可以避免行人被汽车撞到的智能耳机

Pedestrians could soon be safer thanks to smart headphones able to decipher hundreds of street sounds and signal imminent harm.

行人可能很快会更安全，因为智能耳机能够识别数百种街道声音和迫在眉睫的危险信号。

Scientists are developing the pioneering pods in order to slash the surging number of pedestrian deaths.

科学家们正在开发这种开创性的耳机，以减少不断增加的行人死亡人数。

The amount of people walking the streets with their faces glued to screens while listening to music — also known as “twalking” — is increasing.

越来越多的人一边走在大街上，一边听着音乐(也被称为“twalking”)，他们的脸粘在屏幕上。

In the last seven years “twalking” deaths have skyrocketed in the US, with pedestrian fatalities reaching a near 30 year high in 2018.

在过去的7年里，美国“骑车族”的死亡人数激增，2018年行人死亡人数达到近30年来的最高水平。

A research team is now developing the prototype headphones and testing them on New York’s noisy streets in a bid to reverse the trend.

为了扭转这一趋势，一个研究小组正在开发这款耳机的原型，并在纽约嘈杂的街道上进行测试。

Dr. Fred Jiang, a professor of electrical engineering at Columbia University, said: “We are exploring a new area in developing an inexpensive and low-power technology that creates an audio-alert mechanism for pedestrians.”

哥伦比亚大学电气工程教授Fred Jiang博士说:“我们正在探索一个新的领域，开发一种低成本、低功耗的技术，为行人创建一种声音警报机制。”

The headphones have miniature microphones and intelligent signal processing that detects the sounds of approaching vehicles. If a hazard appears near, the system pings an audio alert to the pedestrian’s headphones to warn them of the threat.

这种耳机配有微型耳机和智能信号处理系统，可以探测到驶近车辆的声音。如果附近出现危险，系统就会向行人的耳机发出音频警报，提醒他们受到威胁。

The scientists, writing in IEEE Internet of Things Journal said the research and development of the smart headphones is complex.

科学家们在IEEE物联网杂志上撰文称，智能耳机的研发非常复杂。

It involves developing a low-power data “pipeline” to process all the sounds near to the pedestrian.

研发内容包括开发一个低功耗数据“管道”，来处理行人附近的所有声音。

The pipeline will contain an integrated circuit that identifies relevant sounds and extracts the correct cues to signal impending danger — all while minimizing battery usage.

这条管道将包含一个集成电路，可以识别相关的声音，并提取正确的线索来提示即将到来的危险——同时最大限度地减少电池的使用。

The researchers are also using the most advanced data science techniques to design the high-tech headset.

研究人员还使用最先进的数据科学技术来设计高科技耳机。

Machine-learning models on the user’s smartphone will classify hundreds of acoustical cues from city streets and nearby vehicles.

用户智能手机上的机器学习模型将对城市街道和附近车辆发出的数百种声音信号进行分类。

The researchers, who have won two awards for their prototype, are now testing it both in the lab and on the chaotic streets of New York.

研究人员已经因为他们的模型赢得了两个奖项，现在他们正在实验室和纽约混乱的街道上进行测试。