Please select your home edition
Edition
Leaderboard FD July August September 2023

Science blog: Machine learning and whale song

by Ann Allen, NOAA Fisheries 3 Nov 2018 02:13 AEDT
A humpback whale surfaces near Maui Island © Hawaiian Islands Humpback Whale National Marine Sanctuary / Jason Moore

What do you do when you have too much data? It's not a question I've ever had to ask myself in my scientific career, and is rarely a problem when working with hard-to-study animals like whales and dolphins.

But this is exactly the question I found myself asking shortly after starting my new job with the Pacific Islands Fisheries Science Center's Cetacean Research Program in Hawaii in 2017. I was describing my job to my father, who is not a scientist, and trying to convey what we want to do.

We conduct research to understand the health and status of whale and dolphin species around U.S. Pacific Islands and across the Pacific Ocean, a VAST area. One tool we use to monitor elusive whales across this large region is a network of seafloor mounted recorders called high-frequency acoustic recording packages (HARPs).

The HARPs have hydrophones (underwater microphones) that can record the calls of whales and dolphins, as well as any environmental or human-made noise. We deploy HARPs for months to years at a time and we have to retrieve them to obtain data.

So far, our network extends to 13 different recordings sites, of which several have been monitored for a decade or more. This means we now have more than 200 TB of data, or almost 170,000 hours of recordings. If you were to sit and listen to all of that audio straight through it would take you more than 19 years. So if we monitor the whales by listening to their calls in the recordings, how in the world do you ever listen to that much data?

After telling my father all of this, he asked me, "Well why don't you just get those Shazam or Google people to help you? They teach computers to recognize human songs, I bet they can teach it whale song."

I paused. "Well, because it's just not that simple..."

"That's just not how you..."

"Because people have been..."

"Um well y'know, I don't know."

Thus challenged by my dad, I figured I'd find out if those Google people would be interested in teaching their computers to speak whale.

A friend who works for Google put me in contact with their Artificial Intelligence group. They responded almost immediately and enthusiastically; an ocean's worth of data sounded like it would fit in well with their current projects. They already have a lot of machine-learning techniques they've developed for other purposes that they would be excited to put to use for conservation. Machine learning attempts to mimic the learning process of the human brain by teaching a computer to carry out a task, rather than programming it to do the task step by step.

Automatic detection of whale calls isn't a new approach, but designing a good detector can be quite difficult. The ocean is a noisy place, with lots of sounds from wind, waves, fish, and even the recording equipment itself. And humans contribute an ever increasing amount of noise to the ocean, from things like container ships, oil and gas prospecting, military activities, and construction. The whale vocalizations themselves can also sound very different depending on how far away the whale is from the recorder, how it echoes off of the ocean floor and surface, and whether it has a local variation, or dialect, so to speak.

While some whales, such as blue whales and fin whales, make consistent, stereotyped calls that are relatively straightforward to recognize once you've seen some examples, humpback whales have vocalizations that are extremely variable. These calls make up phrases and themes that form songs. The songs change over time, as the whales incorporate new notes or swap the order of phrases, and many populations of humpbacks sing different songs. So how do you teach a computer to recognize whale song when the song keeps changing?

Researchers have been trying to solve this problem for quite a while. So far all the methods developed come with caveats: poor detection rates, a lot of user input to fine tune the detector for each call type and situation, or a high rate of false detections that require someone to go through by hand and pick out the correct detections from the bad ones.

However, there have been a lot of advancements in the last few years in artificial intelligence, and deep-machine learning. You may have noticed that your phone is much better about not making tie bows, I mean typos, when you use voice recognition these days. This is due to advances in machine learning, some made at Google. So with this in mind I figured I would throw Google in the deep end and see if they could come up with a solution to the complex humpback song problem. I gave them a subset of our audio data, with a collection of 'training labels', or humpback whale calls that are marked by a human (me). This training data is how the computer learns what humpback calls look like.

And it turns out that Google's computers are pretty good at recognizing humpback song! In less time than I thought possible (practically gale force speed by academic standards), they sent me timestamps of humpback whale song in our data. The computer isn't perfect yet, we still need to teach it to better recognize different kinds of noise from sites that didn't have as much training data, but it is already doing far better than expected.

There is so much new, important information we can learn from just knowing when there is humpback whale song throughout our dataset. In the Pacific Islands region, very little is known about humpback whale presence, seasonality, daily singing behavior, or population structure outside of the main Hawaiian Islands. We do not even know if humpback whales ever journey to some of the smaller and more remote uninhabited islands. With this data we can start to address these questions. Additionally, because our dataset spans more than a decade, knowing when and where humpback whales are singing will give us information on whether or not the animals have changed their distribution over the years, especially in relation to increasing human activity in some areas.

The possibilities don't stop there. We're now interested in teaching the computer to find other species of whales and dolphins in our recordings. The hope is that if we give the computer enough examples of different whale calls it will learn to generally recognize 'whale' sounds and get better at distinguishing between whale species and also between 'whale' and 'ship' or other noise. My hope is that in the future Google will provide us with a tool that scientists can directly use. A tool that we can provide new data, from a new location, and training labels for a new type of whale call, and it can then tell us where those calls are in the rest of the data.

So, if you're a scientist with acoustic data and labelled whale calls and want to help teach Google how to speak whale, whale then give us a call!

Check out more details of the story from the Google side:

Related Articles

Course des Caps Update: A promising start!
The eleven crews have begun their grand tour around the British Isles The Course des Caps - Boulogne-sur-Mer - Banque Populaire du Nord, which set sail from the shores of northern France, marks the first race of the 2025 season and kicks off the new edition of the IMOCA Globe Series Championship. Posted today at 8:42 am
AST and B&G extend high-performing partnership
Continuing a collaboration that has been critical to the success of the team AST has announced the renewal of its long-term partnership with B&G, the world's leading sailing navigation and instrument specialists, continuing a collaboration that has been critical to the success of the team over the past eight years. Posted today at 6:55 am
18ft Skiffs: Queensland 18 footer history
Decades of successful ideas and achievement Queensland's revival over recent seasons, which resulted in a two-pronged attack by experienced and young teams in new skiffs at the 2025 Giltinan world championship Posted today at 6:46 am
Marine Auctions: Special July Online Auction
The bidding will end on Tuesday 22 July at 2pm AEST The alternative way of selling any type of vessel or marine asset with proven and successful results. Posted today at 4:03 am
Transpac 2025 underway
Sixteen boats hit the line for the first start, departing LA for Hawaii Sixteen boats hit the line for the first start of three in the 2025 Transpac. Next stop: Hawaii. Posted today at 1:13 am
GKSS Match Cup Sweden & Nordea Women's Trophy D2
A challenging southerly breeze and short three-lap course put teams to task A challenging southerly breeze and short three-lap course put teams to task on the second day of racing at the GKSS Match Cup Sweden and Nordea Women's Trophy in Marstrand, Sweden. Posted on 1 Jul
Admiral's Cup 2025 | Interview with the CYCA Team
A highly experienced team for the revived Admiral's Cup regatta from July 17 The Cruising Yacht Club of Australia is fielding a highly experienced team for the revived Admiral's Cup regatta that will be run from Cowes on the Isle of Wight in the UK, from 17th July 2025. Posted on 1 Jul
Australian Hobie Cat Nationals entries rolling in
With at least three World Champions already entered, the racing is sure to be exciting too! There is excitement in the air for the Pitts Design and Construction 53rd Australian Hobie Cat National Championships. Posted on 1 Jul
Rolex TP52 Worlds in Cascais - Practice Day
Will Platoon Aviation's big breeze, big pressure experience prove key to their fourth world title? Of the three past and present world championship winning crews which completed their final practice today in typically muscular 25 knot breezes and big waves out of Cascais, Portugal it was Harm Müller-Spreer's Platoon Aviation which showed best today. Posted on 1 Jul
Some thoughts on provisioning for distance sailing
A new perspective on provisioning and time spent at sea One of the great joys of distance racing unfurls the moment that the dock lines are untied. Suddenly, the myriad packing lists that inevitably define most trip-planning efforts become about as relevant as a tax return from eight years ago. Posted on 1 Jul
Cyclops Marine 2023 November - FOOTERNorth Sails Loft 57 PodcastSwitch One Design