What is Voice and Speech Recognition?
Voice and speech recognition are two separate biometric modalities that, because they are dependent on the human voice, see a considerable amount of synergy. Both are contactless, software based technologies, and as such are counted among the most convenient biometrics in regular use.
Voice recognition, also commonly referred to a voiceprint, is the identification and authentication arm of the vocal modalities. By measuring the sounds a user makes while speaking, voice recognition software can measure the unique biological factors that, combined, produce her voice. Voiceprints can be measured passively as a user speaks naturally in conversation, or actively, if she is made to speak a passphrase. The key is that true voice recognition measures the minutia of the voice, and is not wholly dependent on a spoken code or passphrase (though spoken passphrases can be used to speed up the process, they are not necessary).
Speech recognition, on the other hand, is a user interface technology. In today’s increasingly mobile and connected world, having hands free interface options is critical. Speech recognition technology, also called voice command, allows users to interact with and control technologies by speaking to them.
Combined, speech and voice recognition can be a powerful duo, able to authenticate and offer hands free interface simultaneously.
Where can I find voice and speech recognition?
Chances are you have speech recognition on your personal mobile device. Each of the main AI assistants—Apple’s Siri, Microsoft’s Cortana, and Android’s OK Google—can perform searches and basic tasks based on voice command. Going a step further, Siri was upgraded in 2015 to be able to recognize who is speaking.
Voice recognition for authentication is increasingly being found in call centers, measuring voiceprints of callers in order to prevent fraud while doing away with tedious security questions. It is also commonly paired with face recognition, its software based cousin, for mobile app authentication.
With the rise of the Internet of Things, voice and speech recognition are finding a niche. As experts search for solutions to secure the machine-to-machine network, voice and speech are commonly evoked as a convenient and secure pairing, ideal for devices on which design space is a premium. All voice and speech recognition needs is a microphone to listen.
How are voice and speech recognition making a difference?
Swisscom, one of Switzerland’s biggest telecoms, recently implemented voice recognition in its call centers. The technology passively authenticates customers who call in, and the company has emphasized the high level of security voice biometrics has brought to its phone operations.
Always on speech recognition is making today’s mobile devices easier to use than ever before. This technology is specifically gaining traction among Internet of Things focused chip-makers.
Voice recognition is strengthening other biometric login solutions. The USAA banking app, for example, uses facial recognition and voice recognition to provide easy and secure multi-factor biometric security, the voice component adding an extra level of liveness detection to the process.