Voice Interface of the Applicant's Assistant Based on Neural Network Methods of Multilingual Speech Thinking

The thesis is devoted to the study of the problems faced by foreign applicants when applying to Russian universities, with an emphasis on the HSE. The paper analyzes the difficulties in interacting with the Admissions Committee and information resources, and also examines approaches to creating voice assistants based on neural networks. The study showed the disadvantages of existing solutions, such as the lack of multilingualism and the ability to record voice messages. As a result of the work, a working model of a Telegram chatbot with language selection and speech transcription functions based on neural network models was developed. In conclusion, the potential of the presented chatbot for improving the adaptation of foreign applicants in Russia and making a scientific contribution to the development of AI assistants is emphasized. The work represents a significant contribution to the field of education and technology, providing new opportunities for effective interaction with international students and improving the admission process to higher education institutions.