Skip to main content

Review: Haikubox

This AI-enabled device can identify the species around your home by their songs and alert you when new ones arrive.
WIRED Recommends
Haikubox on wooden deck next to foliage
Photograph: Haikubox

If you buy something using links in our stories, we may earn a commission. This helps support our journalism. Learn more. Please also consider subscribing to WIRED

Rating:

9/10

WIRED
Finally, a Shazam for birds! Easy to install and set up anywhere. Impressively accurate at identification. Will alert you when new birds are heard (or any birds you set alerts for). Great app with links for exploring your local birds.
TIRED
Pricey.

For bird watchers, being able to identify birds by their song is the holy grail. Some people seem to be naturals, hearing a song once and remembering it forever. If you're like me—not one of those people—you've probably had the thought, “Why isn't there a Shazam for birds?” Surely if Shazam can identify a song with a few seconds of bad audio playing over some blown-out speakers, someone can figure out how to do the same for a bird singing clearly in a nearby tree.

That, in a nutshell, is what the creators of the Haikubox have done—created the Shazam of birdsong.

That in itself is welcome and remarkable, but the Haikubox turns out to be much more than that. It's one of the rare pieces of technology that actually increases your connection to the world around you, rather than cutting you off.

Neural Net

Bird migration started early this year. I know this because my Haikubox told me. Not in so many words, but it started announcing new warblers arriving by the middle of August, which means they're already heading south to their winter grounds in Central and South America.

With a full time job and three kids, I don't have time to get out and go birding every day. I likely would have missed the Cape May warblers when they came through for a couple weeks at the end of August. They never stay long, and I always thought they stuck to the birch clearings a good mile up the road. Thanks to the Haikubox, I know that while they do tend to spend their days elsewhere, they pass by my home in the mornings. I was able to see them because the Haikubox alerted me whenever it heard one.

This is the magic of the Haikubox—it expands your world.

Photograph: Haikubox

For something so remarkable, the Haikubox is decidedly prosaic in appearance. It's a 4 x 6-inch rounded-square box that's about 2 inches thick. On the bottom is a sealed exit for the power cord and a small microphone that records sounds around the Haikubox. While the device is weather-resistant, and I have had no trouble with it in the rain, the company recommends keeping it out of direct sunlight. Don't submerge it. Once you have a good spot, plug it in and connect it to your Wi-Fi network via the Haikubox Connect app. The Haikubox will start recording audio 24/7.

That's the end of the hardware, but that's not where the magic really lies. Once connected to your wireless network, the Haikubox sends its recorded sounds to servers at the world-renowned Cornell Lab of Ornithology.

The Ornithology Lab has thousands of birdsong samples and a neural net for processing them. Neural networks are a form of machine-learning software well suited to recognizing patterns in audio—it's how Siri and Google Assistant understand your voice. Likewise, the neural net can filter out bird songs from background noise. In order to find patterns, it first needs to learn what the pattern is. Cornell's library of birdsong recordings provides the training that the AI needs to learn which sounds are bird songs and which ones are you watering the garden.

Cornell has been tweaking its neural net for some time. If you'd like to experience this without investing in a Haikubox, you can grab Cornell's Merlin Bird ID app, which relies on a small subset of the data and an AI processor similar to what the Haikubox uses. Haikubox creator David Mann told WIRED that the Haikubox uses a modified version of BirdNet, which is called BirdNet for Haikubox.

Neither BirdNet nor BirdNet for Haikubox is perfect, but it's impressively accurate most of the time. Even better, you can use the Haikubox app to help the AI improve.

Multiple Views
Haikubox via Scott Gilbertson

To see which birds your Haikubox has heard and attempted to identify, you can use the Haikubox app for Android, iOS, or the web-based interface. The first time you open the app you set up an account, and then you can log in via that account from any device. The data in each app is the same, and I used all three in the course of my testing. I found the mobile app more useful for notifications, but I preferred to go through and explore the species info in the web app, since I could open eBird and other extra info in background tabs.

Once your Haikubox has had a few hours to record, it will begin to show you the species it thinks it has heard. Each identification gets a nice color image and a couple of links to Cornell's All About Birds and eBird websites for more info on that species. There's also a button to listen to the Haikubox recording and an option to tell the AI whether or not it's correct. You can even say, “No, it's a …” and enter the species name if you're confident you've got it right. If you're not a birdsong expert, don't worry. You can still let Haikubox know when it has recorded wind chimes or sprinklers or other noise that occasionally tricks it.

By default, the Haikubox will show you a screen with the birdcalls it has picked up in the last two hours and has medium confidence about. You can change the duration to the last six hours, 12 hours, or 24 hours. The longer time spans are good for figuring out what's happening at night around your Haikubox. There are also two other confidence settings: high, and no filtering at all. The latter I found too noisy to be helpful. High is useful for eliminating most of the questionable sounds, but I stuck mainly with the default.

Haikubox via Scott Gilbertson

There are three other views available for your overall birdsong data. One is real time, which shows you what's outside right now. In my case, this is almost exclusively red-breasted nuthatches and black-capped chickadees. Then there's a view with the top 10 most frequently heard birds, and finally a screen with all the birds heard this year. I found this one useful for going through and confirming the neural net's guesses. One thing I suggested to Mann is adding a button to listen to a known-correct eBird recording next to what the Haikubox found. This way, you could listen side by side, which might help with the tougher recordings.

Sometimes there are recordings that are tough to figure out. The Haikubox is convinced there's a mute swan around me, which is technically possible, though I have never seen one, nor has anyone else on eBird this summer. In cases like this, I waffled. At first I marked it wrong because the audio was very indistinct, but then it came up again and sounded almost exactly the same, and I wasn't so sure anymore. I marked it “not sure” and started keeping an eye out for a mute swan whenever I'm near the lake shore. That's part of what I like about the device; it provokes you to investigate the world around you more closely.

Another way the app does this is through alerts. The first alert I turned on was for new species. Whenever the Haikubox hears something it hasn't heard before, I get a message on my phone. But you can also turn on alerts for particular species. In the case of the Cape May warbler mentioned above, I was out of town the first time it showed up, so I turned on an alert for every time the Haikubox hears one. After a couple of days of missing them in spite of the alerts, I finally saw some.

For now Haikubox's neural net smarts only work in the US.  The company is currently beta testing Haikuboxes in Europe to work out any issues (like making sure speech algorithms work in multiple languages). Mann says the company hopes to launch in the UK and Europe in late 2023.   

Private Eyes

It's worth mentioning that by default the Haikubox shares its data with Haikubox's parent company, Loggerhead Instruments, as well as Cornell. Mann assured me there are multiple levels of filter designed to ensure that human voices are never stored. By default, you can see other Haikuboxes on the network. I found it fascinating to browse through and see what other people are recording, but if this feels like a violation of privacy to you, it's easy to turn off. Head to your Account page (the person icon in the upper right) and click on the box that says “Make Private,” and then click Save. Everything remains the same in terms of functionality, but other users will not be able to browse your data. In the spirit of sharing and citizen science, I left mine open for all to see.

There are two ways to buy the Haikubox. You can get one with a lifetime membership for $400, or you can buy one for $190 and pay a $70/year membership fee to access the data. It's expensive, but otherwise, bird watching is pretty cheap—all you need is a good pair of binoculars and a guide book, or maybe that's just how I rationalize the expense to myself.

Even if you're not a hardcore bird watcher, the Haikubox is an accessible way to get you to look outside. This is especially true for kids. Mine now check the app every morning to see what the Haikubox heard last night, and then head out to see if they can find them. It will awaken curiosity, and will lead you to a deeper understanding of what's happening in the world around you. 

Updated September 26, 2022: We've noted the worldwide compatibility of the Haikubox.