On Thursday, microphones mounted outside two buildings in Manhattan went live.
Bright yellow signs that say “Recording Underway” announced their arrival.
But these devices are not eavesdropping on your conversations.
A group of researchers from New York University and Ohio State University are training the microphones to recognize jackhammers, idling engines and street music, using technology originally developed to identify the flight calls of migrating birds. Think of it as the Shazam, the smartphone app that can identify songs, of urban sounds.
Snippets of audio, about 10 seconds each, will be collected during random intervals over the course of about a year to capture seasonal notes, like air-conditioners and snowplows. The cacophony will be labeled and categorized using a machine-listening engine called UrbanEars. The sensors will eventually be smart enough to identify hundreds of sonic irritants reverberating across the city.
The goal of the project, Sounds of New York City, or Sonyc, is to create an aural map that a group of researchers hopes will help city agencies monitor and enforce noise pollution, and will empower citizens to assist in the process.
During the first phase of the five-year program, which is primarily financed by a $4.6 million grant from the National Science Foundation, about 100 sensors will be mounted on the second floor of public buildings, first around N.Y.U.’s campus near Washington Square Park and eventually in Brooklyn and other parts of Manhattan.
Each $100 machine contains a small computer processor and a Wi-Fi antenna that connect to a custom microphone, shielded inside a 10-inch-long foam wind guard. The machines are also outfitted with plastic spikes to ward off unwanted interference.
“Pigeon poop is the killer of all microphones,” said Charlie Mydlarz, a British researcher who worked in an anechoic chamber, a room designed to deaden sound, to ensure that the devices met noise measurement standards set by the city.
So far, the contraptions have been tested in the heat, Dr. Mydlarz said. But the cold of winter has yet to arrive.
The antagonizing effects of noise have been well documented: hearing loss, disrupted sleep and cognitive impairment in children, among other ills. One study found that nine out of 10 adults in New York City are exposed to noise that exceed levels considered safe by the federal Environmental Protection Agency. Others have made links to a loss of productivity and property values.
But taming the city’s din has been an elusive ambition. For more than a century, groups like the Society for the Suppression of Unnecessary Noise, the Noise Abatement Commission and Citizens for a Quieter City have been waging war on this unwanted neighbor.
Still, New Yorkers continue to howl about the racket. But there are limitations to the way humans can respond to and even report these issues, said Juan Pablo Bello, the lead investigator for Sonyc.
It takes an average of four days for one of the city’s 53 noise inspectors to respond and review to a complaint, according to the city’s Environmental Protection Department. By then, the house party is over and the cat has calmed down. And grievances filed through the 311 system tend to be dominated by certain groups of residents. Manhattanites are two to three times as likely to report a disturbance than those who live in the other boroughs, he said.
Enter “smart” technology.
Dr. Bello, who is the director of the Music and Audio Research Lab at the Steinhardt School of Culture, Education and Human Development at N.Y.U., is leading a team of specialists in big data, citizen science, data visualization and machine-listening to create a network of sensors that will not only monitor noise, but also possibly track the effectiveness of noise ordinances.
Recording devices in public places might give some New Yorkers pause, but Dr. Bello said conversations picked up by the microphones could not be reconstructed from the recordings. An independent acoustical consultant was hired to make sure.
During the next phase of the project, sensors will transmit data about the time and day that, say, a dog barked or a street performer crooned, and an estimate for the sound level.
Eventually, an app called Urbane will allow users to interact with the data, while another app will complement 311 reporting and possibly help New Yorkers track how complaints are handled.
“There’s a lot that needs to be investigated and a lot of things that need to be solved before it’s in a place to really help people,” Dr. Bello said.
So far, he said, the reception to the project’s potential has been enthusiastic. The team is scouting new sites for sensors, but the process is time-consuming because they need a strong internet signal, a power source and approval from building owners. N.Y.U. has also made efforts to talk to nearby residents to explain the program.
Ted Timbers, a spokesman for the Environmental Protection Department, said officials “look forward to reviewing the data produced through the Sonyc research project to determine if it would be helpful in our enforcement efforts.”
Other cities such as Chicago have deployed a network of sensors that act like a fitness tracker to measure livability through the climate, air quality and noise information.
But according to Claudio Silva, the interim director of the Center for Data Science at N.Y.U., Sonyc researchers are conducting the first large-scale analysis about the sources of noise in the wild.
Dr. Silva will be wrestling with troves of data from the city to contextualize noise with factors like traffic and census information to generate “a full geometry” of the urban landscape.
“On the geek side,” he said, it is exciting to be able to translate technology in ways that “could change the city.”
Making stronger correlations between crime patterns and noise levels, for example, could help shape public policy. Or it could help someone decide where to live or where to start a business.
“Having this information can inform people on making the best decision for their lifestyle,” Dr. Silva said.