Researchers at Stanford University have developed a brain-computer interface (BCI) that can translate neural activity into speech, allowing Paralyzed Individuals to express themselves through a computer screen. The BCI uses tiny brain-insertable sensors connected to a computer algorithm to capture electrical signals from the cerebral cortex and predict intended speech. The system consists of a neural network that decodes phonemes in real-time and a language model that converts those phonemes into words based on statistics in the English language.

During the study, a patient with ALS underwent numerous training sessions to practice speaking sentences, which were then translated by the BCI. With regular sessions over a period of six months, the patient achieved a translation rate of 62 words per minute, faster than existing machine-based speech technology. The vocabulary for the model initially consisted of 50 words and expanded to 125,000 words with a small error rate.

While the BCI is not yet ready for everyday use, it represents a significant step in improving communication speed and enhancing the quality of life for speech-disabled individuals. The ability to communicate more naturally and at a comparable rate to typical conversation can help individuals regain their independence, maintain relationships, and meet their basic needs.

Source: Nature (no URL provided)

The post New Brain-Computer Interface Translates Thoughts into Words for Paralyzed Individuals appeared first on TS2 SPACE.