A team at Indiana University Bloomington has developed a proof-of-concept AI system, Brainoware, utilising brain organoids linked to a computer for basic speech recognition. The brain organoids, miniature collections of nerve cells, were trained to recognise a specific voice from a set of audio clips. While the initial accuracy was 30-40%, it improved to 70-80% after two days of adaptive learning without feedback. The approach aims to overcome challenges like high energy consumption and limitations of silicon chips, exploring the potential of biocomputing using living nerve cells. However, researchers acknowledge the need to address limitations, including the short lifespan of organoids.
Credits: The New Scientist
