Key points from article :
Neuroengineers have created a new system that can translate simple thoughts into recognizable speech, using artificial intelligence and a speech synthesizer.
Researchers examined the brain activity of epilepsy patients who were having electrode implants in their brains to monitor seizures and gather data for their research.
They trained a neural network to recognize patterns in the signals and translate them into robotic-sounding words using a vocoder.
They compared the reconstruction accuracy from low and high neural frequency ranges.
Speech synthesizer from all neural frequencies achieved the highest subjective and objective scores on a digit recognition task.
Results demonstrated the efficacy of deep learning for designing the next generation of speech BCI systems.
Research by Northwell Health Physician Partners Neuroscience Institute published in Scientific Reports.
