科学美国人60秒 SSS 蝙蝠的叫声传递了感情

When we humans talk to other humans, the soundswe make all have very specific meanings. "When Isay apple you immediately imagine something thathas the characteristics of an apple." Yossi Yovel, aneuroecologist at Tel Aviv University in Israel. "Andthe question is, do animals also have something like that?"

人类互相交谈的时候，我们所发出的声音都具有特定的含义。“当我说苹果的时候，你马上就会想到苹果的特点。”尤西·约维尔是以色列特拉维夫大学的神经生态学家。“问题是，动物有这样的能力吗?”

Yovel and his team chose to listen in on bats, which do a lot of vocalizing. In fact, in caves withvast numbers of bats, it's total cacophony 1. [bat cave] "It sounds like a crowd in a footballstadium before the match has begun, or something like that." To simplify the problem, theresearchers eavesdropped 2 on a much smaller colony—just 22 Egyptian fruit bats.

约维尔和团队选择了听蝙蝠的声音，因为蝙蝠会制造出很多声音。实际上，在存在大量蝙蝠的洞穴里，声音总是那么不和谐。“那种声音听起来就像足球比赛开始前聚集在球场里的人群所发出的声音，或者其他类似的情况。”为了简化问题，研究人员对一个较小的蝙蝠群进行了监听，这个群体只有22只埃及蝙蝠。

Over several months, they recorded tens of thousands of calls, [call] along with synced-upvideo—which allowed them to decipher the speaker, the intended recipient 3, the situation, andthe behavior resulting from each call. They then fed their huge database of calls to computers, to test whether machine learning could help make sense of them, using algorithms like theones used for human speech recognition.

数个月以来，他们记录了数十万种蝙蝠的声音并录制了同步视频，这让研究者可以破译出发出声音的蝙蝠、接收声音的蝙蝠、当时的情景以及每个叫声所引发的行为。之后，研究人员将这个庞大的叫声数据库输入电脑，来测试机器学习是否可以理解这些叫声，是否可以利用分析人类言语认知的算法去理解蝙蝠叫声的含义。

Turns out, the algorithms could correctly identify which bat made each call, more often thanchance would predict. "And I can say to some extent who is this bat shouting at, so who isthe addressee of this vocalization?" They could even figure out what a bat might be angryabout, like "hey, stop sniffing 4 me, or this is my food" and how the addressee might respond—meaning there's really quite a lot of information embedded 5 in bat vocalizations. Pretty useful, if you live in the dark. The study is in the journal Scientific Reports.

结果表明，这些算法可以正确识别出哪只蝙蝠发出了叫声，准确率高于预测。“在某些程度上，我可以判断出这只蝙蝠在冲哪只蝙蝠叫喊，即这一声音的受体是谁。”他们甚至发现了蝙蝠会对哪些事感到生气，比如“喂，别再闻我了，这是我的食物”，他们还了解了声音的接受者可能会做出哪种反应，这表明蝙蝠的叫声中包含了大量的信息。如果你生活在黑暗中，这非常有用。该研究结果发表在《科学报告》期刊上。

As for whether we might someday have Google Translate for bat calls? "Well, Google, definitelynot... but perhaps an iPhone." But seriously, "For sure in the next 100 years we will always bebehind, we will never really be able to generate a bat to human dictionary. But definitely we willadvance in that direction." And in doing so—it might shed a little light on animalcommunication in general. And how our own language evolved.

那未来某一天我们是否可以用谷歌翻译蝙蝠的叫声呢?“谷歌，当然不行，不过也许苹果手机可以。”但是认真而言，“毫无疑问，在未来的100年，我们将一直处于落后地位，因为我们永远不能创造出蝙蝠人类词典。但是，我们肯定将向那个方向前进。”这样我们也许可以对动物之间的交流有个大致了解。同时也可以了解我们的语言是如何进化的。