语音活动检测对方言辨识系统的影响研究 关键词：方言辨识；语音活动检测；DD+Hang-over算法 [gap=1145]Key words：dialect identification；voice activity detection；DD+Hang-over algoritlun
基于6个网页-相关网页
The methods of selecting model parameters were studied. Dialect identification experiments which build with GMM were conducted.
建立了基于GMM的方言辨识模型,研究了模型参数的选取,并用该模型进行了方言辩识实验,分析了训练模型时所用特征参数对辩识结果的影响。
参考来源 - 基于统计特性的汉语方言辨识方法研究·2,447,543篇论文数据，部分数据来源于NoteExpress
针对方言辨识系统分类决策能力较弱的问题，提出一种基于集成学习的方言辨识方法。
This paper presents a novel Chinese dialect identification method to solve the poor decision ability existed in most dialect identification system.
从而有效的解决了在汉语方言辨识中码本精度不高的问题，使得系统的识别率有了很大提高。
Also the method effectively solved the problem of low precision code and improved system recognition rate.
本文通过对当地方言语音系统的辨识，阐释了永登方言声韵调的基本特征以及与普通话在语音方面的差异。
This paper, by the distinction of the pronunciation system of the local dialect, discusses the primary phoneme features and the differences between the dialect and mandarin Chinese in pronunciation.
应用推荐