Just so that you know that Facebook hasn’t abandoned its brain-typing project, the company published an update consisting of the current situation report
In the publication released by the company, where the company acknowledges the Reality Labs-backed scientists at the University of California, the team behind the ongoing project. Facebook presented the result of a recent test conducted by the UCSF researchers which aim at decoding people’s speech using implanted electrodes. Their work demonstrates a method in which users can quickly read whole words and phrases from the brain. If at the end of the day the experiment turns out to be perfect, then it will help Facebook to get closer to achieving its dream of a noninvasive thought-typing system.
According to a reliable source, a similar system like the Facebook brain-typing project already exists, however, they often ask users to spell out individual words with a virtual keyboard which beats the idea of ‘brain-typing’. On the other hand, Facebook’s experiment saw its subjects listening to multiple-choice questions while speaking the answers aloud. Although, not exactly the result that the company is aiming at, however, it marks a step further for the company.
An instance made in the report released by Project Nature’s Communication suggests that If participant heard someone ask “ Which instrument do you like listening to?” they probably would respond with answers like ‘drum’, and ‘violin’, while their brain activity has been recorded. Now, what the system does is that- it would guess when they are asking a question and when they were answering it, while it also guesses the content of both speeches alongside.
“The predictions will be shaped by prior context- so that once the system determines which question subjects were hearing, it would narrow the set of likely answers. At that, the system could produce results with 61 to 76 percent accuracy, compared with the 7 to 20 percent accuracy expect by chance.” The verge wrote.
Also, the latest experiment has unfolded new findings, and while it basically reveals a lot of differences from the initial experiment carried out by the UCSF back in 2017: a noninvasive, mass-market cap that lets people type more than 100 words per minute without manual text entry or speech-to-text transcription. Facebook also highlights a Reality Labs-backed headset that reads brain activity with near-infrared light, potentially making a noninvasive interface more likely which is by far a better result than the company had in 2017.
While the project still seems to be far from completion, the company hopes to integrate it with other pending projects like that of the virtual and augmented reality glass. “Being able to decode even just a handful of imagined words — like ‘select’ or ‘delete’ — would provide entirely new ways of interacting with today’s VR systems and tomorrow’s AR glasses,” the Reality Labs post reads. And just like I informed earlier, Facebook isn’t the only big company working on brain-computer interfaces: Elon Musk’s Neuralink recently revealed new work on a threadlike brain-reading implant which is kind of similar to that of Facebook brain-typing project.
Having said all of that, the ongoing project still remains teaser until it is finally ready, and that can only be determined by the company, although it might just be one of that incubator projects that may eventually not see the limelight. However, Facebook does not seems to be giving up on the project anytime soon.