A subject-independent portable emotion recognition system using synchrosqueezing wavelet transform maps of EEG signals and ResNet-18
Key Facts:
The system records brain activity through an EEG cap and uses the AI model DeWave to translate EEG signals into words and sentences.
The technology has demonstrated around 40% translation accuracy on BLEU-1 scale and aims to reach the performance level of traditional language translation programs.
It offers a more adaptable and less invasive alternative to previous technologies, having been tested on 29 participants with diverse EEG patterns.
Source: University of Technology Sydney
The system records brain activity through an EEG cap and uses the AI model DeWave to translate EEG signals into words and sentences.
The technology has demonstrated around 40% translation accuracy on BLEU-1 scale and aims to reach the performance level of traditional language translation programs.
It offers a more adaptable and less invasive alternative to previous technologies, having been tested on 29 participants with diverse EEG patterns.
Source: University of Technology Sydney