 |
 |
 |
 |
 |
 |
 |
 |
|
Voice VerificationVoice verification is conceptually similar to fingerprinting. It is common knowledge that each person’s fingerprints have unique characteristics that can be used to distinguish one person from another. It has also been proven that each person can be identified by the unique features of his or her vocal characteristics and speaking patterns. Biometric voice verification is the process of comparing a voice sample with a stored, digital voice model, or voiceprint, for the purpose of verifying identity. A voiceprint is a digital representation of some of the unique characteristics of a caller’s voice, including physiological characteristics of the nasal passages and vocal chords, as well as the frequency, cadence and duration of the vocal pattern. A voiceprint is not a recording or sound file – it cannot be played back into a voice biometrics system by an imposter. How it works
Enrolment - To start the process, new users call the system, and provide a voice sample as prompted by Spike Server. This voice sample is then converted into a digital template or voiceprint, which is stored in the Spike Server database. Verification - Spike Server’s voice verification is text dependent, meaning the caller must state a specific word, phrase or digit sequence in order to be verified. This is ideal for combining knowledge verification, such as an account number or password, with voice verification. In a typical verification process, the previously-enrolled caller speaks a unique identifier such as a phone number. Through the use of automated speech recognition (ASR), Spike Server recognizes the identifier, and then checks the database to see if the caller has enrolled. Spike Server then scores the voice sample to determine if it is a match. The use of ASR is optional. The caller could provide the unique identifier
to a live agent or could use a touch-tone keypad. |
