突破性的研究：大腦信號(hào)被解碼成書面文本

突破性的研究：大腦信號(hào)被解碼成書面文本

Doctors have been able to turn brain signals for speech into written sentences in a research project that aims to transform how patients with severe disabilities communicate.

在一項(xiàng)旨在改變嚴(yán)重殘疾患者交流方式的研究項(xiàng)目中，醫(yī)生們已經(jīng)能夠?qū)⒋竽X語(yǔ)言信號(hào)轉(zhuǎn)化為書面句子。

The breakthrough is reportedly the first to show how the intention to utter specific words can be extracted from brain activity and converted quickly into text.

據(jù)報(bào)道，這一突破首次表明，如何從大腦活動(dòng)中提取出特定詞語(yǔ)的表達(dá)意圖，并將其迅速轉(zhuǎn)化為文本。

According to The Guardian, the brain-reading software only works with certain sentences that it can be trained on, although scientists believe a more powerful system could be created in the future.

據(jù)《衛(wèi)報(bào)》報(bào)道，大腦閱讀軟件只能與特定的句子一起工作，而科學(xué)家們相信未來(lái)可能會(huì)創(chuàng)造出一個(gè)更強(qiáng)大的系統(tǒng)。

Doctors at the University of California hoped to create a product that could enable people living with paralysis to communicate more easily than existing devices, which pick up eye movements and muscle twitches to control a virtual keyboard, the British newspaper reports.

據(jù)英國(guó)報(bào)紙報(bào)道，加利福尼亞大學(xué)的醫(yī)生們，希望研制出一種能使癱瘓患者比現(xiàn)有的設(shè)備更容易交流的產(chǎn)品，這種設(shè)備可以采集眼球運(yùn)動(dòng)和肌肉抽搐來(lái)控制虛擬鍵盤。

“To date, there is no speech prosthetic system that allows users to have interactions on the rapid timescale of a human conversation,” said Edward Chang, a neurosurgeon and lead researcher on the study.

該研究的首席研究員、神經(jīng)外科醫(yī)生Edward Chang說(shuō):“到目前為止，還沒有一種語(yǔ)音假肢系統(tǒng)可以讓使用者在人類交談的快速時(shí)間范圍內(nèi)進(jìn)行互動(dòng)。”

Three epilepsy patients who were about to have neurosurgery for their condition all had a tiny patch of electrodes placed directly on the brain for a week to map the origins of their seizures, the researchers state in their study.

研究人員在他們的研究中稱，三名癲癇患者即將接受神經(jīng)外科手術(shù)，他們都將一小塊電極直接放置在大腦上一周，以記錄癲癇發(fā)作的起因。

While in the hospital, the patients reportedly allowed Chang to record their brain activity while each was asked nine questions and asked to read a list of 24 potential responses. Chang’s team then built computer models that matched the patterns of brain activity to the questions and the answers.

據(jù)報(bào)道，在醫(yī)院里，患者允許張記錄他們的大腦活動(dòng)，同時(shí)每個(gè)人被問(wèn)9個(gè)問(wèn)題，并被要求閱讀24個(gè)可能的回答。Chang的團(tuán)隊(duì)隨后建立了計(jì)算機(jī)模型，使大腦活動(dòng)模式與問(wèn)題和答案相匹配。

“This is the first time this approach has been used to identify spoken words and phrases,” David Moses, a researcher on the team, told The Guardian. “It’s important to keep in mind that we achieved this using a very limited vocabulary, but in future studies, we hope to increase the flexibility as well as the accuracy of what we can translate.”

“這是第一次用這種方法來(lái)識(shí)別口語(yǔ)單詞和短語(yǔ)，”研究小組成員大衛(wèi)·摩西告訴《衛(wèi)報(bào)》。“請(qǐng)注意，這次我們使用的詞匯量非常有限，但在未來(lái)的研究中，我們希望增加翻譯的靈活性和準(zhǔn)確性。”

Their findings were published in the journal Nature and the work itself was funded by Facebook.

他們的研究結(jié)果發(fā)表在《自然》雜志上，這項(xiàng)研究本身是由Facebook資助的。