Voice recognition software such as Siri and Cortana are a boon to mankind.
From dictating an email to solving complex math problems, the software can do your bidding and perform any task as long as it's possible.
However, there are times when Siri or Cortana just doesn't understand what you need.
Even if you command them to play music or ask for directions, for instance, they won't give you the right result, especially if you're in a noisy place.
So what's the solution?
Instead of cleaning up the audio signal that captures the user's voice, an Israeli startup known as VocalZoom believes voice recognition applications could work a lot better with the help of a new sensor.
The sensor contains a small, low-powered laser that can measure even the slightest vibrations of your skin when you speak.
How The Sensor Works
VocalZoom Chief Executive Tal Bakish says the sensor is quite simple and operates in a way that is similar to acoustic microphones.
Typical acoustic microphones measure the movement of a membrane when pressure or sound waves in the air hit it.
But with the new sensor, the skin is the membrane. During speech, the sensor measures the movement of this membrane using a laser aimed at key areas of the face, such as the lips, mouth, throat, cheek and behind the ear.
Bakish says the membrane moves due to the sound waves inside the mouth, which are not affected by background noise. The sensor only hears the "voice" of the speaker.
Future Plans
The startup is still currently building the sensor which they say will be first integrated into helmets and headsets.
There, it will be used alongside speech-recognition technology that depend on microphones to reduce misunderstanding.
So far, VocalZoom has raised at least $12.5 million in venture funding. Developers expect the product to be first used in motorcycle helmets or headsets used by warehouse workers.
For instance, the sensor might be handy when asking directions while riding your Harley.
Additionally, Bakish believes the sensor could be added to cars by 2018 so users can give vocal commands while driving.
"If people can talk to their car in the same way they talk to their friends, it will become very easy to sell connected-car services and products," says Bakish.
A Chinese company called iFlytek, which specializes in speech recognition technologies, plans to have a prototype headset with the VocalZoom sensor at the end of August this year.
VocalZoom also has several joint agreements with other automotive companies, although Bakish won't name them on record.
Photo: Kārlis Dambrāns | Flickr
