Abstract

Brain–computer interfaces (BCIs) are systems that transform the brain's electrical activity into commands to control a device. To create a BCI, it is necessary to establish the relationship between a certain stimulus, internal or external, and the brain activity it provokes. A common approach in BCIs is motor imagery, which involves imagining limb movement. Unfortunately, this approach allows few commands. As an alternative, this chapter presents another approach, an internal language-related stimulus known as imagined speech, which is the action of imagining the diction of a word without emitting any sound or articulating any movement. This neuroparadigm is more intuitive, less subjective, and ambiguous, which are very relevant advantages; however, the cost to properly process the brain signal is not trivial. This chapter describes the main components of an EEG-based imagined speech BCI, along with key works, emerging trends, and challenges in this research area. Regarding the challenges, we present four of them in the pursuit of decoding imagined speech. The first challenge involves accurately recognizing isolated words. The second one is the automatic selection of a subset of EEG channels aiming to reduce computational cost and provide evidence of promising locations for studying imagined speech. The third challenge introduces an innovative approach to addressing scenarios where a new word needs to be added to the vocabulary after the computational model has been trained. Lastly, the fourth challenge concerns the online recognition of words from continuous EEG signals. Despite advances in the area, there is still much work to be done. Important initial steps have been taken in terms of the application of novel techniques for preprocessing, artifact removal, feature extraction, and classification which are the stages to be taken to process the collected signal. Additionally, the community has shared datasets and organized evaluation forums to accelerate the search for solutions.

Citations

Authors (5)