July 17, 2014 -
SpeechPro announced it recently participated in the 2014 Speaker Recognition i-vector Machine Learning Challenge, which tested the performance of algorithms frequently used for voice biometrics.
Organized by the National Institute of Standards and Technology, the challenge analyzes the development of emerging techniques for using i-vectors for speaker detection as relating to conversational telephone speech.
Conducted in April, NIST recently released the results which were assessed by comparing more than 12 million pairs of voice tracks to confirm each speaker’s identity.
SpeechPro’s VoiceKey technology uses algorithms that were created by its parent company, Speech Technology Center. These algorithms joined more than 100 submissions from worldwide companies, universities and laboratories, including IBM and 3M.
“There’s been an increased interest in the use of voice biometrics and speaker recognition for law enforcement agencies as well as for commercial applications in banking or call center industries in recent years,” said SpeechPro president Alexey Khitrov. “Discovering the true identity of a voice offers an unprecedented opportunity to protect access to information as well as secure it.”
SpeechPro recently partnered with Avaya to make its voice biometric platform available on Avaya’s DevConnect Marketplace.