A breakthrough in neurotechnology: a completely paralyzed woman has been able to “speak” using only her thoughts, thanks to a brain-computer interface (BCI). She was implanted with 253 microelectrodes that read brain signals, while AI converts them into speech at a rate of 90 words per minute—faster than many people can type. The study, published in Nature, shows that the system can even recognize words that were not included in its training data.
The technology recreates the patient’s voice using old recordings, preserving her natural tone rather than producing a synthetic sound. The electrodes capture neural activity in the brain’s speech areas, and machine-learning algorithms decode the signals into text and speech. The system achieves 75% accuracy in word recognition, making meaningful conversations possible.
While the method currently requires invasive surgery, scientists are working on non-invasive BCIs, such as wearable sensors that detect signals through the skin. For example, in 2024, an MIT team introduced a prototype helmet with electrodes capable of decoding simple commands with 60% accuracy. In the future, such technologies could eliminate the need for keyboards and voice commands, allowing communication directly through thoughts.
This breakthrough marks a major step forward for people with paralysis and could lead to broader applications of BCI. Imagine controlling devices, communicating, or even sharing emotions without speaking a single word. However, challenges remain: how can we ensure brain data security and address ethical concerns?
#space #science #educational #technology
