VOICE AUTHENTICATION

Introduction:

More and more people are on the move and need to be able to securely access information over the phone, the Internet or company Intranet.

Traditional access methods such as user ID’s and Password protocols can easily be compromised. Using Secure Access, once an employee has been verified, they will automatically gain access to confidential information such as voice mail, and e-mail or anywhere where the person accessing a secure network needs to be verified. Traditional methods such as PIN's and passwords are easily compromised. The potential cost of a breach in a company's Intranet security could be catastrophic from both a monetary and reputation perspective. Furthermore excessive numbers of passwords and be easily forgotten and slow to reset. Speaker Verification is based on the simple principle that a caller's own voice is the easiest, individual bound and most convenient way to identify remote service users. Because each human being has a unique voiceprint, as unique as a fingerprint, it can be used for verification purposes with the highest degree of convenience.

Speaker Verification enables call centers, telecom companies, and any company that relies on remote services to verify the identity of their users quickly and efficiently: a voice password, like the customer's own name or telephone number, is matched against it's reference templates, which was stored within the service provider database at the enrollment process.

Enabling Technology:

When the customer calls, he or she is given a prompt or asked for a password, which can be as simple as his/her name. The vocal sample is switched to the Voice verification System for analysis and matched with the prerecorded voiceprint. Verification is immediately returned to the agent or an Interactive Vocal Response in case of operator free services. The customer doesn't need to memorize anything because the own voice, rather than a specific number, is being verified.

In more details: voice Saista Sarl works by digitizing a profile of a person's speech to produce a stored model voice print, rather like a template, which is referred to each time that person attempts to access secure data or premium services. The position and movement of the glottal tissues, lips, jaw and tongue correspond with speech movements in the vocal tract. Saista Sarl technology reduces each spoken word into segments: sub-word like syllables, phonemes, triphones or similar units of sound, composed of several dominant frequencies called formants, which remain relatively constant over that segment. Each segment has three or four dominant tones that can be captured in digital form and plotted on a table or spectrum. This table of tones yields the speaker's unique voice print. The voice print is stored as a table of numbers, where the presence of each dominant frequency in each segment is expressed as a binary entry. Since all table entries are either 1 or 0, each column can be read bottom to top as a long binary code. When a person speaks his or her voice password, the code word or words are extracted and compared to the stored model for that person. When a user attempts to gain access to protected data, their voice password is compared to the previously stored voice model and all other voice prints stored in the database. Each speech sound in the user's voice password is queried in an anti-speaker database. Since some characteristics of a person's voice are the same as another's, the system authenticates the user by comparing the user's common features with those in the anti-speaker database and eliminating those common elements from the sample to be authenticated. When all features matching others are removed, the system is left with only the unique features of the user's voice. These unique features, compared with the enrolled voice password, are the characteristics which determine successful authentication.