Inner Speech Classification using EEG Signals: A Deep Learning Approach

Bram van den Berg; Sander  van Donkelaar; Maryam Alimardani

Inner Speech Classification using EEG Signals: A Deep Learning Approach

Bram van den Berg^*, Sander van Donkelaar, Maryam Alimardani

^*Corresponding author for this work

Research output: Contribution to conference › Paper › Scientific › peer-review

Abstract

Brain computer interfaces (BCIs) provide a direct communication pathway between humans and computers. There are three major BCI paradigms that are commonly employed: motor-imagery (MI), event-related potential (ERP), and steady-state visually evoked potential (SSVEP). In our study, we sought to expand this by focusing on “Inner Speech” paradigm using EEG signals. Inner Speech refers to the internalized process of imagining one’s own “voice”. Using a 2D Convolutional Neural Network (CNN) based on the EEGNet architecture, we classified the EEG signals from eight subjects when they internally thought about four different words. Our results showed an average accuracy of 29.7% for word recognition, which is slightly above chance. We discuss the limitations and provide suggestions for future research.

Original language	English
Publication status	Published - 2021
Event	2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS) - Magdeburg, Germany Duration: 8 Sept 2021 → 10 Sept 2021 https://www.ichms2021.de/

Conference

Conference	2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS)
Country/Territory	Germany
City	Magdeburg
Period	8/09/21 → 10/09/21
Internet address	https://www.ichms2021.de/

Cite this

@conference{d056bab774064e2988745e1cee3951c5,

title = "Inner Speech Classification using EEG Signals: A Deep Learning Approach",

abstract = "Brain computer interfaces (BCIs) provide a direct communication pathway between humans and computers. There are three major BCI paradigms that are commonly employed: motor-imagery (MI), event-related potential (ERP), and steady-state visually evoked potential (SSVEP). In our study, we sought to expand this by focusing on “Inner Speech” paradigm using EEG signals. Inner Speech refers to the internalized process of imagining one{\textquoteright}s own “voice”. Using a 2D Convolutional Neural Network (CNN) based on the EEGNet architecture, we classified the EEG signals from eight subjects when they internally thought about four different words. Our results showed an average accuracy of 29.7% for word recognition, which is slightly above chance. We discuss the limitations and provide suggestions for future research.",

author = "{van den Berg}, Bram and {van Donkelaar}, Sander and Maryam Alimardani",

year = "2021",

language = "English",

note = "2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS) ; Conference date: 08-09-2021 Through 10-09-2021",

url = "https://www.ichms2021.de/",

}

TY - CONF

T1 - Inner Speech Classification using EEG Signals

T2 - 2021 IEEE 2nd International Conference on Human-Machine Systems (ICHMS)

AU - van den Berg, Bram

AU - van Donkelaar, Sander

AU - Alimardani, Maryam

PY - 2021

Y1 - 2021

N2 - Brain computer interfaces (BCIs) provide a direct communication pathway between humans and computers. There are three major BCI paradigms that are commonly employed: motor-imagery (MI), event-related potential (ERP), and steady-state visually evoked potential (SSVEP). In our study, we sought to expand this by focusing on “Inner Speech” paradigm using EEG signals. Inner Speech refers to the internalized process of imagining one’s own “voice”. Using a 2D Convolutional Neural Network (CNN) based on the EEGNet architecture, we classified the EEG signals from eight subjects when they internally thought about four different words. Our results showed an average accuracy of 29.7% for word recognition, which is slightly above chance. We discuss the limitations and provide suggestions for future research.

AB - Brain computer interfaces (BCIs) provide a direct communication pathway between humans and computers. There are three major BCI paradigms that are commonly employed: motor-imagery (MI), event-related potential (ERP), and steady-state visually evoked potential (SSVEP). In our study, we sought to expand this by focusing on “Inner Speech” paradigm using EEG signals. Inner Speech refers to the internalized process of imagining one’s own “voice”. Using a 2D Convolutional Neural Network (CNN) based on the EEGNet architecture, we classified the EEG signals from eight subjects when they internally thought about four different words. Our results showed an average accuracy of 29.7% for word recognition, which is slightly above chance. We discuss the limitations and provide suggestions for future research.

M3 - Paper

Y2 - 8 September 2021 through 10 September 2021

ER -

Inner Speech Classification using EEG Signals: A Deep Learning Approach

Abstract

Conference

Fingerprint

Cite this