Voice, face secure data in personal electronic devices
Security is becoming more and more of an issue as personal electronic devices grow increasingly sophisticated. Users do not want others to gain access to their private data in their smartphones, tablets, and other computers. In the same vein, people don't wish others to be able to access the systems in their automated homes.
Until recently, the main security solutions have required the user to enter a PIN code or a more complex password. Some products, like Android smartphones, allow the user to "draw" a shape with her finger. Although these methods may seem to be relatively unobtrusive, they become tiresome when one has to perform them multiple times a day. There's also the problem of forgetting whatever PIN, password, or pattern one has set up.
Another alternative is to add a fingerprint sensor, but this requires additional hardware that consumes valuable real estate on the device and that can cost between $5 and $10. Furthermore, these sensors don't always work as well as one might hope, sometimes requiring the user to try multiple times before giving up in frustration and falling back on the PIN/password option.
More recently, there has been a growing interest in biometric identification technologies, such as speech identification and facial recognition. Using just one of these techniques in isolation can result in an unacceptable level of "false negatives" (blocking access to the right person) and "false positives" (granting access to the wrong person). However, using both of these techniques together—referred to as "biometric fusion"—yields extremely high levels of accuracy.
Sensory Inc. specialises in developing speech and vision recognition, identification, and verification technologies. Traditional speech recognition requires the user to first press a button to alert the system that a voice command will be coming. Sensory pioneered TrulyHandsFree speech recognition technology, which is always-on and can identify spoken commands (even in noisy environments such as parties) while having a small memory footprint and consuming very little power.
Now Sensory has launched TrulySecure, which combines speech identification and facial recognition to give a state-of-the-art security solution. The TrulySecure technology suite is available to OEM product developers in the form of an SDK that unites Sensory's speaker verification technology with its newly developed, proprietary, face recognition engine.
Since devices like smartphones already include microphones and cameras, and since TrulySecure runs on the device's host/application processor, there are no additional hardware costs. As a result, Sensory says that implementing TrulySecure is only about a quarter the cost of a fingerprint-based solution (depending on volumes and commitments).
Of particular interest is the fact that TrulySecure is an on-device biometric identification system that does not rely on a connection to the cloud. Many users prefer this approach because they do not wish for their biometric data to be replicated and stored outside of their personal devices.
TrulySecure takes account of local environmental conditions. In noisy environments with good visibility, for example, TrulySecure will give more emphasis to facial identification. By comparison, in quiet environments with poor lighting, TrulySecure will give more emphasis to speech verification.
The level of security is application dependent. As can be seen in this video, an application like Facebook may be satisfied with only facial recognition. By comparison, an online banking application may demand a much higher level of confidence.
Last, but not least, users can define their own "voiceprint" or "passphrase," and both TrulyHandsFree and TrulySecure can support multiple users with different passphrases, where each user is granted access only to his or her settings, memory, and data on the shared device.
- Max Maxfield