Fig. 4 | Scientific Reports

Fig. 4

From: Speech prediction of a listener via EEG-based classification through subject-independent phase dissimilarity model

Fig. 4

Impacts of the number of cross-subject phase alignments required to train the subject-independent phase dissimilarity model. The average accuracy is shown per subject and segment for different numbers of templates needed to train the subject-independent model. On average, a minimum of nine templates is required to overcome the accuracy of the subject-specific model for each segment.

Back to article page