Researchers at the University of Technology Sydney’s GrapheneX-UTS Human-centric Artificial Intelligence Centre have achieved a groundbreaking milestone by developing a portable, non-invasive device capable of translating silent thoughts into text. This pioneering invention holds the potential to revolutionize communication for individuals facing challenges such as paralysis, strokes, or other conditions preventing verbal expression. Additionally, the device opens new avenues for seamless communication between humans and machines, enabling the control of robots or bionic arms.
In a recent study, participants wore a cap equipped with an electroencephalogram (EEG) to capture electrical activity in their brains silently reading texts. As soon as the sensors detected thought processes, they translated the brainwaves into random text. The recorded brain waves were then transformed into coherent text using a sophisticated artificial intelligence model called DeWave. This innovative breakthrough provides a unique mind-reading experience and offers hope for those seeking alternative communication methods.
During tests, participants were prompted to think specific phrases silently, resulting in the AI accurately transcribing the thoughts into text. For instance, a participant’s mental prompt, “Good afternoon! I hope you’re doing well. I’ll start with a cappuccino, please, with an extra shot of espresso,” was successfully translated into “Afternoon! You well? Cappuccino, Xtra shot. Espresso.”
CT Lin, Director of the GrapheneX-UTS HAI Centre, led the research and stated, “This research represents a pioneering effort in translating raw EEG waves directly into language, marking a significant breakthrough in the field. It is the first to incorporate discrete encoding techniques in the brain-to-text translation process, introducing an innovative approach to neural decoding. The integration with large language models is also opening new frontiers in neuroscience and AI.”
Although certain statements proved more challenging, the study achieved a 40% success rate among the 29 participants evaluated. The model demonstrated proficiency in matching verbs but faced difficulties with nouns, often producing synonymous pairs. Despite these challenges, the researchers consider the results meaningful and indicative of the device’s potential.
Importantly, the device’s use of EEG signals through a cap, rather than invasive brain implants, makes it non-invasive, cost-effective, and easily transportable. In comparison to other initiatives like Elon Musk’s Neuralink, which recently received approval for human trials, this breakthrough device represents a significant step forward in brain-computer interface technology. Neuralink aims to develop brain implants for paralysis patients, and while it has its own merits, the UTS device stands out for its non-invasiveness, affordability, and portability, making it a promising advancement in communication technology.
