Can we learn to talk to sperm whales? | David Gruber | TED

翻译人员: Qingyue Sun 校对人员: Helen Chang

You are about to hear the sounds of the largest-toothed predator

你即将听到

地球上最大的有齿肉食动物的声音。

on the planet:

这种比校车还大的动物，

an animal bigger than a school bus

with perhaps the most sophisticated form of communication

也许拥有现有的

that has ever existed.

最复杂的交流方式。

(Video: whale clicking)

（视频：鲸鱼发出的滴答声）

These are the sounds of the mighty sperm whale,

这是健壮的抹香鲸的声音。

a fellow mammal that can dive almost a mile,

它是一种能下潜 将近一英里的哺乳动物，

hold its breath for more than an hour

能屏住呼吸超过一小时，

and lives in these amazingly complex, matriarchal societies.

并且住在奇迹般复杂的母系社会里。

These clicks you heard,

你所听到的这些滴答声

called codas,

被称作尾波，

are just a facet of what we know of their communication.

只是我们所知的 它们交流方式的一部分。

We know these animals are communicating,

我们知道这些动物在交流，

we just don't yet know what they're saying.

但只是还不知道它们在说什么。

Project CETI aims to find out.

CETI项目希望发现这个奥秘。

Over the next five years,

在未来的五年里，

our team of AI specialists,

我们团队的AI专家们，

roboticists, linguists

机器专家们，语言学家们，

和海洋生物学家们，

and marine biologists

aim to use the most cutting-edge technologies

希望用最先进的技术

to make contact with another species,

来和其他物交流，

and hopefully communicate back.

希望得到回应。

We believe that by listening deeply to nature,

我们相信，通过用心地聆听自然，

we can change our perspective of ourselves

我们能改变我们对自身的看法，

and reshape our relationship with all life on this planet.

并重塑我们与地球上其他生物的关系。

This of course seems like an impossible goal.

这当然听起来是个不可能实现的目标。

People have been trying to make contact with other animals

数百年里，人们尝试着 和其他动物沟通。

for hundreds of years.

How could we do what others could not,

我们如何做到别人做不到的，

especially given that I'm sitting here on my couch in New York City

特别是当我现在坐在 纽约市里我的沙发上，

in the middle of a pandemic and protests?

在疫情和抗议之中？

I've spent the last 20 years as a marine biologist and oceanographer,

二十年来我一直是个 海洋生物学家和海洋摄影师，

studying the ocean from all different perspectives,

从不同角度学习着海洋，

from microbes to sharks.

包括从微生物到鲨鱼的角度。

I've assembled interdisciplinary teams

我组织过跨学科的团队，

that have built the first shark-eye camera

建造了第一个鲨鱼眼摄像机

to see the world from a shark's perspective,

来通过鲨鱼的视角看世界，

and have collaborated with engineers

并且和工程师们合作过

to design robots so gentle that they don't even stress a jellyfish.

来设计轻柔到不会让水母紧张的机器。

But it wasn't until 2018

但直到2018年，当我在

when I was on fellowship

拉德克利夫高等研究所做研究员时，

at the Radcliffe Institute for Advanced Study

that I realized that perhaps the best way to understand the ocean

我才意识到也许理解海洋

and its inhabitants

和它的居民的最好方式

wasn't just by seeing the world through their eyes,

不只是通过它们的视角看世界，

but by listening --

而是通过倾听——

by really, deeply listening.

通过真正、用心的聆听。

I became interested in sperm whales when I heard their sounds.

当我听到抹香鲸的声音时 我开始对它们感兴趣。

They sounded like they were coming from another universe;

它们听起来像来自另一个宇宙，

a siren song being broadcast from the darkest reaches of the sea.

是海洋至黑处传来的海妖的歌声。

These weren't the typical harmonious whale songs

这些不是我所习惯的，

that I had been accustomed to.

鲸鱼典型的和谐歌声。

These sounded more like digital data transfer.

这些听起来更像电子数据传播声。

We assembled the future Project CETI team

我们组建了现在的CETI项目团队，

and began discussing how to use the most advanced technologies

并开始讨论如何使用最先进的技术

to communicate with whales.

来与鲸鱼沟通。

One of the principal conclusions

主要结论之一是

was that machine learning had a really good chance

机器学习有很好的机会

of understanding the patterns of sperm whale communication.

去理解抹香鲸交流的规律。

And the time to apply these technologies was now.

现在是应用这些技术的时候了。

Cracking the interspecies communication code

破解物种间的通讯密码

didn't just seem possible,

不仅看起来有可能实现，

it almost seemed inevitable.

并且几乎是不可避免的。

But how can analyzing patterns help us converse with whales

但是分析规律如何帮助我们与鲸鱼

and other animals?

和其他动物沟通呢？

Well, step one is to understand the elements of sperm whale communication.

第一步是了解抹香鲸交流的要素。

你们听到的这些尾波 不是我们所认为的句子，

These codas you heard don't appear to be sentences as we know them,

but there's clear structure in how these animals communicate.

但这些动物的交流方式有明确的结构。

Sperm whales send codas back and forth to each other

抹香鲸有序地互相发送

尾波，

in sequences,

and there are regional dialects like British and Australian accents.

还有地域方言， 就像英国和澳大利亚的口音。

This is exactly why machine learning is such a powerful tool.

这就是为什么机器学习 是如此强大的工具。

These approaches analyze patterns in relationship and map meaning to them.

这些方法分析规律间的关联 并得出意义。

Just a few years ago, scientists used machine learning

就在几年前，科学家们通过机器学习

to translate between two totally unknown human languages.

翻译了两种完全未知的人类语言。

Not by using a Rosetta Stone or a dictionary,

不是用罗塞塔石碑或字典，

but by mapping them on patterns in higher-dimensional space.

而是将它们映射到高维空间的规律中。

But for machine learning to work effectively,

但是为了让机器学习有效，

it needs data --

它需要数据——

it needs lots and lots of data.

它需要非常、非常多的数据。

In the past half-century,

在过去的半个世纪里，

marine researchers have painstakingly collected

海洋研究人员煞费苦心地收集

and hand annotated just a few thousand sperm whale vocalizations,

并手工注释了几千种抹香鲸的叫声，

but in order to learn sperm whale communication,

但是为了学习抹香鲸的交流，

we'll need to collect millions,

我们需要收集数以百万计、

if not tens of millions

甚至数以千万计

of carefully annotated sperm whale vocalizations

仔细注释的抹香鲸的声音

并和其行为相对应。

correlated with behaviors.

We'll do it with noninvasive, autonomous, free-swimming robots,

我们将使用非侵入式、 自主、自由游泳的机器人、

aerial-aquatic drones,

航空-潜水式无人机、

bottom-mounted hydrophone arrays

海底水听器阵列

and more.

等等。

We'll work with our close partners at the Dominica Sperm Whale Project

我们将与多米尼加 抹香鲸项目的亲密伙伴合作

to cover a 20-square-kilometer area

来覆盖20平方公里，

that is frequented by over 25 well-known families of sperm whales.

那里有超过25个著名的 抹香鲸家族经常出没。

We're going to put specific focus on the interactions of mothers and calfs,

我们将把重点放在 母亲和幼崽之间的互动上，

training our machine learning algorithms

来训练我们机器学习算法

to learn whale language from the bottom up.

从下往上学习鲸鱼的语言。

All this data we'll have sent through a pipeline

所有这些数据都是通过管道传输的，

and analyzed by the Project CETI translation team.

并由CETI项目翻译团队进行分析。

The raw audio and context data will go through our machine learning engine

原始音频和环境数据 将通过我们的机器学习引擎，

where it's going to be first sorted by structure.

在那里它首先要被按结构排序。

The linguistics team will then search for things like syntax

然后语言学团队 将搜索诸如语法之类的东西

and time displacement.

和时间位移。

For example,

例如，

if we find an event where a whale was talking about something yesterday,

如果我们发现一条鲸鱼 在讨论着昨天发生的事情，

that alone would be a major finding,

光是这个就能成为一个重大发现了，

something that has thus far only been shown in humans.

那是目前为止只在人类中发生的事情。

And once we've really mastered listening,

一旦我们真正掌握了聆听，

we're going to try really carefully to talk back

我们会尝试很小心地回应，

even on the most simplistic level.

即使在最简单的层面上。

Finally, Project CETI will build an open-source platform

最后，CETI项目 将构建一个开源平台，

where we will make our data sets available to the public,

我们将在那向公众提供我们的数据集，

encouraging the global community

鼓励全球社会

to come along on this journey for understanding.

都参与进这段理解的旅程中。

These animals could be the most intelligent beings on this planet.

这些动物可能是 这个星球上最聪明的生物。

They have a neocortex and spindle cells --

它们有一个新皮层和梭形细胞——

structure that in humans control our higher thoughts,

那是在人类身上控制高级思想的结构，

emotions, memory, language and love.

包括情感、记忆、语言和爱。

And all the platforms that we develop can be cross-applied to other animals:

我们开发的所有平台 都可以交叉应用于其他动物：

to elephants, birds,

在大象、鸟类、

primates, dolphins --

灵长类、海豚——

essentially any animal.

几乎所有动物身上。

In the late 1960s,

在20世纪60年代末，

我们的团队成员 罗杰·佩恩发现鲸鱼会唱歌。

our team member, Roger Payne, discovered that whales sing.

（录音：鲸鱼歌唱）

(Recording: whale singing)

A finding that sparked the Save the Whales movement

这一发现引发了拯救鲸鱼运动，

led to the end of large-scale whaling

终结了大规模的捕鲸，

and prevented several whale species from extinction

并阻止了一些鲸鱼物种的灭绝。

just by showing that whales sing.

这光是展示鲸鱼能唱歌就能做到的。

Imagine if we could understand what they're saying.

想象一下如果我们能理解 它们在说什么。

Now is the time to open this larger dialogue.

现在是开启这一更大对话的时候了。

Now is the time to listen deeply

现在是时候用心聆听

and show these magical animals,

来向这些神奇的动物，

and each other,

和彼此，

newfound respect.

展示新建立的尊重。

Thank you.

谢谢。