Not only do many people walk around with their eyes glued to their smartphones, but a lot of them also wear headphones, making them oblivious to the sound of oncoming cars. With that in mind, scientists are developing headphones that warn their wearers when vehicles are approaching.
First of all, there are already microphone-equipped devices that silence third-party headphones whenever sufficiently-loud noises are detected. On a busy city street, however, these would constantly be getting triggered, even when the user was in no danger. Additionally, by the time the person realized that their music had been muted, it might be too late for them to avoid being struck by a car.
That's where the new headphones are intended to come in. They're being designed by scientists from Columbia University's Data Science Institute, working with colleagues from the University of North Carolina, and New York's Barnard College.
The current prototype contains multiple miniature microphones which detect sound in all directions around the user, transmitting the audio data to a machine learning-based app on their smartphone. That app in turn will ultimately be able to compare all of the current sounds to those that it's been "trained" on, recognizing any that are associated with fast-approaching vehicles. When such sounds are detected, it will alert the user with a distinctive audio alarm that's specifically designed to get their attention in a chaotic environment.
The whole system reportedly uses very little battery power, and should be inexpensive to manufacture. It is now being tested both in the lab, and on the streets of New York City.
"We hope that once refined, the technology will be commercialized and mass-produced in a way that will help cities reduce pedestrian fatalities," says Columbia's Asst. Prof. Fred Jiang.
Source: Data Science Institute at Columbia University