Binary decisions and emotional EEG based virtual assistant for speech impairments

Abstract

The potential for brain computer intervention to provide effective and exciting technology for persons with impairments is growing because of the constant advancements in information technology. To help those who have difficulty communicating in a high school environment, our research will develop a wearable assistive device called SPEAK. It is a compact, portable equipment that is lightweight and simple to use. It comes with the Muse 2 headband and a deep learning based assistance program that operates on a mobile device. Sometimes people are unable to communicate vocally for a variety of reasons, yet their minds (brains) nonetheless speak normally. This approach would be really helpful and perfectly acceptable to help such mute individuals who are unable to communicate. This research puts forth the effort to detect and collect the frequency ranges of two significant analytical components that have been particularly chosen, such as binary decision yes/no analysis and emotional analysis. The examined frequencies are compared to the freshly emerging brain signal excitements in order to recognize and comprehend the phrases in accordance with the relevant brain signal. The virtual assistant produces voice for the muted person whenever a word is recognized from a brain signal and is transformed into speech sound.