The Real Difference between Speech and Brain-Computer Interfaces: Two Teams of Researchers Revisit a Conversational Technology for 60 to 70 wpm
In separate studies, both published on 23 August in Nature1,2, two teams of researchers describe brain–computer interfaces (BCIs) that translate neural signals into text or words spoken by a synthetic voice. The BCIs are able to read speech at rates of 78 and 62 words per minute. The new technologies are both faster than before, but natural conversation usually lasts around 160 words per minute.
The results of the studies were similar despite the slightly different approaches. The error rate was 9.1 percent for a limited 50-word vocabulary and 20.8 percent for a 125,000-word vocabulary for the study. After four months, the brain signals can be converted to words with a rate of about 18 words per minute. The UC San Francisco and Berkeley algorithm was able to decode at a median rate. The error rate for a 1,024-word vocabulary was 25 percent while the error rate for a 118-word vocabulary was 8.2 percent.
Although a 23 to 25 percent error rate isn’t good enough for everyday use, it’s a significant improvement over existing tech. In a press briefing, Edward Chang, chair of neurological surgery at UCSF and co-author of the UCSF study, noted that the effective rate of communication for existing technology is “laborious” at five to 15wpm when compared to the 150 to 250wpm for natural speech.
Chang said at the briefing that the 60 to 70 wpm mark is a real milestone for the field.
That said, these studies are more proof of concept than a technology that’s ready for prime time. The treatments require long sessions to train, which may be an issue. The researchers from both teams were hopeful that training would be less intensive in the future
The devices must also be tested on many more people to prove their reliability. Judy Illes, a researcher at The University of British Columbia in Canada says that data is only understood in context, regardless of how elegant or technically sophisticated it is. “We have to be careful with over promising wide generalizability to large populations,” she adds. I do not know if we are there yet.
It has to be easy for people to use the tech at home, without requiring caregivers to go through complicated training. Brain implants are also invasive, and in these particular studies, the BCI had to be connected via wires to a device on the outside of the skull that was then attached to a computer. There are also concerns about electrode degradation and the fact that these may not be permanent solutions. To get to consumer use, the tech will have to be rigorously vetted, which can be a lengthy and expensive process.
Brain-reading devices allow paralysed people to think about talking using their thoughts: A case study of an elderly woman whose brain was saved from speaking after a brain stem stroke
Chang said that the potential benefit of this tech is tremendous, if it can be safely and widely implemented. “We are thinking about that quite seriously and what the next steps are.”
Christian Herff believes these devices could be products in the near future.
Bennett told reporters that those who are nonverbal can still remain connected to the bigger world, while maintaining friends and family.
In a separate study2, Edward Chang, a neurosurgeon at the University of California, San Francisco, and his colleagues worked with a 47-year-old woman named Ann, who lost her ability to speak after a brainstem stroke 18 years ago.
Although the implants used by Willett’s team, which capture neural activity more precisely, outperformed this on larger vocabularies, it is “nice to see that with ECoG, it’s possible to achieve low word-error rate”, says Blaise Yvert, a neurotechnology researcher at the Grenoble Institute of Neuroscience in France.
Chang and his team also created customized algorithms to convert Ann’s brain signals into a synthetic voice and an animated avatar that mimics facial expressions. They used recordings from her wedding video to personalize the voice to sound like Ann after her injury.
The researchers were given feedback by Ann after she gave a speech about hearing a voice similar to her own. It was important for me to speak for myself.
And the participants of both studies still have the ability to engage their facial muscles when thinking about speaking and their speech-related brain regions are intact, says Herff. For every patient, that will not be the case.
Source: Brain-reading devices allow paralysed people to talk using their thoughts
Can Tropical Winds Survive a Warming Temperature That Could Break Down Tropical Fores? A Theoretical Analysis of Wind Tunnel Experiments
“We see this as a proof of concept and just providing motivation for industry people in this space to translate it into a product somebody can actually use,” says Willett.
The wind-tunnel experiments could help athletes run the fastest marathon ever, and an analysis could explain why birds are the colors they are.
As the climate warms, tropical forests around the world are facing increasing temperatures. But it’s unknown how much the trees can endure before their leaves start to die. A team of scientists have gathered multiple data sources together to try and answer this question, and think that a Warming of 3.9 C would lead to many leaves reaching a tipping point which would break down. This scenario would likely cause significant damage to these ecosystems’ role in vital carbon storage and as homes to significant biodiversity.