Speech recognition is the lightest , fastest, and most human of all interface technology. Human beings, when utilizing a speech recognition interface, often become so quickly familiar with it that they attribute human characteristic to the program. They also are uniquely intolerant of speech recognition error.

We have developed a technology which recognizes heavily accent, and nonstandard speech. This technology is robust, and although currently under development we anticipate that it will valuable to many broad sectors of industry. This work arose from our systems being required to operate in the high noise, low speaker intelligibility environments of airport and construction, and is based on a series of algorithms and mathematical techniques that have been successfully applied in other areas of industry to categorize, sort and classify nonstandard speech, heavily accented speakers, and distorted or dropped speech elements. This will be a robust and usable speech technology for persons with dysarthria with applications to the broad market, including speech engine enhancement and biometric score and speaker verification enhancement.

We are currently moving into Phase II of our research and anticipate that our product will be released in the spring of 2012. Please feel free to contact us , using our online form for information, and update. And as always, we welcome any feedback, comments, suggestions or ideas. Please feel free to contact us.