This mind-reading AI tool turns thoughts into text! Better than Elon Musk's Neuralink?
Researchers at the University of Technology Sydney have unveiled the world's first mind-reading helmet, translating thoughts into text. A breakthrough for communication and human-machine interaction.
We may have read fiction and seen Hollywood movies where some people can read other people's, and pets', thoughts, but now, we have a living example of a tool that actually goes ahead and does it, courtesy artificial intelligence (AI). It is nothing short of a miracle! Researchers at the University of Technology Sydney's GrapheneX-UTS Human-centric Artificial Intelligence Centre have introduced a groundbreaking device, the world's first mind-reading helmet. This portable and non-invasive invention translates silent thoughts into written text, offering a vital means of communication for individuals with conditions hindering speech, such as paralysis or stroke. Additionally, it holds promise for seamless interaction between humans and machines, potentially enhancing control over robots and bionic arms. With AI likely to keep making the kind of progress it is making now, clearly, the sky is the limit as far as this tool is concerned. The participants wore a cap equipped with an electroencephalogram (EEG) to capture brain activity while silently reading texts. The sensors detected thinking moments, recorded brain waves, and utilised an AI model named DeWave to transform these waves into coherent text. A video demonstration showcased a participant thinking a complex message, with the AI accurately rendering the thought into written form.
Encoding Techniques in Brain-To-Text Translation
Director of the GrapheneX-UTS HAI Centre, CT Lin, led the research, highlighting the pioneering nature of the effort. This breakthrough incorporates discrete encoding techniques in brain-to-text translation, pushing the boundaries of neural decoding. The integration with large language models opens new frontiers in neuroscience and AI. You can check out the study here.
Despite challenges, the study achieved a 40% success rate among the 29 participants, with the model demonstrating proficiency in matching verbs. Notably, difficulties arose in precise noun translations, leading to synonyms pairs. However, the model's meaningful results aligning keywords and forming coherent structures indicate its potential. Also read: Utilize the power of generative AI to increase productivity: Know how Google Bard can help
While the EEG signals from the cap result in a noisier output compared to implanted electrodes, the study surpassed previous benchmarks in EEG translation. The researchers emphasised the non-invasive, cost-effective, and easily transportable nature of their technology, distinguishing it from Elon Musk's Neuralink, a brain-chip startup undergoing human trials for paralysis patients.
Musk's Neuralink aims for surgical insertions of brain-computer interface (BCI) implants, initially enabling control of a computer cursor or keyboard through thoughts alone. Musk envisions broader applications, including treating conditions like obesity, autism, depression, and schizophrenia. However, it has raised multiple ethical concerns as it has reportedly led to harm and even deaths of animals in experiments. The helmet is clearly free of those concerns and rising success rate makes it a big rival for Neuralink.
So, how is this different from the brain implants in Neuralink? No bloodletting in it for sure. This is how it is done.
How it is done
1. Participants silently read passages of text while wearing a helmet
2. It recorded electrical brain activity through their scalp
3. This is done using the old fashioned electroencephalogram (EEG).
4. The EEG wave is segmented into distinct units
5. These then capture specific characteristics and patterns from the human brain. This is done by an AI model called DeWave developed by the researchers.
6. What is does is translate EEG signals into words and sentences.
7. However, this is possible only after learning from large quantities of EEG data.
8. In effect, it incorporates discrete encoding techniques in the brain-to-text translation process and introduces an innovative approach to neural decoding.
Distinguished Professor Lin said, “This research represents a pioneering effort in translating raw EEG waves directly into language, marking a significant breakthrough in the field”. He added, "The integration with large language models is also opening new frontiers in neuroscience and AI."