STT Technology for Measuring Chinese Pronunciation Accuracy
Mei Rianto Chandra, Clara Herlina Karjo, Yi Ying · 2025
This paper explores the use of speech-to-text (STT) technology to assess the correctness of Chinese pronunciation among learners using the Baidu Translate application. STT technology offers an unbiased and adaptable solution for language learning. The paper examines the accuracy of advanced STT systems in identifying and converting spoken Chinese into written text, incorporating 310 sounds utilized by 31 respondents. It also explores the use of STT technology in educational settings, providing learners with prompt feedback and improving their learning experience. The study discovered that Baidu Translate STT technology significantly improves the effectiveness and reliability of pronunciation evaluations when compared to traditional methods. However, in certain instances, the Baidu Translate application struggles to distinguish homophones. The research highlights the potential of STT technology to revolutionize language learning and open possibilities for wider implementation in language instruction.