Machine Translation Evolution from Natural Translation to GPT Technology

Nowadays, new technologies offer various forms of instant communication, such as emails, instant messaging services, and videoconferencing platforms. However, the different languages around the world remain a significant obstacle to the effective transmission of messages. This challenge has driven the evolution of translation systems, leading to the creation of various translators that aim to meet user needs with increasingly accurate translations. This article provides an overview of machine translation (MT), including speech-to-speech translation (S2ST), exploring its origins and development in the information society, and comparing it with NLP technology like ChatGPT. It also discusses the advantages and potential disadvantages of ChatGPT for the future of foreign language learning and human translators. Key concepts such as natural language processing and automatic speech recognition are examined concerning MT and ChatGPT. Additionally, the article introduces the “Speech Translator,” a mobile application developed in App Inventor 2, used for analysis in a case study. The methodology for this application is developed in four phases: trend analysis, trend characterization, resource construction, and resource evaluation and/or documentation.