Your perception of sound and mine are very, very different. That's why my favorite headphones sound tinny and awful to you, and yours sound woofy and messy to me. People's ears vary so much in physiology that it's like we each get a randomized graphic equalizer at birth, with up to 20 decibel swings each way as we go up and down the audible frequency spectrum. Even your left and right ear are different from one another. Nura's adaptive headphones measure these differences with a short test, then tune themselves so that they sound amazing for every listener. We had a chance to pass them around the Gizmag office and speak with Nura co-founder Kyle Slater. And while it wasn't a surprise that they sounded fantastic, what really blew us away was how terrible they sounded when we tried each other's sonic profiles.
I'm not a huge headphone buff. I probably would have paid no attention to Nura if my Facebook feed hadn't lit up with glowing recommendations from a very particular type of friend: pro musicians. One of them, pianist Luke Howard, called them "pretty magical," and he's not the kind of guy to exaggerate, even a little bit.
So when I found out inventor Kyle Slater was coming through town on a whirlwind tour promoting what was already a wildly successful Kickstarter campaign, I asked if he'd come by the Gizmag office and show us what all the fuss was about.
You and I hear things differently
Nura's revolutionary idea is based on a fact most people have never heard of before. Slater explains: "People's ears, and what they hear, are very different. Your eardrum and the three bones and the cochlea is very different for everyone. The difference between people is a lot bigger than people realize.
"A 200 Hz tone, you might be particularly sensitive to that, where I might not be. And I might be much more sensitive to a 1 kHz tone than you are. If the headphones aren't matched to your hearing, you miss instruments. You miss a lot of detail you should be able to hear."
Think of a graphic equalizer, with individual volume sliders for a range of different frequencies. Now think about this: people can differ as much as 20 decibels in their sensitivity to a particular frequency. That's 8 clicks on an iPhone volume slider.
If you're 20 decibels more sensitive to a high frequency than me, then a pair of headphones that sound nice and clear to me will sound unbearably tinny to you. And what's more, as much as there's a difference between people, there's a similar normal deviation between your own two ears.
No wonder it's hard for some people to find headphones or speakers they like. And this is the point Slater and the Nura team started at when they were trying to build the perfect set of headphones. "Usually as an engineer, if you're talking about building something, you can write down on paper a set of specifications for what would be the ideal device to make," he told us. "With headphones, I found I couldn't even write down a set of specifications, because they all change with each person."
They came up with the idea of a set of headphones that adapt to every ear, that use EQ to boost the frequencies you can't hear so well, and tone down the ones you're extra sensitive to.
How Nura Headphones analyze your ears
Great idea, but how? Slater and co-founder Luke Campbell fell back on specialist knowledge they'd built – Slater through a Ph.D focused on psychoacoustics, Campbell through a Ph.D in hearing science. They discovered a unique way to use a simple hearing test.
"It's called the otoacoustic emission, it's a test that's used for screening newborn babies for deafness, " explained Slater. "We use it in a slightly different way. What actually happens is that when you send different sounds to the ear, it vibrates the membrane, which vibrates the three bones, which stimulates the cochlea. The cochlea then sends an electrical signal to the brain.
"But it also sends a very faint signal back through those three bones to the membrane, which vibrates it slightly.
"It's not a reflection off the eardrum, it's actually generating a sound corresponding to what you're hearing. We have a microphone in the main earpiece that picks it up. What that signal contains is basically an indication of your sensitivity to the frequencies of sound we're feeding into each of your ears. Once the system knows what your hearing curve looks like, it then equalizes the signal to each ear, to lift the levels of those frequencies you're not hearing so well."
With that, Slater handed me a set of Nura prototypes. They were odd-looking but solid, with an earbud poking out from the middle of an over-the-ear design. Putting them on, the first thing you notice is how well they isolate external sound. Almost totally.
So, how do they perform?
Plugging them in to the Nura app, the next thing you hear is a little frequency sweep, which the headphones use to identify people who have previously used them.
Since the headphones didn't recognize me, the next step was to do that 30-second otoacoustic test. They played a series of tones, getting quicker as the pitch rises, and then a graph of my hearing in each ear popped up on Kyle's phone screen.
It matched what I know about my hearing. Years of ride cymbal abuse on a drum kit have taken the edge off the top frequencies in my right ear. I used to leave my right earplug slightly out of my ear, for better immersion in the sound on stage. Because these things actually measure the very signal your brain is getting from your hearing mechanisms, they account for hearing damage as well as natural variance.
From that point, we played a bunch of songs, switching between a flat EQ profile that Kyle told me is effectively the Nura phones doing nothing to the EQ – which sounded comically bad to me – and a second profile that's tailored to my own hearing, which sounded a ton better. Great, even. Warm and crisp and clear, with so much detail that you really do feel like you start noticing instruments that might have blended into the mix before.
The Nura headphones deliver bass through a completely separate channel to the rest of the sound, with a large vibrating bass speaker in the headphone cup while the higher frequencies are handled exclusively by the earbuds. Thus you can "feel" the bass a little more than with most headphones, but the sound loses none of its detail. And they're pretty much totally silent to people sitting next to you.
They sounded terrific, but what really dropped my jaw was when we passed them around the office and saw just how different a sound they created for each of the guys on the team. Michael Irving's profile showed extreme sensitivity to very high frequencies in the left ear, as well as some other interesting spikes. His profile sounded absolutely awful to me, but he said the Nura phones sounded amazing.
Kate Seamer was also blown away, but she's way less sensitive to low frequencies than me, so her profile sounded woofy and muddy when I tried it. It was like this for everyone. My profile sounded actively horrible to everyone else, as well. The difference was astounding, it just blew me away to realize how differently we all perceive sound.
For comparison, I switched to my own office headphones, a set of Koss BT540i bluetooth jobs I use all day as video mixing headphones, - and was surprised to find they didn't sound massively different to the Nura set when it's tuned for my ear.
Kyle figures that's because as a longtime semi-pro musician, I'm more aware of what I need from audio gear than most people, and I've done a good job of finding headphones with a sound that's well matched to my hearing. I briefly consider whether he's just buttering me up with compliments, but the real proof for me is in how awful everyone else's profiles sound, and the way the headphones instantly recognize me each time I put them back on.
Besides, before taking this idea to consumers, Kyle and the Nura team took it straight to the toughest audio critics of all – mixing engineers in the music industry.
Taking Nura to the toughest critics
"We're about music, and the journey from the artists through the sound engineer to the listener," Kyle told me. "All the hard work that people put into that music. Obviously we're proud of our product as a technology company, but we don't want to go and tell the music industry it's great. We thought it would be arrogant not to go to them, that it would be far better to show it to them, be honest about what our product's doing, and listen to their feedback.
"I was in LA last week with Frank Tétaz and Gotye, they had a listen. I was over in London and showed Warp Records and Young Turks, Abbey Road, Mix mag, great places. The reaction was awesome.
"I was really concerned with how the music industry would view what we're doing – we hoped the perception would be that we're delivering their music exactly the way they want it heard, rather than using technology to mess with their mixes.
"But look at the current ecosystem – every set of headphones, every set of speakers, every room is asserting its own sound profile on the music. We're trying to make headphones that have no sound of their own, that deliver exactly what the artists are making – we don't want to have a say in the music. We're in hearing science, not the artistic side and we don't believe we should impose a sound on top of that. We see our role as very passive."
Fundraising goals smashed, time to move onto production
While Kyle's in the office, the Nura Kickstarter clicks over US$300,000 in pledges, having smashed its $100,000 goal in a matter of hours after launch. As I write, it's now sitting at more than $650,000 with a month left to run, meaning the Nura team is funded up and getting ready to roll on production.
There's work yet to do. The prototypes have chunky circuit boards hanging off them that'll be minimized and integrated on the production jobs, but the external design is getting close. I asked if that electronic gear is going to make for a heavy headset.
"It won't be the electronics that make them heavy," said Slater. "If they're heavy it'll be because of material choices. Metal is heavier than plastic. But we want something robust, that we can guarantee we can manufacture. Having honest materiality, that's what our industrial designers call it. It means it costs us more, but we can guarantee it won't break and they'll be robust."
Likewise, Kyle says the team hasn't finalized the sound of the headphones yet, either: "The profile that we're using, the way we're choosing to take your objective data and turn that into the sound for each ear, that's something we'll continue to refine. It's something that with this platform, we can refine. We'll be working with the music industry to develop the golden profile, the golden mapping, if you like. For this prototype we've done it in a theoretical way that most people say is more clear and sounds much better."
Presumably they can also be upgraded in terms of how many frequencies they test for in the otoacoustic training routine. Right now, they measure your hearing response at just 11 different frequencies to build your hearing curve. Given how far the response can swing between those points, I'd be happy to go through a much longer initial tuning process to tune them at 30 or 40 frequencies. As much fun as it is showing the technology quickly to your buddies, I think most serious audiophiles would be very happy to wait longer for an even more precisely tweaked curve.
The Nura headphones are scheduled to start shipping in April next year, assuming all goes to plan. They'll retail at $399, priced to match the premium market, but early Kickstarter backers are being offered them for half that price for the time being. (Full disclosure: I couldn't resist, I plonked down an early bird order before Kyle left the office, my first Kickstarter backing.)
The truth is, all headphones have their own sound. Most people, if they're willing to put in the time, will be able to find a set that sounds great to them, a set that by random chance happens to match up reasonably well with their individual hearing profiles, like I have. Nura's ingenious technology takes all the guesswork out. They sound amazing to everyone here who's tried them, they've turned my understanding of hearing upside down, and we're very much looking forward to seeing where this technology goes from here.
Nura's Kickstarter is live now.
To overcome that problem I bought a pair of simple, noise blocking headphones (non-electronic) - $14.00 at Home Depot. They're a little large and obtrusive - you might be embarased to use them in some public places - but they've improved my ability to use my cell phone 110%. I can drive on the interstate with the windows open and talk in a normal tone on the phone. There are expensive, electronic noise-canceling headphones available that are cosmetically more appealing, but they're anywhere from $150 to $400. Give it a shot. For $14 ya can't go wrong!
Wouldn't it be great if that technology were someday transferred to hearing aids? I'm sure I can speak for the multitude with hearing loss like I have when I say that today's best hearing aids are both expensive and inadequate. This is probably a much bigger market opportunity than headphones!
Self-tuning hearing aids for under a grand! What a concept! GO, NURA, GO!!
On the one hand, it should not matter what the condition of a user's ears are--they are what they are when exposed to the sonic environment of the world around them. If that environment is reproduced and delivered to the user's ears faithfully, then that is what the user should perceive as the "best quality" sound experience. Sonic accuracy is an objective parameter. The subjective part comes in when, for example, one expresses a preference for the sound of one guitar over another. But once the choice is made, the "ideal" reproduction chain should not affect that chosen guitar sound, however bright or dull it may be.
On the other hand, if the headphones can compensate for hearing loss, or can compensate for some other playback system non-linearities (since the "ideal" condition may be otherwise unattainable), then OK.
Again, I don't want to take away whatever "pleases" the listener, but I question the basic premise of altering sound depending on the physical parameters of healthy ears.
I would suggest that the extent of the equalization required by most people is a good deal less than 20dB.
But the feedback approach to characterization is interesting.
Thanks, Dave
Surely you set up your EQ curve on your playback equipment to what suits you, as you listen through your own headphones. As I see it , you don't need to spend $300.