【摘要】：人的語(yǔ)音聲調(diào)的準(zhǔn)確識(shí)別,可以提高語(yǔ)音信號(hào)處理效果,保證人機(jī)通信的順利進(jìn)行。聲調(diào)識(shí)別時(shí),需要獲取不同的聲調(diào)模式,將待識(shí)別的聲道進(jìn)行比對(duì),而傳統(tǒng)的基于RNN-RBM語(yǔ)言模型的識(shí)別方法只能獲取語(yǔ)音音素、單詞以及語(yǔ)句,不能獲取其對(duì)應(yīng)的標(biāo)準(zhǔn)聲調(diào)模式,無(wú)法完成比對(duì),降低了識(shí)別的精度。提出基于K-means初始化EM算法的語(yǔ)音聲調(diào)識(shí)別方法。通過(guò)建立聲調(diào)信息高斯混合模型,準(zhǔn)確的對(duì)基頻信息概率密度函數(shù)進(jìn)行擬合,采用最大化(EM)算法提取基頻特征參數(shù),并以此為基礎(chǔ)獲取更多的聲調(diào)模式,利用K-means初始化EM算法,消除EM算法對(duì)初始值選取較為敏感的問(wèn)題,再對(duì)高斯混合模型階數(shù)進(jìn)行預(yù)測(cè),提高EM算法執(zhí)行聲調(diào)識(shí)別精度。仿真結(jié)果表明,采用改進(jìn)的聲調(diào)識(shí)別方法進(jìn)行聲調(diào)識(shí)別,識(shí)別準(zhǔn)確率較高,具有一定的實(shí)用性。
[Abstract]:The accurate recognition of human voice tone can improve the effect of speech signal processing and guarantee the smooth progress of communication. In tone recognition, different tone patterns need to be obtained and the tracks to be recognized are compared. However, the traditional recognition method based on RNN-RBM language model can only obtain phoneme, word and sentence. The accuracy of recognition can not be reduced because the corresponding standard tone mode can not be obtained and the comparison can not be completed. A speech tone recognition method based on K-means initialized EM algorithm is proposed. By establishing the mixed model of tone information Gao Si, the probability density function of fundamental frequency information is fitted accurately, and the feature parameters of fundamental frequency are extracted by maximization (EM) algorithm, and more tone modes are obtained based on this model. Using K-means to initialize EM algorithm, the problem that EM algorithm is sensitive to initial value selection is eliminated, and then the order of Gao Si mixed model is predicted to improve the accuracy of tone recognition of EM algorithm. The simulation results show that the improved tone recognition method has high accuracy and practicability.
【作者單位】： 甘肅農(nóng)業(yè)大學(xué)信息科學(xué)與技術(shù)學(xué)院;
【分類(lèi)號(hào)】：TN912.34

【參考文獻(xiàn)】

相關(guān)期刊論文 前7條

1 譚萍;邢玉娟;;噪聲環(huán)境下文本相關(guān)說(shuō)話人識(shí)別方法改進(jìn)[J];西安工程大學(xué)學(xué)報(bào);2016年05期

2 張毅;黎小松;羅元;吳承軍;;基于人耳聽(tīng)覺(jué)特性的語(yǔ)音識(shí)別預(yù)處理研究[J];計(jì)算機(jī)仿真;2015年12期

3 高霞;李瑞俊;;EM算法在不完全數(shù)據(jù)參數(shù)估計(jì)中的應(yīng)用[J];集寧師范學(xué)院學(xué)報(bào);2015年03期

4 張玉梅;胡小俊;吳曉軍;白樹(shù)林;路綱;;語(yǔ)音信號(hào)序列的Volterra預(yù)測(cè)模型[J];物理學(xué)報(bào);2015年20期

5 吳蔚瀾;蔡猛;田W，

本文編號(hào)：2384592