Article

DEVELOPMENT OF A SYSTEM FOR REAL-TIME TRANSLATION AND VOICE SYNTHESIS OF ENGLISH AUDIOVISUAL CONTENT INTO UZBEKISTAN BASED ON ARTIFICIAL INTELLIGENCE

Omadjon UrishevPhD, Fergana State Technical University Fergana, UzbekistanMurodjon Mamurovich AkhmedovStudent, Fergana State Technical University Fergana, Uzbekistan

Zenodo (CERN European Organization for Nuclear Research)repository2026en

ABI

Abstract

This paper describes the creation of a real, time AI, based system for converting English audiovisual content into the Uzbek language with voice synthesis that is synchronized at the same time. The system combines Automatic Speech Recognition (ASR), Neural Machine Translation (NMT), and Text, to, Speech (TTS) technologies. OpenAI Whisper, Google Translate API, and Tacotron2 were used to models to get the best output both in terms of accuracy and the naturalness of the voice. The system proposed gives an opportunity to the user to hear English video content in the Uzbek language with synchronized speech. It is a very effective solution for content localization, education, and media applications.

Topics

Speech Recognition and Synthesis Speech and Audio Processing Educational Technology and Pedagogy

Identifiers

DOI: 10.5281/zenodo.18810332

Citations and references

Cited by 00 references

Metrics — AkademScholar · Coming soon