【摘要】：作為人機(jī)交互技術(shù)中重要的成員之一，語音情感識別技術(shù)被廣泛應(yīng)用在教育、醫(yī)療、通信、計(jì)算機(jī)、自動(dòng)化等行業(yè)。同時(shí)，語音情感識別涉及的知識面很廣，涵蓋了計(jì)算機(jī)科學(xué)與技術(shù)、模式識別、語音學(xué)、心理學(xué)、統(tǒng)計(jì)學(xué)和信號處理等學(xué)科。具有良好的研究基礎(chǔ)和廣闊的發(fā)展前景。 目前，語音情感識別的相關(guān)研究已經(jīng)取得很多的成果，同時(shí)也存在著許多研究困難。通過改進(jìn)分類算法準(zhǔn)確率可以提高語音情感識別產(chǎn)品和系統(tǒng)的性能，使其提供更好的服務(wù)和用戶體驗(yàn)，提高一些行業(yè)的工作質(zhì)量和效率，對促進(jìn)行業(yè)發(fā)展具有重要意義。 在語音情感識別方面，SVM算法具有很好的分類性能，而AdaBoost算法可以進(jìn)一步提升SVM算法的分類準(zhǔn)確率。 本文以SVM和AdaBoost算法為基礎(chǔ)，提出一種新的集成學(xué)習(xí)算法，即選擇性AdaBoostSVM算法，算法思路為：首先使用AdaBoost算法訓(xùn)練若干個(gè)SVM分類器，再通過Kmeans算法對這些分類器進(jìn)行聚類，得到若干個(gè)代表分類器，，然后對每一個(gè)測試樣本，均使用Knn算法從訓(xùn)練集中找出其最近鄰的若干訓(xùn)練樣本，并將這些訓(xùn)練樣本放入代表分類器中測試，最后選出測試準(zhǔn)確率最高的分類器作為當(dāng)前測試樣本的最終分類器。 本文在EMO-DB德語語音庫、CASIA中文語音庫和SAVEE英語語音庫下測試算法效果。先找出三個(gè)語音庫在五倍交叉驗(yàn)證和十倍交叉驗(yàn)證下的最優(yōu)SVM參數(shù)，再分別對三個(gè)語音庫作五倍交叉驗(yàn)證和十倍交叉驗(yàn)證的測試。 實(shí)驗(yàn)結(jié)果表明，本文算法對三個(gè)語音庫的分類準(zhǔn)確率均有提升。 在五倍交叉驗(yàn)證中，本文算法對三個(gè)語音庫的分類準(zhǔn)確率較單一SVM算法分別提升了1.86%、1.51%和3.77%，較AdaBoostSVM算法提升了0.35%、0.78%和0.13%，分類準(zhǔn)確率達(dá)到87.56%、81.75%和76.75%。 在十倍交叉驗(yàn)證中，本文算法對三個(gè)語音庫的分類準(zhǔn)確率較單一SVM算法分別提升了1.46%、0.74%和1.86%，較AdaBoostSVM算法提升了0.21%、0.37%和1.86%，分類準(zhǔn)確率達(dá)到87.29%、81.20%和76.51%。 說明本文提出的選擇性AdaBoostSVM算法在提升語音情感識別準(zhǔn)確率上是可行的。
[Abstract]:As an important member of human-computer interaction technology, speech emotion recognition technology is widely used in education, medical treatment, communication, computer, automation and other industries. At the same time, speech emotion recognition involves a wide range of knowledge, covering computer science and technology, pattern recognition, phonetics, psychology, statistics and signal processing and other disciplines. It has a good research foundation and broad development prospect. At present, the research of speech emotion recognition has made a lot of achievements, but also there are many difficulties. By improving the accuracy of classification algorithm, we can improve the performance of speech emotion recognition products and systems, make them provide better service and user experience, and improve the working quality and efficiency of some industries, which is of great significance to promote the development of the industry. SVM algorithm has good classification performance in speech emotion recognition and AdaBoost algorithm can further improve the classification accuracy of SVM algorithm. Based on SVM and AdaBoost algorithm, a new ensemble learning algorithm, selective AdaBoost SVM algorithm, is proposed in this paper. The idea of the algorithm is as follows: firstly, several SVM classifiers are trained by AdaBoost algorithm, and then these classifiers are clustered by Kmeans algorithm. A number of representative classifiers are obtained, and then, for each test sample, the Knn algorithm is used to find some training samples from the training set, and the training samples are put into the representative classifier for testing. Finally, the classifier with the highest test accuracy is selected as the final classifier of the current test sample. In this paper, the algorithm effect is tested in EMO-DB German language corpus, CASIA Chinese language corpus and SAVEE English phonetic database. The optimal SVM parameters of the three speech banks under five times cross validation and ten times cross validation are found out first, and then the five times cross validation and ten times cross verification tests of the three speech banks are performed respectively. The experimental results show that the classification accuracy of the algorithm is improved. In the five-fold cross-validation, the classification accuracy of the three corpora in this paper is 1.86% and 3.77% higher than that of the single SVM algorithm, 0.35% and 0.13% higher than that of the AdaBoost SVM algorithm, respectively, and the classification accuracy is 87.56% and 76.75%. In the 10-fold cross-validation, the classification accuracy of the three corpora in this paper is 1.46% and 1.86% higher than that of the single SVM algorithm, 0.21% and 1.86% higher than that of the AdaBoost SVM algorithm, and the classification accuracy is 87.29% 81.20% and 76.51% higher than that of the AdaBoost SVM algorithm. It shows that the proposed selective boost SVM algorithm is feasible in improving the accuracy of speech emotion recognition.
【學(xué)位授予單位】：華南理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TN912.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 林奕琳;韋崗;楊康才;;語音情感識別的研究進(jìn)展[J];電路與系統(tǒng)學(xué)報(bào);2007年01期

2 黃程韋;趙艷;金峗;于寅驊;趙力;;實(shí)用語音情感的特征分析與識別的研究[J];電子與信息學(xué)報(bào);2011年01期

3 張永;張衛(wèi)國;徐維軍;;基于數(shù)據(jù)分割和集成學(xué)習(xí)的大規(guī)模SVM分類算法[J];系統(tǒng)工程;2009年03期

4 李亞軍;劉曉霞;陳平;;改進(jìn)的AdaBoost算法與SVM的組合分類器[J];計(jì)算機(jī)工程與應(yīng)用;2008年32期

5 王曉丹;孫東延;鄭春穎;張宏達(dá);趙學(xué)軍;;一種基于AdaBoost的SVM分類器[J];空軍工程大學(xué)學(xué)報(bào)(自然科學(xué)版);2006年06期

6 王曉丹;高曉峰;姚旭;雷蕾;;SVM集成研究與應(yīng)用[J];空軍工程大學(xué)學(xué)報(bào)(自然科學(xué)版);2012年02期

7 李書玲;劉蓉;張鎏欽;劉紅;;基于改進(jìn)型SVM算法的語音情感識別[J];計(jì)算機(jī)應(yīng)用;2013年07期

8 張石清;趙知?jiǎng)?戴育良;楊廣映;;支持向量機(jī)應(yīng)用于語音情感識別的研究[J];聲學(xué)技術(shù);2008年01期

9 韓文靜;李海峰;阮華斌;馬琳;;語音情感識別研究進(jìn)展綜述[J];軟件學(xué)報(bào);2014年01期

10 胡洋;吳黎慧;高磊;蒲南江;;基于SVM的語音情感識別研究[J];電子測試;2011年09期

相關(guān)博士學(xué)位論文 前3條

1 劉佳;語音情感識別的研究與應(yīng)用[D];浙江大學(xué);2009年

2 程麗麗;支持向量機(jī)集成學(xué)習(xí)算法研究[D];哈爾濱工程大學(xué);2009年

3 趙臘生;語音情感特征提取與識別方法研究[D];大連理工大學(xué);2010年

本文編號：2135356