Alarms and sirens
Nothing screams urgent like the sound of a siren but working out the direction the sound is coming from is often tricky. New sirens are being tested but making them sound convincing is just one of several problems to overcome.
You might not know it but your brain is doing physics calculations all the time. Your brain listens to intricate differences and details in noises in order to pinpoint sounds to within 20-40 cm from 10 metres away.
To tell whether a sound comes from the left or right, your brain compares the differences between what it hears in each of your two ears. The slight difference in distance a sound has to travel to reach each ear creates an almost unperceivable delay as well as a slight difference in volume. Unfortunately, sounds can come from several places that have indistinguishable time and volume differences, and form what scientists call the ‘cone of confusion’.
Your brain also relies on your earlobes altering sounds coming from the front differently to sounds coming from behind you. “All the bumps and convolutions of your external ear are completely unique to an individual. Some frequencies get amplified and others get attenuated. And brains learn, as they develop, how the external ear modifies the sound” says Deborah Withington, retired Professor of Professor of Auditory Neuroscience.
Most alarms and sirens fall into the frequency range 1-3 kHz where the human hearing is the most sensitive, but this frequency range is also a black spot in our sound location abilities. Locating sounds through time differences works best for sounds with frequencies less than 1kHz and volume differences only work for frequencies above 3kHz. To add to the complications, our ability to determine sounds in front or behind us only applies to sounds above 5kHz.
Sounds are located more accurately if they have a combination of frequencies that utilise these three methods, known as ‘broadband sound’. “The more frequencies you’ve got the more accurately you can pinpoint a sound because it makes use of all the cues.” says Deborah.
Transferring this knowledge over to emergency vehicle sirens is far from simple. “Designing a good siren is at the interface of physics and behaviour” says Ken Catchpole, Director of Surgical Safety and Human Factors Research at Cedars-Sinai Medical Center. “One issue is getting people to respond in the best way” he says. “They need to recognise it’s a siren, respond to where that sound is but also figure out what sort of vehicle is coming along. You need a lot more space for a fire service vehicle than you do for a police car.”
Volume is also an issue. A loud siren will alert people far away but too loud and you’re potentially damaging the hearing of people not in cars or the vehicle crew. Too quiet and it doesn’t give people enough time to react. Sirens need to cut through the background distractions of music, speech or road noises and get past muffling car soundproofing. Current sirens resort to high pitched frequencies but these high frequencies are especially prone to damping from car sound proofing.
Broadband sound sirens may seem like the answer, but they have their own problem “it sounds different” says Deborah. We’ve become too accustomed to what a siren should sound like.
Alternatives sirens are being tested, “you can mix tonal sounds with some of the broadband sound mixed in and potentially get much better localisation” says Ken. But while tests continue to be carried out on sirens one area where broadband sound has been proven to make a real difference is fire alarms. “We’ve done research on ferries, cruise liners, airplanes, building and hotels looking at evacuations in a smoke filled environment. If you have broadband sound over the exit in smoke it cuts evacuation times by 75%. That’s live or die.” says Deborah.
The folowing links are external