英语语音合理性自动化识别模型研究

Research on automatic recognition model of english speech reasonableness

传统英语语音合理性自动化识别模型无法准确有效反映英语语音合理性,结果准确性较低。为此有必要对英语语音合理性自动化识别模型进行研究。规划英语语音合理性自动化识别模型的设计流程,按照规划的流程对采集的语音信号进行加重、分帧、加窗和端点监测操作处理。在处理后的语音信号中提取特征参数,根据提取出的特征参数进行评分计算,实现英语语音合理性自动化识别,完成英语语音合理性自动化识别模型的设计。实验数据表明,设计的英语语音合理性自动化识别模型更能准确反映真实情况,增强25%准确度,并具备极高的有效性。

The automatic recognition model of traditional English phonetic rationality can not accurately and effectively reflect the rationality of English pronunciation,and the accuracy of the result is low.Therefore,it is necessary to study the automatic recognition model of English phonetic rationality.The design flow of the automatic recognition model for the sound rationality of the English is planned,and the speech signals collected are processed in accordance with the planned process.The feature parameters are extracted from the processed speech signals,and the automatic recognition of English speech rationality is realized,and the automatic recognition model of English speech rationality is designed.The experimental data shows that the automatic recognition model of English speech rationality can reflect the real situation more accurately than the traditional automatic recognition model,and enhances the accuracy of 25%,and has high effectiveness.