Automatic Recognition of English Loanwords in Russian Speech (Using the IT Field as an Example)
Loading...
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Данная работа исследует вопрос автоматического распознавания английских заимствований в русской речи, уделяя особое внимание сфере информационных технологий (IT). В работе анализируются теоретические и практические аспекты автоматического распознавания речи (ASR) и рассматриваются трудности, связанные с английскими заимствованиями и профессиональным жаргоном. Исследование фокусируется на адаптации англицизмов в русском языке и оценивает, насколько эффективно такие системы, как Yandex SpeechKit и Whisper Open AI, справляются с распознаванием этих заимствований. В рамках работы был создан специализированный корпус и разработаны передовые методики обучения, что позволило значительно продвинуться в теме распознавания речи с вкрапленными английскими заимствованиями. Результаты исследования обогащают тематику данной сферы и направлены на развитие взаимодействия человека и машины, особенно в корпоративных и технически насыщенных средах.
This thesis explores the issue of automatic recognition of English borrowings in Russian speech, with a particular focus on the field of information technology (IT). The study analyzes the theoretical and practical aspects of automatic speech recognition (ASR) and examines the challenges related to English Borrowings and professional jargon. The research focuses on the adaptation of anglicisms in the Russian language and evaluates how effectively systems such as Yandex SpeechKit and Whisper Open AI handle the recognition of these borrowings. As part of the work, a specialized corpus was created, and advanced training methodologies were developed, which allowed significant progress in the recognition of speech containing English borrowings. The results of the study enrich the field and aim to enhance human-machine interaction, especially in corporate and technologically intensive environments.
This thesis explores the issue of automatic recognition of English borrowings in Russian speech, with a particular focus on the field of information technology (IT). The study analyzes the theoretical and practical aspects of automatic speech recognition (ASR) and examines the challenges related to English Borrowings and professional jargon. The research focuses on the adaptation of anglicisms in the Russian language and evaluates how effectively systems such as Yandex SpeechKit and Whisper Open AI handle the recognition of these borrowings. As part of the work, a specialized corpus was created, and advanced training methodologies were developed, which allowed significant progress in the recognition of speech containing English borrowings. The results of the study enrich the field and aim to enhance human-machine interaction, especially in corporate and technologically intensive environments.