with Xerox, they’ve lowered the error rate to only five percent.

Called Clarissa, the software has been designed to be a general purpose “procedure reader.” This will help with the lengthy complex checklists of steps astronauts must follow when performing experiments.

“Just try to analyze a water sample while scrolling through pages of a procedure manual displayed on a computer monitor while you and the computer float in microgravity,” challenges astronaut Michael Fincke, who recently completed a six-month stay on the International Space Station.

Scientists from NASA and Xerox demonstrated the technology at the Association for Computational Linguists’ 25th annual meeting at the University of Michigan, Ann Arbor.

“Clarissa is a fully voice-operated ‘virtual crew assistant,’ enabling astronauts to be more efficient with their hands and eyes and to give full attention to the task while they navigate through the procedure using spoken commands,” said Beth Ann Hockey, project lead on the team that developed Clarissa at NASA Ames.

Early versions of Clarissa tried to analyze all dialogue spoken by astronauts in testing. To say that could be dangerous to astronauts were the system widely integrated with the functions of a space station or craft would be understatement.

In 2004, Clarissa lead implementer Manny Rayner of NASA Ames contacted Xerox researcher Jean-Michel Renders of Xerox Research Centre Europe about a possible collaboration. They hoped that Xerox’s experience in machine learning, linguistics and text categorization would increase the system’s accuracy on the ‘open microphone’ task.

Xerox delivered. The Clarissa system can now more accurately analyze what is said, and recognizes words, sentences, and context of usage. A sophisticated machine-learning algorithm weighs the various pieces of positive and negative information. Clarissa can better recognize differences between system commands and innocuous conversations.

David Utter is a staff writer for Murdok covering technology and business. Email him here.