Boon Edam Blog | October 2020
When it comes to the security of our buildings, voice recognition has often been underutilised when compared to other security access control options, such as fingerprint biometrics and facial recognition.
Yet, as our approach to security continues to change, especially as we continue to find ourselves in an increasingly contactless security world, evidence shows this trend has begun to change drastically. This shift poses an important question - could voice authentication software prove to be the safest security option for those looking to secure key areas of their building?
While voice recognition has long had its ties to the security industry, its uses have been confined to traditionally non-physical environments. Much like fingerprint security in recent years, voice recognition has often been incorporated into the security of mobile phones, ensuring only those authorised will be able to unlock the device via their voice.
The idea of voice recognition to prove ownership is something the banking industry has invested heavily in over the years, with UK banks such as Lloyds and Halifax investing in the technology. This allows people to access their accounts and verify ownership by using their voice, providing a simple way to access their accounts without the need to remember password and pin codes.
When it comes to physical security, voice recognition was typically found in hi-tech home security installations, where homeowners can monitor their home through the installation of security devices that could be controlled through voice biometrics. Nowadays though, the barrier of entry to voice recognition security in the home has been greatly lowered, with the likes of Amazon Alexa and Google Home being able to control your smart home security system with ease.
These advancements have also been great news for building managers, with voice recognition security systems becoming a more attractive and affordable option for those wishing to incorporate their usage within their buildings to enhance security.
Despite the changes and advancements in voice signature technology over the years, its fundamental characteristics haven’t changed much.
In its simplest form, voice signature recognition software works by recording a sample of a person’s speech and turning it into a digital format. Once digitised, this sample can be stored in a database, which can then be used as a reference point for when a user requests access. If the user's voice matches one found in the database, then the user will be granted access.
Unlike fingerprint biometrics though, an individual’s voice can be affected by a variety of factors, which is why voice recognition software incorporates both a physiological and behavioural component to greatly reduce the potential margin of error during authorisation.
The physiological component focuses on voice biometrics digitally recreating an individual’s vocal tract. By focusing on the physiological component of an individual’s voice, such as the shape of their nose and mouth, a digital vocal tract can be created. As no two vocal tracts are alike, doing so ensures every voice collected and stored is truly unique
The behavioural component is also used to help create a unique print by focusing on the movement of the user’s mouth during speech. Variations of the movement can affect a person's manner of speech, such as their accent and tone. In monitoring this alongside the physiological component, voice recognition software ensures that no two voices collected are alike, granting only those who are authorised with access.
Not only is voice recognition security secure, but its ability to be truly contactless makes it arguably one of the safest security option when it comes to providing a contactless solution for both staff and visitors.
Its popularity as a biometric security option has been shown to outshine its touchless peers, with a Unisys survey ranking voice recognition as the most preferred security option, with 32% of consumers preferring it to fingerprints (27%), facial recognition (20%), hand geometry, such as gesture control (12%) and iris scans (10%).
While all touchless solutions are, by default, contact-free in nature, voice recognition is desirable, removing the need for close contact of the face or hands, reducing the risk of exposure even further. This, combined with its non-invasive nature and ease of use, highlights why voice recognition may not just be one of the safest contactless options - but a popular one too.
Continued investment in the voice recognition market also highlights this, with the market projected to reach a total of over 19 billion pounds by 2025, up from 3.9 billion when the market was valued in 2016.
This number is set to increase and with the contactless world not looking to change anytime soon, we may soon see voice recognition security software continue to grow in popularity.