根据维吾尔语词重音的位置规律进行音节级标注,提取不同的特征参数(能量、基频等),对各个特征参数作单流、分流及特征级融合识别实验,对比分析各特征参数对重音检测识别率的影响。对融合后的高维单流特征采用主成分分析作降维、去冗余处理,并作识别实验。参照识别精确率结合语音语言学知识对实验结果进行分析。 According to the location rules of Uyghur word stress, syllabic dimensioning is performed, different characteristic parameters (energy, fundamental frequency, etc.) are extracted, single flow, shunt and feature level fusion identification experiments are performed on each characteristic parameter, Rate of impact. After the integration of high-dimensional single-stream features using principal component analysis for dimensionality, to redundant processing, and for identification experiments. With reference to the recognition accuracy combined with phonetic linguistic knowledge to analyze the experimental results.