Simulation Tools and Techniques. 11th International Conference, SIMUtools 2019, Chengdu, China, July 8–10, 2019, Proceedings

Research Article

Analysis of Phonemes and Tones Confusion Rules Obtained by ASR

Download
66 downloads
  • @INPROCEEDINGS{10.1007/978-3-030-32216-8_45,
        author={Gulnur Arkin and Askar Hamdulla},
        title={Analysis of Phonemes and Tones Confusion Rules Obtained by ASR},
        proceedings={Simulation Tools and Techniques. 11th International Conference, SIMUtools 2019, Chengdu, China, July 8--10, 2019, Proceedings},
        proceedings_a={SIMUTOOLS},
        year={2019},
        month={10},
        keywords={Non-native speakers of Chinese Speech recognition Confusion rules Pronunciation evaluation},
        doi={10.1007/978-3-030-32216-8_45}
    }
    
  • Gulnur Arkin
    Askar Hamdulla
    Year: 2019
    Analysis of Phonemes and Tones Confusion Rules Obtained by ASR
    SIMUTOOLS
    Springer
    DOI: 10.1007/978-3-030-32216-8_45
Gulnur Arkin1, Askar Hamdulla1,*
  • 1: Xinjiang University
*Contact email: askar@xju.edu.cn

Abstract

This paper is based on the exploration of the effective method of erroneous phoneme pronunciation of Chinese mandarin learners whose mother tongue is Uyghur and the solution of major problems of language education, concerning the learner’s pronunciation, it uses a different method, namely data-driven approach, and the Automatic Speech Recognition (ASR) is also used to recognize phonemes of the pronunciation of Chinese mandarin learners whose native language is Uyghur. The phoneme sequence is identified and then the standard pronunciation phonemes corresponding to the recognized phonemes are used as the target phonemes to obtain the mapping relation of each target phoneme and recognition phoneme, thus the possible phoneme error categories and possible erroneous rules in Uyghur learners’ pronunciation can be obtained, which may give some help to the Uyghur learners to learn the Chinese auxiliary language system and the corresponding pronunciation evaluation model.