Language Identification of Minority Language based on GMM-UBM Model

摘要：

An approach to language identification of minority language based on GM.VI-lBM model is described in this paper. In the training stage, a new method of double threshold for voice activity detection is used to effectively remove noise and extract useful voice frames. Then we extract the MFCC feature parameters, and train IIBM model and the GMM model of 6 languages; In the testing stage, utterances with different durations and Chinese loan words of six minority languages are selected. We analyze each language identification rate and the results with different duration testing data, and then we give some explanations of error identification in terms of phonetics. We also analyze the impact of Chinese loan words on the results.

关键词： Minority Language Voice Activity Detection GMM-UBM Language Identification Chinese Loan Words

作者: Libo Zuo Jian Yang Linyu He

作者单位: School of Information Science and Engineering, Yunnan University, Kunming, Yunnan, China

会议类型: 国际会议

会议名称: 2010 International Conference on Circuit and Signal Processing(2010年电路与信号处理国际会议 ICCSP 2010)

会议地点: 上海

会议语种:英文

页码: 397-400

在线出版日期: 2010-12-25（万方平台首次上网日期，不代表论文的发表时间）