摘要
为了解决英语翻译器中语音识别的精度较低的问题,提出一种基于智能算法的英语翻译器语音识别方法。首先采集语音信息,再提取语音信号特征;然后基于谱分解模型对语音信号特征进行预处理;最后设计一种翻译器语音识别方法,实现英语翻译器语音识别方法的设计。仿真测试结果表明,采用设计方法后,英语翻译器语音识别准确率最高为96.13%,识别时间最长仅为1.77 s,说明本方法的准确配准能力较强,识别效率较高,具有一定应用价值。
In order to solve the problem of low accuracy of speech recognition in English translators, an intelligent algorithm based speech recognition method for English translators is proposed. First, the speech information is collected, and then the speech signal features are extracted;Then, speech signal features are preprocessed based on spectral decomposition model;Finally, a translator speech recognition method is designed to realize the design of English translator speech recognition method. The simulation test results show that the highest speech recognition accuracy of the English translator is 96.13%, and the longest recognition time is only 1.77 s after using the design method. This shows that the method in this paper has strong accurate registration ability, high recognition efficiency and certain application value.
作者
温湛靓
WEN Zhanliang(Xi’an Aeronautical Polytechnic Institute,Xi’an 710089,China)
出处
《自动化与仪器仪表》
2022年第12期162-165,共4页
Automation & Instrumentation
基金
西安航空职业技术学院2022年度科研计划项目《数字化转型背景下民航运输与服务人才培养路径研究》(21XHSK-02)。
关键词
智能算法
谱分解模型
端到端建模
英语翻译器
语音识别
频谱特征分解
intelligent algorithm
spectral decomposition model
end-to-end modeling
english translator
speech recognition
spectral feature decomposition
