Aryabhatta Infosys

the payroll people...

Voice Recognition Systems PDF Print E-mail

Principles of voice biometrics

Our voices are unique to each person (including twins), and cannot be exactly replicated.

How does voice biometrics work

Speech includes two components: a physiological component (the voice tract) and a behavioural component (the accent). It is almost impossible to imitate anyone's voice perfectly. Voice recognition systems can discriminate between two very similar voices, including twins.

The voiceprint generated upon enrolment is characterised by the vocal tract, which is a unique a physiological trait. A cold does not affect the vocal tract, so there will be no adverse affect on accuracy levels. Only extreme vocal conditions such as laryngitis will prevent the user from using the system.

During enrollment, the user is prompted to repeat a short passphrase or a sequence of numbers. Voice recognition can utilize various audio capture device (microphones, telephones and PC microphones). The performance of voice recognition systems may vary depending on the quality of the audio signal.

To prevents the risk of unauthorised access via tape recordings, the user is asked to repeat random phrases.

Benefits of voice biometric systems

  • Ability to use existing telephones
  • Can be automated, and coupled with speech recognition systems
  • Low perceived invasiveness

Weaknesses of voice biometric systems

  • High false non-matching rates

Applications of voice biometrics

Voice biometric systems are mostly used for telephony-based applications. Voice verification is used for government, healthcare, call centers, electronic commerce, financial services, customer authentication for service calls, and for house arrest and probation-related authentication.


You are here  : Home Products Biometric Devices Voice Recognition