Everyone has a unique voiceprint: Loquendo Speaker Verification technology uses the voiceprint as a biometric to ascertain an individual’s identity. Voice authentication is an easy-to-use, versatile yet non-intrusive technology. It is highly accurate and doesn’t require specific equipment.
All you need is voice!

A world of advantages!
Loquendo Speaker Verification (Loquendo SV) is the ideal solution for security issues associated with remote access to over-the-phone services. It leaves you free to create easy-to-use, personalized services that are free from the frustration that is often associated with managing the many passwords required
in day-to-day activities. This technology is flexible, non-intrusive and secure – which means high user acceptance.
Developers can integrate Loquendo SV technology into their existing solutions without deploying any particular biometrics equipment. This means you can give your customers peace of mind while containing costs!
It is available in a wide variety of languages, for the broadest possible customer reach.

How it works
Loquendo SV involves two distinct operation phases:

• Registration or enrollment - the speaker’s voiceprint is captured and memorized. This step is carried out once only at the start of the first session and involves the user uttering a selection of keywords.
  
  

• Verification - the user’s identity claim is accepted or denied or Identification - the user's identity is automatically detected.
  The voice which is being verified or identified is matched with the previously acquired models.

Main Characteristics
Loquendo Speaker Verification technology leverages the power of Loquendo ASR.
Supports several operative modes:

• Text-dependent – verification is carried out on the basis of a specific text, which can be suggested by the system or chosen by the user;
  

• Text-prompted – the system provides a “random” text for the user to repeat. This mode rules out any possibility of fraudulent access by means of recordings, etc.
  

• Free Speech - the user is free to say anything.

• Loquendo SV leverages best-of-breed Loquendo ASR (Automatic Speech Recognition) acoustic models to guarantee semantic matching: the knowledge verification mechanism verifies both the characteristics of the user’s voice and the content of the vocal password. This combination considerably increases the system’s overall security levels.
• The field of applicable vocal passwords is specified along with the relevant recognition grammars, which means highly precise application customisation.

System monitoring and calibration tools cover the specific security requirements of each application. Fully flexible and customisable in terms of:

• Number of repetitions and amount of content in the training phase
• Number of repetitions in the verification phase.

Developers keep system requirements low and benefit from maximum integration freedom: The size of each Voiceprint file is limited to a few dozen Kbytes. Loquendo SV interfaces with any kind of DB architecture and technology, offering total integration freedom.