Silent EEG-Speech Recognition Using Convolutional and Recurrent Neural Network with 85% Accuracy of 9 Words Classification.

In this work, we focus on silent speech recognition in electroencephalography (EEG) data of healthy individuals to advance brain-computer interface (BCI) development to include people with neurodegeneration and movement and communication difficulties in society. Our dataset was recorded from 270 healthy subjects during silent speech of eight different Russia words (commands): 'forward', 'backward', 'up', 'down', 'help', 'take', 'stop', and 'release', and one pseudoword. We began by demonstrating that silent word distributions can be very close statistically and that there are words describing directed movements that share similar patterns of brain activity. However, after training one individual, we achieved 85% accuracy performing 9 words (including pseudoword) classification and 88% accuracy on binary classification on average. We show that a smaller dataset collected on one participant allows for building a more accurate classifier for a given subject than a larger dataset collected on a group of people. At the same time, we show that the learning outcomes on a limited sample of EEG-data are transferable to the general population. Thus, we demonstrate the possibility of using selected command-words to create an EEG-based input device for people on whom the neural network classifier has not been trained, which is particularly important for people with disabilities.

A pilot study of language and culture mediation in medical interpreting at border crossing points in Moscow, Russia.

The paper analyzes provision of medical interpreting services at border crossing points in the Russian Federation. It is argued in the article that medical interpreting at border crossing points should be viewed as linguacultural mediation in the context close to emergency situations, which requires specific interpreter's competences. The article aims at indenting relevant competences of medical interpreters at border crossing points through conducting an experiment on quality of medical services provision to migrants and refugees in terms of effective interpreter assisted doctor-patient communication. The research methodology rests on cluster, factor and discriminant analysis and integrated two stages: desk and field stages. To conduct the survey part an open-ended questionnaire has been developed which included 7 items. The survey was anonymous and involved native speakers of Arabic, Pashto, Dari, Uzbek, Tajik languages. The conducted research made it possible to identify factors that negatively influenced ultimate assessment of medical services provision at border crossing points. These factors are two-fold: the fists category relates to organizational issues such as lack of specific medicines and doctors with narrow specialization, stressful atmosphere and lengthy border crossing procedure; second category includes linguistic factor such as interpreter's knowledge of communicants' cultural background, ability to act as mediator, goal oriented communicative skills. The research results reveal that medical interpreting at border crossing points should be regarded as intercultural mediation as takes place in unstandardized setting and involves a range of culturally and socially significant issues. As the research outcomes the preliminary list of medical interpreter's competences has been drawn.