How many times you wondered what it’s like if you had a “Knight Rider” car which you could talk to and just tell it what to do. This car identify you when you come, just as if it was your human assistant, and perform simple tasks at your request. How many times you had to operate a device while your hands are full, simple tasks, like answer a call  while holding a baby, or turn ON the light while carrying a heavy box.

Many applications and technologies try to address those needs today as well as Virtual assistant applications such as Siri(Apple) and Cortana(Microsoft). These applications fail to reach what we call the “usability threshold”. In the world of User interface (UI) and User experience (UX), if a device like a mouse, keyboard or touch screen has even a small error probability, the user gets frustrated and stop using it. By the way, it doesn’t matter if it’s a complex task to operate this device. Take for example keyboard. This is still the interface of choice for typing. Even in smartphones. Keyboard is an extremely complicated task to learn. But users are willing to go for it, because once you master this, it’s convenient and seamless to use. Gestures using touch screen, experts for Laughed at Apple when first lunched, saying this is not a natural way to communicate. Today we all use it.

In parallel to these un-natural-UI devices, new technologies like voice control and eye tracking evolved. These are considered much more natural interfaces for humans, but no one uses it. Why? Because it’s still suffer from robustness issues and errors. It’s easier to master, but the fact that we need to repeat words, even if it’s once every 10 words, or that we need to move our eyes again and again when the light conditions are bad for the camera, creates the feeling we have no control of the device. Or, as Steve Jobs said, the device should feel like an amplification of ourselves, an extension of our body, which we have full control of.

VocalZoom has developed a multi-functional Human-Machine-Communication (HMC) device, which enable accurate voice control and voice authentication. This super small and low cost sensor, has the ability to measure micro-meter vibrations on a human face and extract the voice out of these vibrations. The human face vibrate only because of the human voice and is not affected by background noise and the voice of other speakers. This way the sensor can listen to the speaker in front of the sensor. Using VocalZoom, Voice control software such as google voice, Siri and Cortana, can perform in a very accurate way and exceed the usability threshold.

Another important usage of the sensor is for voice authentication. The sensor not only improve the accuracy by having a cleaner signal, it also guarantee the voice comes from the speaker in front of the sensor. This is an advantage over conventional microphones which hear the voice from all directions and can detect other people speaking in the background. The market for these applications become larger as online payments become common as well as self-service kiosks such as ATMs.

The ability to measure very small movements and accurate distance, enable many additional applications such as replacing the need for buttons, vibrations measurement in machines to detect failures, proximity sensing and 3D imaging applications. Also battery operated devices require accurate indication for speech to turn the device ON just when required (Smart phones, BT headsets etc.).

VocalZoom today partner with lead customers in the consumer electronics and the automotive, and first products, which use the VocalZoom technology, expected in the market before end of 2016.


Written by: Tal Bakish, CEO at VocalZoom. Mail:





Related Post