Classification of auditory stimuli from EEG signals with a regulated recurrent neural network reservoir

Abstract

The use of electroencephalogram (EEG) as the main input signal in brain-machine interfaces has been widely proposed due to the non-invasive nature of the EEG. Here we are specifically interested in interfaces that extract information from the auditory system and more specifically in the task of classifying heard speech from EEGs. To do so, we propose to limit the preprocessing of the EEGs and use machine learning approaches to automatically extract their meaningful characteristics. More specifically, we use a regulated recurrent neural network (RNN) reservoir, which has been shown to outperform classic machine learning approaches when applied to several different bio-signals, and we compare it with a deep neural network approach. Moreover, we also investigate the classification performance as a function of the number of EEG electrodes. A set of 8 subjects were presented randomly with 3 different auditory stimuli (English vowels a, i and u). We obtained an excellent classification rate of 83.2% with the RNN when considering all 64 electrodes. A rate of 81.7% was achieved with only 10 electrodes.

EEG2Mel: Reconstructing Sound from Brain Responses to Music

“…The raw input representation showed to not give the best results in our deep CNN regressors but it did remain well above the chance performance rate of 10%. Meanwhile, the mel-spectrogram worked better across both the raw and PSD input representations with performance comparable to studies where only EEG classification was conducted [13,17,18]. Classifying outputs to their stimuli classes reveals fidelity of semantically relevant reconstructions.…”

“…The first four minutes of all recordings were used and cut up into five second chunks. To balance the train and test Other EEG studies also keep generalization within participants [5,13], and a recent study has shown that weak correlations across participants in the NMED-T could be why generalizing to unseen participants is difficult without many recorded participants in the dataset [14].…”

P-CRITICAL: a reservoir autoregulation plasticity rule for neuromorphic hardware

“…Many LSM users implement a non-spiking readout layer W O using machine learning methods [13,[21][22][23][24] or even n-layers formal neural networks [25]. In this work, we use a single formal layer with a softmax activation function as the output layer as we focus on the reservoir component of the LSM.…”

“…While W R is not trained, many authors include unsupervised neuroplasticity rules [22,23,[26][27][28] as a way to either keep biological realism or to provide a higher computational performance. Similarly, several studies looked at the initialization of W R in combination with various topologies such as small-world [27,[29][30][31] and scale-free networks [31,32].…”

EEG-based Brain-Computer Interfaces (BCIs): A Survey of Recent Studies on Signal Sensing Technologies and Computational Intelligence Approaches and their Applications

“…Using RNN with LSTM architectures could take temporal dependence into account for EEG timeseries signals and could achieve an average classification accuracy of 93.0% [131]. In the research of applying RNN to auditory stimulus classification, [132] used a regulated RNN reservoir to classify three English vowels a, u and i. Their result showed an average accuracy rate of 83.2% with the RNN approach outperformed a deep neural network method.…”