计算机工程与应用2011，Vol.47Issue(25)：152-155,164,5.DOI:10.3778/j.issn.1002-8331.2011.25.040
基于高斯混合模型的自然环境声音的识别
Natural sounds recognition using GMM distribution
摘要
Abstract
A recognition method for natural sounds based on Gaussian Mixture Model (GMM) distribution is proposed. Mel-Frequency Cepstral Coefficients (MFCCs) are used to analyze natural sounds for their feature extraction.The expectation maximization algorithm is used to learn a Gaussian mixture model distribution of MFCCs for the set of audio feature vectors that describe each sound.Minimum classification error criterion and vote rule are used to yield higher recognition accuracy for natural sounds.Experimentally,compared with ^-Nearest Neighbor(KNN) method,GMM is able to achieve a higher accuracy rate for discriminating 36 classes of natural sounds.The classified accuracy rate of GMM reaches to 95.83%.关键词
Mel频率倒谱系数/高斯混合模型/自然环境声音的识别/投票裁决Key words
Mel-frequency cepstral coefficients/Gaussian mixture model/natural sounds recognition/vote rule分类
信息技术与安全科学引用本文复制引用
余清清,李应,李勇..基于高斯混合模型的自然环境声音的识别[J].计算机工程与应用,2011,47(25):152-155,164,5.基金项目
国家自然科学基金(the National Natural Science Foundation of China under Grant No.61075022) （the National Natural Science Foundation of China under Grant No.61075022）
福建省教育厅A类科技项目(No.JA09021). （No.JA09021）
