计算机科学 ›› 2016, Vol. 43 ›› Issue (8): 297-299.doi: 10.11896/j.issn.1002-137X.2016.08.060
罗元,孙龙
LUO Yuan and SUN Long
摘要： 为提高说话人确认系统在噪声环境下的鲁棒性,在利用听觉外周模型改进Mel频率倒谱系数(Mel FrequencyCepstral Coefficient,MFCC)的基础上,结合感知线性预测系数(Perceptual Linear Predictive Coefficient,PLPC),以类间区分度为依据,在特征域对两种声纹特征进行融合,提出一种新的声纹特征提取方法,并对基于该特征的说话人确认系统的噪声鲁棒性进行研究。针对不同信噪比的语音信号进行了融合特征与原始特征的对比实验,结果表明,融合特征在模拟餐厅噪声环境中的错误率更低,较MFCC与PLPC分别降低了2.2%和3.1%,说话人确认系统在噪声中的鲁棒性得到提升。
| [1] Pols L C W,Schouten M E H.Perception of tone,band,and formant sweeps[M]∥The psychophysics of Speech Perception.Springer Netherlands,1987:231-240 [2] Hsu W,Sun J.The Effectiveness of Linear Prediction Residual to the Verification of Voiceprint and the Recognition of Chinese Tone[C]∥IEEE International Symposium on Multimedia.IEEE,2010:353-356 [3] Patil H A,Basu T K.Comparison of subband cepstrum and Mel cepstrum for open set speaker classification[C]∥Proceedings of the IEEE INDICON 2004 India Annual Conference,2004.First.IEEE,2004:35-40 [4] Srinivasan A.Speaker identification and Verification using Vector quantization and Mel frequency Cepstral Coefficients[J].Engineering and Technology,2012,4(1):33-40 [5] Hermansky H.Perceptual linear predictive (PLP) analysis ofspeech[J].The Journal of the Acoustical Society of America,1990,87(4):1738-1752 [6] Li Y G,Ouyang X Z,Z F.A Study on Robust Speech Recognition Based on Gamatone feature[C]∥ National Conference on Man-Machine Speech Communication Chinese Information Processing Society of China.2013:26-29(in Chinese) 李银国,欧阳希子,郑方.基于Gammatone特征的鲁棒语音识别研究[C]∥十二届全国人机语音通讯学术会议(NCMMSC’2013).2013:26-29 [7] Luo Y,Chen J,Zhang Y.An Auditory Feature Extraction Algorithm Based on Gama-chirp Filter Banks[J].Information and Control,2013,2(5):590-594(in Chinese) 罗元,陈君,张毅.基于伽马啁啾滤波器组的听觉特征提取算法[J].信息与控制,2013,42(5):590-594 [8] Makhoul J,Cosell L.LPCW:An LPC vocoder with linear predictive spectral warping[C]∥ IEEE International Conference on Acoustics,Speech,and Signal Processing(ICASSP’76).IEEE,1976:466-469 [9] Varga A,Steeneken H J M.Assessment for automatic speech recognition:II.NOISEX-92:A database and an experiment to study the effect of additive noise on speech recognition systems[J].Speech Communication,1993,12(3):247-251 [10] Bao H J,Zheng F.Combined GMM-UBM and SVM speaker identification system[J].Tsinghua University (Science & Technology),2007,8(S1):693-698(in Chinese) 鲍焕军,郑方.GMM-UBM和SVM说话人辨认系统及融合的分析[J].清华大学学报(自然科学版),2007,8(S1):693-698 |
| No related articles found! |
|
||
首页