The Innovative Speaker Verification Service

Nov. 24, 2011 - PRLog -- One of the challenges in the field of speech technologies is to determine who corresponds to the given speech signal. Systems which solve this problem fall into two classes - verification and identification technologies. Verification - voice-based access control algorithm performing user authentication by voice password, and identification - determining user’s identity from the specified, limited list of people. Nowadays, voice identification and verification systems are in demand around the world.

Person’s voice has individual characteristics, it is unique. Therefore, recent development companies of biometric systems define identity of the speaker by voice. Increasingly, there are systems in which voice performs protection function. Access rights to network, facilities, banking systems, cloakrooms, personal information, etc. are verified via voice. In practice, it turned out that password protection was not reliable. Text passwords and keys are inferior to voice-based identification and verification systems. The very voice is the universal key that can not be faked, broken or stolen.

Identity verification of user by voice presupposes compatibility of a predetermined speech sample with a newly given one. User pre-sets any phrase as a password, and then he utters his password. The system checks whether the uttered voice conforms to a pre-established standard. If a match is confirmed, then access to the system opens. Thus, the voice verification provides a greater degree of reliability in comparison with the password verification via keyboard. Another's voice can not be adulterated; while password falsification is available and realizable.

Personal identification by voice is the process of determining whom the voice belongs to from a limited list of people. “Having examined” sound record library, system will identify the closest voice to the sample.

Speaker verification is an innovative form of voice biometrics. The specialists from the Belarusian company Speech Technology have created their own algorithm of text-dependent speaker verification which is notable for its accuracy and reliability.

Now everyone willing to try and use the system can do this due to the launch of a unique on-line speaker verification service based on SpeteQ Verify technology http://verification.speteq.com/demo/. The speaker verification service is working in beta-mode now though it’s available twenty-four-hour. Anyone can check how it works – all you need is a microphone.

SpeteQ Verify technology can be integrated in any software or system due to its usability. Furthermore, it can ensure a high level of security due to the measuring of the biometric features of a person during identification, authorization, etc.

The service also includes an adaptive recording system which makes it possible to calibrate the recorder against the ambient noise. It works perfectly in the communication channels where voice compression is used.

Speech Technology specializes in the development of innovative and cutting-edge solutions for voice and speech processing. In the nearest future we are going to launch a number of services which will highlight the capabilities of speech technologies.

In the meantime, we are glad to ask everyone concerned to test our verification system! http://verification.speteq.com/demo/

# # #

Speech Technology Ltd. designs and develops science intensive software for recording, processing and analysis of speech data. Our products make it possible to recognize speech, change voice and speech rate, implement voice biometrics, analyze audio/video data and many others. Our solutions are warning systems, stenographing systems, audio editors, video surveillance systems and many others. Our team of developers is professionals in the range of implementation of cross-platform applications from mobile devices to server stations. Detailed information is on our web site: http://speetech.com/