No abstract

A direct-speech brain-computer interface (DS-BCI) acquires neural signals corresponding to imagined speech, then processes and decodes these signals to produce a linguistic output in the form of phonemes, words, or sentences. Recent research has shown the potential of neurolinguistics to enhance decoding approaches to imagined speech with the inclusion of semantics and phonology in experimental procedures. As neurolinguistics research findings are beginning to be incorporated within the scope of DS-BCI research, it is our view that a thorough understanding of imagined speech, and its relationship with overt speech, must be considered an integral feature of research in this field. With a focus on imagined speech, we provide a review of the most important neurolinguistics research informing the field of DS-BCI and suggest how this research may be utilized to improve current experimental protocols and decoding techniques. Our review of the literature supports a cross-disciplinary approach to DS-BCI research, in which neurolinguistics concepts and methods are utilized to aid development of a naturalistic mode of communication.

Classification of electroencephalography (EEG) signals corresponding to imagined speech production is important for the development of a direct-speech brain–computer interface (DS-BCI). Deep learning (DL) has been utilized with great success across several domains. However, it remains an open question whether DL methods provide significant advances over traditional machine learning (ML) approaches for classification of imagined speech. Furthermore, hyperparameter (HP) optimization has been neglected in DL-EEG studies, resulting in the significance of its effects remaining uncertain. In this study, we aim to improve classification of imagined speech EEG by employing DL methods while also statistically evaluating the impact of HP optimization on classifier performance. We trained three distinct convolutional neural networks (CNN) on imagined speech EEG using a nested cross-validation approach to HP optimization. Each of the CNNs evaluated was designed specifically for EEG decoding. An imagined speech EEG dataset consisting of both words and vowels facilitated training on both sets independently. CNN results were compared with three benchmark ML methods: Support Vector Machine, Random Forest and regularized Linear Discriminant Analysis. Intra- and inter-subject methods of HP optimization were tested and the effects of HPs statistically analyzed. Accuracies obtained by the CNNs were significantly greater than the benchmark methods when trained on both datasets (words: 24.97%, p < 1 × 10–7, chance: 16.67%; vowels: 30.00%, p < 1 × 10–7, chance: 20%). The effects of varying HP values, and interactions between HPs and the CNNs were both statistically significant. The results of HP optimization demonstrate how critical it is for training CNNs to decode imagined speech.

No abstract

Imagined speech has recently become an important neuro-paradigm in the field of brain-computer interface (BCI) research. Electroencephalogram (EEG) recordings during imagined speech production are difficult to decode accurately, due to factors such as weak neural correlates and spatial specificity, and signal noise during the recording process. In this study, a dataset of imagined speech recordings obtained during production of eleven different units of imagined speech is used to investigate the relative effects of different features on classification accuracy. Three distinct feature-sets are computed from the data: a linear feature-set, a non-linear feature-set, and a feature-set comprised only of mel frequency cepstral coefficients (MFCC). Each featureset is used to train a decision tree classifier and a Support Vector Machine classifier. The results indicate that the use of MFCC features provides greater discrimination of imagined speech EEG recordings in comparison with the other features evaluated, and that phonological differences between imagined words can serve as an aid to classification.