口语练习中反馈信息智能化系统设计研究

Research on Intelligent System Design of Feedback Information in Spoken Practice

笔者针对口语练习中存在的问题展开分析和研究,提出了口语练习中语音特征提取系统研究。采用MALTAB对语音信号提取进行倒谱分析,提取参考语音和跟读语音的相同类型特征参数进行研究。根据语音特征,将参考语音与跟读语音在相同表现框架下,以波形、图表形式动态呈现,多角度、多层次展现差异,鼓励口语练习者继续改进发音,缩小差异。

In view of the problems existing in oral practice, the author makes an analysis and research, and puts forward a systematic study of speech feature extraction in oral practice. The cepstrum analysis of speech signal extraction is carried out by using MALTAB to extract the same type of feature parameters of reference speech and follow-up speech. According to the phonetic characteristics, the reference speech and the follow-up speech are presented dynamically in the form of waveforms and graphs under the same performance framework, showing differences from different angles and levels, encouraging oral practitioners to continue to improve their pronunciation and narrow the differences.