A software developed in the Netherlands can decipher the sound being spoken to person,and even who is saying them from scan of the listener's brain.
To train the software ,neuroscientist used functional magnetic resonance imaging(fMRI) to track the brain activity of seven people while they listened to three different speakers saying simple vowel sounds.The team found that each speaker and each sound created a distinctive neural fingerprint in a listener's auditory cortex-a brain region that deals with hearing.The fingerprint was used to create rules that could decode future activity and determine both who is being listened to,and what they are sayin.
Elia Formisano of Maastricht University,who led the group,claims that it is a sort of speech-recognition device and is the first with which one can really distinguish two human voices.The team hopes to match recent advances in using fMRI to identify what a person is looking at from his brain activity.It is now working to make the system robust enough to recognize more complex sound without training,and in a noisy environment. This approach could help improve voice recognition software.