this season is shaping up to be a lot more unpredictable than recent years. you're watching d. w. news line from berlin. coming up next are documentary series doc film. we're taking a closer look at the language of wales. i'm terry mark and for me and all of us here at the w, thanks for watching. ah, the green you feel worried about the meals of the on the green fence? we need to change. join me for the size of the green transformation for me to use for the plan. ah,

ah, i hear off canada pacific coast researchers are hoping to make a long hel, dream come true. behavioral biologists and i t experts have teamed up to create programs aimed at deciphering acoustic signal from animals. could artificial intelligence helped to identify patterns and the sounds made by marine mammals. we soon start to understand what it is that whales are talking about. the the coastal waters between alaska in the us and canada, vancouver island, are where a group of killer whales or orchids,

their summers. a team of scientists from the deep l research expedition are preparing to embark deep al stands for deep learning applied to animal linguistics. i was a computer scientist, m a nerd from germany's university of atlanta and has spent years working on automatic recognition for human speech patterns. can the same methods be adapted to animal languages? underwater microphones, embedded in tubes, serve at the expeditions, ears for 3 summers. teams of computer scientists and biologists have set out to record org calls and document whale behavior. rachel chang from the life nets institute for zoo and wildlife research in berlin is looking for patterns between behavior and animal vocalizations that could help to decipher whale language.

i assume they would have a very different system which fits into our language around $300.00 or kind of are identified as northern residents, as they spend each summer along the coast of alaska and northern british columbia. they live in small family groups and are very communicative to different families use different dialects. and is it possible to discern the meaning of individual calls? here the researchers lower the highly sensitive microphones into the water. the 8 hydro phones can record sounds of to a frequency of 100 killer hurt that's far higher pitched than what is audible to the human ear. thanks to the network of hydrophone and acoustic triangulation,

the researchers will later be able to calculate the positions of the whales to avoid disturbing the whales with engine noise. the research trying around is equipped with an electric motor. while visibility is limited under water, sound waves are transmitted over considerable distances. that's why a communication system with loud calls is clearly beneficial. they're very tied together and when direction change, circling, body twisting, plus a love co location. they're socializing like us to rob and frequent twisting and then jump to jump on to each other. sounds like they talk about the plan. what are we going

to do next? or cause only spend about 5 percent of their time at the surface, which makes systematic observation quite difficult. the research team uses drones to document the behavior of the animals diving expeditions with whales are prohibited in canada. the scientists are looking for the smallest meaningful units of communication which whale is calling? and which one answers are some sounds repeated more than others. biologist, alexander hall, examines the recording. each call comprises a series of brief pulses with sound like melodic curves to human ears. each shift in sound could be meaningful while recording the whale calls, the researchers also document the behavior of the animals.

so there for individuals in this group here that are circling, there were 4 back there with the male. there's one that i can see on the corner of my eye coming towards it. and the mom and cas $81011.00 individuals minimum. the more data available, the easier it is to train deep learning programs to decipher away language. it's therefore a major advantage for the researchers that over 20000 hours of orca curls have already been collected. more than for any other animal species. the whale researcher jared towers have been tasked by the canadian government with observing different oracle populations. back in the 1900 seventy's scientists here began documenting individual animals as well as their group

structures. anyone always trade underneath here shows just gone off. i had the howard has no problem telling the or because apart the scratch here and i want a one away persisted for least a couple years. so the, the way that we identify the individuals, no matter what population they belong to, if it's by appearance and you get used to looking for a certain features on a individual killer. well, the dorsal fins and the patches around a thin have different shapes, enabling scientists to catalogue each of the northern residents. every family forms a lifelong bond. what we're looking at with all of these families. and that'll email leading the

group and the fathers of their offspring don't play much of a role in that family. jared towers works for a federal institution that safeguards canada waters. his former boss, john ford, was the 1st to distinguish between the disparate calls of the whales. ford's research revolutionized our understanding of the communication system used by the resident archive we're listening to or calls in a class wheels. that's the exciting part of underwater listening. as you're getting a window into their, into their life that you would never, you never see john for discovered that the whales used about 50 different

called different families prefer different types of call. these were nicknamed dialects and used to help identify individual families when they're missing, the very stereotype signals is simply keep in touch with everybody in the group name, exchange them. they're constantly monitoring each other's location because they can tell where they are. because of their very directional hearing, and they can monitor the sort of behavioral state, the excitement level, the arousal state of all the other animals in the kindred. northern resident orca visit, pebble beaches daily. we can hear them rubbing right now. they're making socializing sounds and you can hear the pebbles getting pushed around. and they just love all sides of their body. just under the,

in the cell part of the being only a few or groups worldwide engage in this sort of body rob. this behavior is not genetic, rather it's a tradition passed on within families of the northern residents, just like their language. me back on the german canadian research boat. it's a challenge to locate each family among the $300.00 individuals that make up the northern resident population. the orchids are constantly on the move in an expansive water, the size of belgium, the expedition coverson area from vancouver island to the southern tip of alaska. steegal indicated where schools of sam and might be. and this is where orchids often hunt to, ah, calls from the club, can be heard on the underwater microphones. the team tries to determine the

position of the whale that comes from the left. that comes from the right. they compare the calls with a catalogue compiled by john ford, but they encounter discrepancies. and it looks quite the calls of class and 9 are used by different whale families from the calls from the same class should be almost identical. but these differ in length, melody and harmonics. human analysis so far has amounted to only a rough classification. looking at the fact of rams, i'm very sure that we can, you know, achieve something that rival the human performance, to well families approach. what calls are they exchanging through a quarter after

a hear the call back and thought to the front turn around. the researchers are interested in which group is calling, which one answers and which sounds they're using. ah, so look here that i for pushing the boat carolina and 23. there's a lot of variation also. an interesting way here is that 823 families through the course, or is it different? yes, and how people differentiate different. if you have a lot of quotes inside to try and testify, the programmers use algorithms or classifiers to automatically analyze millions of

ways calls in order to compare recurring sound patterns with recurring behavioral patterns. same method is used to decipher the meaning of individual words in foreign languages. for m a mood and his team, it's no easy feat to automatically filter out the week or a call from the constantly fluctuating noise of the ocean. by matching, thousands of all caught up on the input side with the target value on the output side. they train neural networks. these layers of neural networks then learn bit by bit which sounds are typical for the purple peaks indicate a high probability of having found a whale call it the 1st time and machine has been trained to automatically detect

orca sounds. still clustering calls into similar sound groups remains a big challenge. when comparing tens of thousands of recordings always calls are sorted by what they have in common. but the cluster is still too crude to form reliable classes. further programming is required. the team suspects that the northern resident or cause have a complex language system and made and process more intricate sound patterns than humans. but how much is even known about the acoustic capacity of their brains? this is telegraph cove on vancouver island, where elma newt and rachel cheng have an appointment. only a handful of researchers worldwide have expertise on whale and dolphin brains.

neural scientists, laurie moreno is one of them, is an autoplay. this is an orca scholar. a cranium bride opened a worker and in the front you will find the melon, where they do the good they directly location. and this part right here, this large part is where the break with that. okay, yeah, yeah. human and whale brains generally share the same the malian architecture. both cerebral corta sees are wrinkled and complex, but the orca brain is 5 times bigger among the largest of any animal. and unlike the human brain, the oracle also has a para limbic globe. the, the part of the brain really fascinates me is this part, right?

yes, we have this special extra load called the parallel flow. so it's connect feelings, emotions, and thoughts. and it is so complex in the order of brain, it has become its own low. okay, so that could mean that a lot more of the communication has to do with their feelings, with their emotions that communicating feelings. communicating emotions are very important. ok to north wales have a far larger acoustic area than human laurie. marino's team also discovered a 2nd acoustic cortex now we know that they also have a sound processing area in the same area that we process language based on what you said about how the process so nor can we say that we need

a much finer window when we analyze so it's not just fine resolution that we need. we need a way to actually to actually do the analysis at a much, much more complex level than our brains come to. the new brain scan methods have reveals that orca had a far more complex acoustic system than previously believed. could this be true for other marine mammals that use sound for their communication and orientation in contrast to other or groups, the northern residents at some point stopped hunting, seals and other marine mammals. today, they feed almost exclusively on the largest salmon and canadian waters. the chinook

stalks of the fish, however, have been declining rapidly since the late 1990 the the northern residents have to travel increasingly long distances to find the salmon. that filled remains. and despite stringent fishing quotas for the species, their numbers continue to decrease. the news. jared towers from fisheries and oceans canada monitors. if people are complying with the fishing regulations as seem to be the ones that famine aren't wasteful. abundant, not a killer whales aren't getting enough and either the population to go down a little bit. the northern resident or cars are classified as a threatened species,

families low. they don't simply go out and try catching other fits. they abandon learns, or they haven't been found and all interior prairies or it's not like they're going to go and start killing harbor skills. just wondering on the famine around the substantial human appetite for salmon has decimated stocks to critically low levels . ah, the ruins of the salmon fishing industry serve as a reminder of the ones abundant fish stocks. wild canadian salmon were in great demand around the world. the survival up in northern resident or cause is tightly linked to the fate of the shinnecock salmon. mm. last, barrett, leonard is a geneticist and behavioral ecologist at the vancouver aquarium. he visits the

resident or cause every summer to document their long term health and measure how their bodies size and shape compared to previous years. the to ensure the images are comparable from year to year. the drone hover is exactly 30 meters above the whale, almost over the groups. calf is almost a year old and it still suckling from its mother. the social skills of orchids are illustrated by how they deal with their prey. you've got a species that is highly social, that shares all of its food. so they're compulsive shares, killer whale, so they catch fish though, you know, one rallied half at the headed off to somebody else and that would buy them. and then another one. i'll come in and take a little by the adults sometimes lose living sam into the younger

wales in this way, the juveniles learn to hunt their prey in a playful way. at the vancouver aquarium researchers compare the images from different years. one indicator of the whale state of health are the white patches above their eyes. if there are not enough that preserved in the neck area, the angle along the ice spots will change. comparing the images side by side researchers observe a life threatening weight loss. me in response to the decline of natural stock. the fishing industry installed aqua farms in the regions, the remaining wild salmon migrate through nets protect the farmed salmon from workers and other predators. me. the predominantly norwegian operators introduced farmed atlantic salmon species and with them new viral disease

is not previous found in canada as pacific northwest. these fish farms raised as much salmon as possible in a fine space. it's all about maximizing profit. by product feces, germs that escape into open waters growing concern of outbreaks among the wild. chinook salmon. if the primary food source of the resident orca population continues to decline the whales own survival will also be endangered. in a while, researcher alexandra morton is studying the new diseases affecting the salmon. she visits the research, try me around to share the results of her analysis or addressing results my lot if you'd like to see them,

virus research where we found it and the same in terms such a big problem for the salmon and the whales because of where they're located, where do we do a molecular test? to basically just looks the genetic sequence are part of the genetic sequence of the virus. salmon can viruses through their gills. the pathogens then infect their blood stream causing oregon failure. me. when the salmon farms moved in the orca stayed away that made alexandra morton suspicious. she took samples near the farms which were then analyzed in special laboratories. she meant the organs of both wild and farm salmon. 95 percent of the fish she dissected was sick. a large proportion of the wild

tammond also became infected through contaminated water. the pathogens escaped from the fish barb with a type occurrence and spread to the regions the young, wild salmon have to pass through as they migrate. to kilometers further north, the a $42.00 family searches for prey along the shore lines. the research expedition has reached the fjords. fisher channel, once famous for its ample fishing grounds. the whales hunt by day and by night they don't light for orientation biologist floor and sullivan compares her recordings to known call types and notes

down behavior to the or tend to use certain calls in a social context or more when hunting. all recordings are later re analyzed in germany with the help of deep learning programs in their search for prey. the northern resident or time now often cover 80 to 120 kilometers a day. hunting roots that have been passed on for generations to identify individual wales and examined changes in group structures. programmer unwell. schmidt and biologist james field are working on automatic thin recognition software. it works similarly to facial recognize programs, the process of ident fine whales by their fins. and dialect was used years ago in the famous case of an orphaned cath known as a 73 or springer. when her mother died,

then 2 year old springer was separated from the rest of her family. eventually 300 miles further south. the lone calf was found off the coast of the adul scientist could tell from her calls that she was in northern resident springer was brought back and reunited with her family. so while researchers had been able to identify whale dialects for 20 years, now understanding the language of whales represents a whole new challenge the call systems used by different well families change little from one generation to the next spring or now has to camp of her own spirit and storm, roughly 18 years old or typically spend their entire lives with their mothers.

for the 1st 2 years they are suckled. and over the following 8 years, they learn what they need to do to survive. another orca group has joined springer's family. together they search for salmon close to the shore. the they are hunting strategies vary depending on the local environment. this gives them a wealth of experience. bad, like their communication system is half down through generations. the, their group behavior during the hunt seems to be coordinated. the whales exchange a stream of call the the orders are still able to find enough food by spending more and more time hunting. for a calf,

it's one of many lessons in sam and hunting. for the scientists, it's a rich source of data. the human activity greatly impacts the whales habitat on this container as well as the orca. know that org or call us to call the room the sound of passing ships doesn't frighten the whales, but the noise does interfere with their communication and compromises their hunting ability. the canadian government is investigating the impact of ship noise on marine mammals . so we listen for their vocalizations which can cover a range of 10 kilometers or so. man made noise in the world's ocean doubles

every 10 years. one of the concerns we have about vessel noise is its potential effect on the eco location capabilities of the whales. the whales vision is very limited, underwater because of murkiness and the water, especially at night. they can't seem typically in these waters more than a whales length or so. so perhaps 10 meters max. in a quiet condition, the wheels can probably detect should not sammons at a range of perhaps 200 meters. but that might be significantly reduced by masking of by boat noise. similar to bat orca scan their surroundings by admitting clicking sounds and listening to their echoes. the noise from ships math those echoes to reduce its own engine noise. the expedition boat runs on an ultra quiet electric drive. sounds like an end, but it's in there,

we have the boat and there's another way we can hear the on one it's it's much clearer. so we have this electrical noise here, 10 to a hertz, and we have a look at the engine. although electric engines are far quieter, there is nevertheless some interference. rachel chang inspects the static noise and examines the control cables. are now much better now. the sound of electric motors is still rare in the ocean, which is perhaps why the ship triggers the whale of interest. the boat stops in order to maintain the requisite distance of 100 meters, curious, or cause or not as cautious and inspect the boat and its crew. the

scientists wonder whether the whales might use distinct cause for different boats. in any case, there is immense curiosity on both sides or cause are extremely social creatures and love to touch one another. the thing that's a 23 and a 25 group floating away pretty soon on the water surface where others just mingling around i do jumps also part of their communication or just a bit of fun. many

of their behavioral patterns are still only partially understood. this is also due to the fact that or cause rarely roam close to the surface where humans can observe them some distance away. another group of workers suddenly appears marine biologist, alias undo, haul, wants to get an overview with surface. go. now, if we could have harvard site this is a completely different type of orca. a golf, unlike the resident whales, these transient or kind don't feed on salmon, they hunt other marine mammals. you know, i don't hear, you will call it,

but it seems there hunting there for in another contrast to northern residence, transient whales hunt almost silently. that is because dalton seals and other whales, their natural prey, have excellent hearing. that means they know exactly when they're in danger. the transient orca is also inspect a rocky plateau for prey. it's a tactic that appears to be successful. 6 0, we hear that the blood at the surface reveals a kill. the transience begin to celebrate their catch. and now exchange many calls back and forth acoustic. they're very different from the residence.

curious if they classify or can actually pick up the different the the homes of the roughly 500 transients and 300 northern residents overlap. but their calls different so much. it is if they're speaking different languages, dolphins, and other marine mammals can tell the difference. while they immediately flee from transient whales, they rush intentionally toward the residence as soon as they hear their calls, to hunt and play with them. below deck m a nerd and christiane bag la work on the automatic call classification. yes, you're welcome. and it shows here the detection of the seal was have 2 seconds, 2nd, 2 seconds. you can clearly see that the incoming so if there isn't all they will be

to texas quite what i like is that the confidence is so high. so the algorithm is quite sure that yes, yes, the automatic call detection work, the challenge is still to match one group of course with another in human language, the equivalent of these sounds, sequences might be simple statement me . the northern residents consist of 32 families. they typically use different call systems, but are still able to communicate with each other when they meet up. looks like the other group is joining them. the calls between the groups go back and forth announcing their arrival in a kind of greeting ceremony. the

thing they're saying here here if they're so close but i know that it's not the same call, let's say what was really one call that the the individual was retention or was the same was doing like combination meetings between family groups used to be more frequent when sammons dogs were more abundant and they had to hunt last seems to be a lot more melodic sounds or vocalizations or a lot longer. i would say the more complex they exchange, another series of calls before the meeting, slowly breaks up. these sounds might signify goodbye, but the researchers will need to compare many other similar exchanges before they

can interpret it more decisively. there are thick banks, a fog between vancouver island and the mainland. at the end of summer approaches. tracking down the whales is more of a challenge. the researchers only option is to use acoustic localization with the hydro phones under calm c conditions, the sound of the orca calls can range up to 10 kilometers emma knew it takes the location forecasts. 1200 square meters over there at 1 o'clock. 123, or localisation is also important in order to interpret the call

is like 2 or 3 year 2 or 3 there. and then it's always the same call what we're here right now. so it's probably, i'm here and the other one recognizes this one. 0 friend is over there. what i would do is try to cluster exactly those calls where i'm convinced that i've heard the same. and then from the sequence, one could make the assumption that the signal and response signal signal response to the last 3 expeditions is coming to an end with data from a total of 20 weeks, a field research on board. a few months later, the researchers meet at the pattern recognition lab at the university of along and neuron germany. rachel chang has an appointment with the programming team. manuel schmidt has run an automatic comparison of 1000000 calls from your company and take a look at the clustering. i made some changes and it's

a little bit better than in 3 here. there it looks very similar, but it's makes was, and i still need training background noise is hinder. the automatic sorting of various called a newly programmed noise filter increases the precision difference from the artist. there are still small errors, but on such a large scale where millions of calls are being automatically compared, they become less significant. i think we're, we're at the level where we can run through a couple of tapes now and several tapes that you can use those class the sequences in order to find those language pattern. rio as low as like

a semantic structure. and those can be interpreted as a call followed by an answer followed by another call. the new clustering results are better than expected. rachel chang's task is now to assign recurring call patterns to certain behaviors. she compares the new clusters with the existing call catalogs and discovers reappearing matches that might be meaningful. the algorithms have learned to differentiate the cause of different oracle families that would normally take human researchers years of training. 9 c in the future, this will allow automatic detection of which family is calling and the ability to follow the subsequent dialogues between families. the next step is comparing millions of call sequences,

which calls appear together frequently accompanied by which similar behavior. when decoding human languages such connections might reveal meaningful possibilities, such as fit and share or table and plate. the deep learning programs cannot be had work completely independently without human control, there could be a huge increase in assignment errors. the 1st matches appear, the same call sequences were used in a similar context. but there are exceptions. the comparison far are not conclusive. so the way of hearing caustic ms point us to this thing is socializing . well, so find the 2nd time in the socializing context. the same pattern we expect to find them in the 3rd one also or in more than we didn't find in this.

that doesn't disprove it. i mean there could be different explanations. we just don't have enough data to say that the german canadian research team is not yet able to create a kind of dictionary of the orca language. but now they do have tools to compare. we'll calls in a more detailed and systematic way than was previously possible. and as more recordings, the whale calls are shared and made available for training deep learning machines. the factor it will be to recognise the subtleties in communication patterns. the humans have long faced limitations when it comes to understanding what animals might be talking about. but with the help of ai, a new era of research might allow us to decipher the secrets behind their communication. the

me the in awe give start, deliver, deliver, deliver. go. good morning to get ready. log in. to get started, deliver deliverables. china seed carriers work 7 days a week. and if they're too slow, their wages get global 3090 minutes.

oh, the fight against the corona virus pandemic now has the rate of infection in developing what measures are being taken? what does the latest research say? information and context ah, i recently updated that special monday to friday on dw, the little guy. this is a 77 percent platform for african issues to share ideas. you know, on this channel we're not. and then it gets tuffy. young people clearly have the pollution, the future. 77 percent now every weekend on the w.

mm. ah excuse . ah . this is the w news lied from the sky of relief from germany's political establishment. after the far right fails to win a key state election under america. conservatives, when big fights off the challenge from the populace alternative for germany. the ball was the last real test. before national elections in september, also coming up, judges in the netherlands to begin hearing evidence against for men accused of shooting down malaysia airlines flights. an age 70 families are bracing, harrowing.