ADM-201 dump PMP dumps pdf SSCP exam materials CBAP exam sample questions

霍金声音背后的秘密 – 译学馆
未登录,请登录后再发表信息
最新评论 (0)
播放视频

霍金声音背后的秘密

Stephen Hawking's Voice and the Machine That Powers It

好了 这是我们最新一期的彭博西部资讯
Well, it is time now for our latest Bloomberg west series of
感官生物工程系列节目
bioengineering the senses.
让我们和医学前沿技术来一次近距离接触
A closer look at the technology on the frontiers of medical science.
今天 我们一起探究物理学家史蒂芬霍金背后的机器
Today, we are looking at the machine behind physicist Stephen Hawking
以及智能技术是如何帮助他交流的
and how smartful technology is helping him communicate.
药物没能治好我
Medicine has not been able to cure me.
所以我依靠技术的帮助来交流和生活
So I rely on technology to help me communicate and live.
这个史蒂芬霍金教授所依赖的重要系统
That vital system that Professor Stephen Hawking relys on
最近完成了升级
has recently done upgrade.
过去需要三到四分钟才能完成的任务
Tasks that used to take 3-4 minutes
比如打开文档 现在只要十秒钟
like opening a word document now take around 10 seconds.
英特尔与霍金合作了数十年
Intel has been working with Hawking for decades.
最近的一次升级已经过去两年了
And the latest upgrade was over two years in the Miega.
我们完全改变了
So we completely change the way that
系统的交互方式 这样就不需要用鼠标进行操作
his interfacing with these aspects the system to have him not to rely on the mouse,
所以不论他正在做什么
so based on whatever he happens to be doing,
我们都能以此预测出他接下来想做的事情中最合理的那一个
we would surface the most logical things that he might want to do.
这也同样适用于霍金的文本语音转换系统
That also applies to Hawking’s text-to-speech system.
你可以从Swiftkey程序开始了解
You can start up with SwiftKey
其软件应用于智能手机上执行输入联想任务
whose software runs predictive task dap on smart phones
这个程序将转换系统的语速提高了一倍
has double talking speech rate
它将霍金的最新著作纳入学习范围
It considers with Hawking’s last writing
检查当中是否有错误
whether there were any errors
然后尝试预测霍金将会用到哪个词
and then tries to predict which character he might choose next.
为了做到这个 Swiftkey程序必须学习
To do that, the SwiftKey program has to learn
霍金是怎样写作和说话的
how Hawking writes and speaks
大量分析他的文献
by analyzing dozens of his documents.
有些是已经出版的 有些没有
Some published, others not.
我们的测试显示
The tests we ran showed that
一段给定的文字由
of a given body of text made up of
一定数量的字符组成
a certain number of characters.
霍金教授实际上只需要输入15%-20%的字符
Professor Hawking would actually only have to enter around 15 to 20 percent of those characters.
剩下的部分都能够被预测出来
And the rest would be inferred.
接下来是疯狂的部分
Now here is the crazy part.
霍金在他的电脑上进行的所有操作
Everything Stephen Hawking does on his computer
都由其面部的一条肌肉触发
is triggered by one muscle in his cheek.
感应中心可以监测到他面部肌肉的提拉动作
And inference center detects when his cheek muscle moves up.
就像我们智能手机能感应到
Just the way our smart phones can sense
它被放在脸边的动作
when it is close to the face.
未来这项技术
Future versions of this technology
将覆盖更大范围的动作
are aiming to take a bigger range of movement into account.
他可以通过动作来说是或者不是
But he can actually to they say yes and no by moving
即通过他的眉毛或者翘起的嘴角
you know, his eyebrows and by pulling his mouth.
我们现在正在研究的一个项目就是
So one of those things we have been working on is,
运用动作捕捉系统 捕捉不同的动作
using essentially a Canada system to detect these different movements.
所以现在我们可以 比如说
So now we can, for example
通过说“不”来取消指令
undo things by saying “No”
不再需要回到屏幕上
rather than having to go a manipulate the whole screen
点击取消按钮了
to get to a back space button, for example.
英特尔甚至在开发脑控制交互界面
Intel is even trying to develop brain control interfaces,
可以帮助那些
which could help people who can
完全不能活动肌肉的人
no longer move any muscles at all.
这项技术一定会应用到霍金身上
That will necessarily happen to Stephen Hawking.
2016年该技术的市场估值高达550亿美元
But with the market forecast to reach 55 billion dollars in 2016,
它代表了电脑系统的下一个发展方向
it represents the next leap for a system of technologies.

发表评论

译制信息
视频概述

讲述机器是如何帮助霍金交流的

听录译者

void

翻译译者

立冬_十年

审核员

审核员 LD

视频来源

https://www.youtube.com/watch?v=OTmPw4iy0hk

相关推荐