Scientists have taken a step forward in their ability to decode what a person is saying just by looking at their brainwaves when they speak.
科學家們在解碼人講話時的腦電波活動的能力上又前進了一步。
They trained algorithms to transfer the brain patterns into sentences in real-time and with word error rates as low as 3%.
他們開發算法將大腦活動實時轉換成句子，錯誤率低至3%。
Previously, these so-called "brain-machine interfaces" have had limited success in decoding neural activity.
之前這些所謂的“腦機接口”在解碼神經活動方面取得的成功非常有限。
The study is published in the journal Nature Neuroscience.
這項研究發表在《自然神經科學》雜誌上。
The earlier efforts in this area were only able decode fragments of spoken words or a small percentage of the words contained in particular phrases.
早期在這方面的努力只限於解碼口語單詞的片段或包含在特定短語中的一小部分單詞。
Machine learning specialist Dr Joseph Makin from the University of California, San Francisco (UCSF), US, and colleagues tried to improve the accuracy.
美國加州大學舊金山分校的機器學習專家Joseph Makin博士和他的同事們試圖提高準確率。
Four volunteers read sentences aloud while electrodes recorded their brain activity.
實驗中，四個志願者大聲朗讀句子，同時用電極記錄他們的大腦活動。
The brain activity was fed into a computing system, which created a representation of regularly occurring features in that data.
然後把大腦活動輸入到一個計算系統，系統生成的數據中包含了經常出現的腦部活動特徵。
Limited language
有限的語言
These patterns are likely to be related to repeated features of speech such as vowels, consonants or commands to parts of the mouth.
這些模式可能與重複的語音特徵有關，如元音、輔音或嘴部活動。
Another part of the system decoded this representation word-by-word to form sentences.
系統的另一部分逐字逐句地將這種表示法解碼成句子。
However, the authors freely admit the study's caveats. For example, the speech to be decoded was limited to 30-50 sentences.
然而，作者坦率地承認了這項研究需要注意的地方。例如，要解碼的語音應限制在30-50個句子之內。
Although we should like the decoder to learn and exploit the regularities of the language, it remains to show how many data would be required to expand from our tiny languages to a more general form of English, the researchers wrote in their Nature Neuroscience paper.
研究人員在發表在《自然神經科學》雜誌上的論文中寫道，儘管我們希望解碼器學習並利用語言的規律，但我們還要知道需要多少數據才能把我們的微小語言擴展到更普遍的英語。
But they add that the decoder is not simply classifying sentences based on their structure.
但他們補充說，解碼器並不是簡單地根據句子結構對句子進行分類。
They know this because its performance was improved by adding sentences to the training set that were not used in the tests themselves.
科學家們知道這一點，因為通過在訓練組中添加測試中沒有用到的句子，解碼器的結果得到了改善。
The scientists say this proves that the machine interface is identifying single words, not just sentences. In principle, this means it could be possible to decode sentences never encountered in a training set.
科學家們說，這證明了機器接口識別的是單個單詞，而不僅僅是句子。原則上，這意味著有可能解碼在訓練集中從未遇到過的句子。
閱讀更多 雄鷹戰力 的文章
