These AI Headphones Can Warn Wearers Of The Passing Traffic

Aadhya Khatri - Jan 02, 2020


These AI Headphones Can Warn Wearers Of The Passing Traffic

AI is used to interpret the sounds of passing traffic from as far as 60 meters away from the user, providing information on where exactly these vehicles are

Headphones with their noise-canceling ability are now produced by most of the best headphone brands. They can seal off wearers from the surrounding world by putting them in their own bubbles. While you are outside with your headphones or earbuds on, the outside world seems to disappear, but it is not actually the case. You can still get yourself in an accident if you cross the road without paying attention to the traffic as you are too immersed in the virtual world the headphones create.

We have had enough cases of pedestrians endanger themselves as a result of wearing their headphones while on the street, this is why experts are turning to AI to remind users of the danger.

These scientists are from Columbia University and what they created is PAWS (short for Pedestrian Audio Warning System), which can warn headphone wearers of the danger of passing traffic while they are trapped in their own virtual world.

headphones-vehicle-alert
PAWS (short for Pedestrian Audio Warning System) can warn headphone wearers of the danger of passing traffic while they are trapped in their own virtual world

The tech makes use of AI to interpret the sounds of passing traffic from as far as 60 meters away from the user, providing information on where exactly these vehicles are. The experts hope that they can help to raise awareness of how dangerous it is for pedestrians and protect them against the traffic.

The test headphone with the AI technology features a series of affordable microphones positioned in different places. The onboard custom integrated circuit will then extract the possible sound of vehicles and then transmits them to a smartphone app.

The AI in the app analyzes the input and then determines which sound is from a passing vehicle and which is not. To be able to accomplish this task, the machine learning system was trained with an extensive dataset of sounds from the environments as well as different kinds of vehicles.

The prototype is still wired but we may see a Bluetooth headphone comes along soon enough if this idea catches on.

Comments

Sort by Newest | Popular

Next Story