This breakthrough represents a significant advancement in restoring communication for individuals who have lost it. The research achieved a speed three times faster than the previous record and is approaching the natural conversation rate of 160 words per minute. Furthermore, this study relies on evolving technology, as companies compete to develop advanced brain implants and generative artificial intelligence, instilling optimism among the authors for even better outcomes.
Jaimie Henderson, the paper’s senior author and a professor of neurosurgery at Stanford University, expressed his satisfaction with the results, stating, “We had hoped for a result like this, and being able to demonstrate it was extremely gratifying.” He likened the progress in brain implants to the evolution of television, where increasing the number of pixels led to a sharper image. Henderson predicts that devices with more electrodes will provide a higher resolution picture of brain activity.
In recent years, there has been significant momentum in the field of connecting brains to electronic devices, popularized by Elon Musk’s Neuralink. Multiple companies are developing technology to read the brain’s instructions to the body and utilize computer programming to carry them out. Remarkable achievements have already been made, such as enabling a paralyzed man to climb stairs. While brain-computer interfaces, also known as BCIs, are not yet commercially available, they have undergone clinical trials involving over 40 participants.
The focus of the Nature study, Pat Bennett, is currently 68 years old and suffers from amyotrophic lateral sclerosis (ALS), a degenerative disease that can lead to paralysis. Bennett previously worked as a human resources director and enjoyed equestrian activities before her illness progressed.
Although Bennett has limited mobility and finds it challenging to type, the muscles responsible for speech in her mouth and throat no longer function, preventing her from producing understandable sounds.
Bennett communicated her thoughts on her condition, stating, “When you think of ALS, you think of arm and leg impact. But in a group of ALS patients, it begins with speech difficulties. I am unable to speak,” as mentioned in a press release from Stanford.
To convert Bennett’s attempts at speech into text, the research team utilized two small implants containing approximately 120 electrodes, which were inserted into her brain to monitor neural activity. The team trained an algorithm to recognize the words she intended to say by having her attempt to speak sentences on a computer screen for four months. This data was then combined with a language model that predicts words based on context.
Using a vocabulary of 125,000 words, the research team successfully decoded Bennett’s attempted speech at a rate of 62 words per minute, with a 24 percent word-error rate. While this level of accuracy allows for understanding the general idea of a sentence, the authors noted that the error rate is currently too high for everyday use, and the system as a whole is not yet practical for patients.
The authors discovered that the error rate decreased as additional channels, or electrodes reading the brain’s signals, were added. Companies like Neuralink, Paradromics, Precision Neuroscience, and Blackrock Neurotech are actively working on devices with more channels to obtain a higher resolution picture of the brain’s signals.Follow Google News
Denial of responsibility! VigourTimes is an automatic aggregator of Global media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, and all materials to their authors. For any complaint, please reach us at – [email protected]. We will take necessary action within 24 hours.
