Researchers have developed a brain-computer interface (BCI) that translates neural signals into audible speech, allowing a paralyzed woman to communicate nearly instantly. This advancement employs AI algorithms that decode spoken thoughts in real-time, significantly improving on previous BCI technology that required completing entire sentences before generating sound. The device was surgically implanted in the woman, Ann, enabling her to convey nuances of natural speech. The innovation represents a major breakthrough toward creating BCIs that can facilitate practical communication for individuals with speech impairments.
Unlike previous efforts, which could produce sounds only after users finished an entire sentence, the current approach can simultaneously detect words and turn them into speech within three seconds.
The findings, published in Nature Neuroscience on 31 March, represent a big step towards BCIs that are of practical use.
BCIs that stream speech in real time are 'the next level' in research because they allow users to convey the tone and emphasis that are characteristic of natural speech.
Researchers personalized the synthetic voice to sound like Ann's own voice from before her injury, by training AI algorithms on recordings from her wedding video.
Collection
[
|
...
]