Speaker verification technology provides accurate and secure identification in an easy and cost-effective manner. 

With this technology, accuracy is especially important as it is used to identify speakers for security and other critical reasons.

 

Persay Ltd., provider of biometric speaker verification technology, is providing optimal accuracy according to a comprehensive evaluation done by the University of Canberra, Australia who rated the accuracy of their technology as number one.

 

For the evaluation, Persay's VocalPassword system (their most recent version and an alpha version of their upcoming release) were tested thoroughly in comparison to the same testing done in 2005.  

 

The system was praised for its improvements and enhanced capabilities especially when it came to accuracy. "In some cases the Equal Error Rate was reduced almost 50 percent, compared to the system tested in 2005, exceeding 99 percent accuracy," a news release noted.

 

VocalPassword is a text-dependent speaker verification system; meaning in order to properly identify someone, their unique voiceprint, stored in the system, is compared against their spoken pass phrase. The system is language and accent independent, for enhanced accuracy and secure identification.  

 

Dr. Ran Gazit, CTO, Persay, commented in a statement to the press, "The results of this comprehensive set of tests have reassured our confidence in the system's performance, showing excellent verification accuracy under various conditions, including increased robustness to background noise and mobile network environments."

 

Gazit also commented that Persay will continue to work on improving text-dependent speaker verification technology as well as text-independent speaker verification technology, offered from their FreeSpeech product.

 

For more information on Persay Ltd. and their product offerings, visit www.persay.com

 

For more on the evaluations, please visit http://www.3sh.net/client-evaluations.html.