We are always being told that passwords we create are far too simple and should be complex. However, complexity means they become harder to remember. Consequently, people write them down or use an online password manager, which is not always easily accessible.
Two-factor authentication is a recommendation, ‘the something you know plus the something you have’. Many of these are free and can operate from your smartphone.
But, have you ever tried to move your authentication to a new phone? Even with the old phone it is hard, let alone trying to do it when your phone is dead or stolen! This is a ‘put you off ever doing it again’ procedure.
Biometrics were seen as the way ahead, but many devices just can’t support the options, even if they have some form of biometric capability. They have never been very reliable, having a high degree of false/positive allowance.
Then there is the challenge of the different ways we connect, not everything is accessed by a PC. There are dial in systems, mobile apps, gaming consoles and multiple OS platforms, which might not support your chosen authentication method.
The identifiers we do all carry are our brains and voices. With voice recognition widely adopted in most devices, our voice could be the key to secure access.
Until recently voice recognition has been either in the ‘command and control’ space, where you give short, clear instructions and the service goes off to find the answers which match as closely as possible, Alexa, Siri and others fall into this category, or, the ‘educate me’ platforms, where you spend hours teaching the system how you speak, adding any complex words you use. These systems frequently are industry-specific, but you also find it in Word and other such applications.
Both of these processes are not ideal for complex speech, as either they take too long to train or are greatly affected by the clarity of speech, so they struggle with heavy accents or background noise.
The new form of speech recognition, Advanced Speech Recognition (ASR), is looking to address both these areas and might be the breakthrough we have been looking for not only in user authentication but also many other areas where the voice is the power to command.
The challenge with any voice system is its ability to understand what has been said and act upon it. This means it must be accurate, and given a normal human only understand/listens to 97% of what is being said to them, ASR systems need to be getting to these levels. Unfortunately, there are very few which can.
If we consider the use of ASR as an authentication tool, there are no forgetting passwords or needing a smartphone, it works on any platform, so no more vendor lock-in! It is also simple for anyone to use, which can’t be said for some authentication methods, especially if you are elderly or visually impaired.
Now consider all the other areas of our daily lives and business where simply speaking to a device could transform our world. ASR is a game-changer not only in the data security world but also in every corner of our businesses. Integration into existing systems is available using open API’s (Application Programming Interface), but the new applications in education, finance and medical arenas, is truly exciting for this technology.
There is no holding back on the power of speech.
“
