计算机科学 ›› 2018, Vol. 45 ›› Issue (4): 278-284.doi: 10.11896/j.issn.1002-137X.2018.04.047
赵利博,刘奇,付方玲,何凌
ZHAO Li-bo, LIU Qi, FU Fang-ling and HE Ling
摘要: 为实现对腭裂高鼻音等级的自动识别,通过对语音信号小波处理和特征提取方法的综合研究,提出基于小波分解系数倒谱特征的腭裂高鼻音等级自动识别算法。目前,研究人员对腭裂语音的研究多基于MFCC、Teager能量、香农能量等特征,识别正确率偏低,且计算量过大。文中对4种等级腭裂高鼻音的1789个元音\a\语音数据提取小波分解系数倒谱特征参数,使用KNN分类器对4种不同等级的高鼻音进行自动识别,将识别结果与MFCC、LPCC、基音周期、共振峰和短时能量共5种经典声学特征的识别结果作比较,同时使用SVM分类器对不同等级的腭裂高鼻音进行自动识别,并与KNN分类器进行对比。实验结果表明,基于小波分解系数倒谱特征的识别结果优于经典声学特征,且KNN分类器的识别结果优于SVM分类器。小波分解系数倒谱特征在KNN中的识别率最高达到91.67%,在SVM中达到87.60%,经典声学特征在KNN分类器中的识别率为21.69%~84.54%,在SVM中的识别率为30.61%~78.24%。
[1] CHEN R J.The state and consider about speech therapy of cleft palate in China[J].International Journal of Stomatology,2012,39(1):1-5. [2] ARIAS-LONDOO J D,GODINO-LLORENTE J I,SAeNz-LECHON N,et al.Automatic Detection of Pathological Voices Using Complexity Measures,Noise Parameters,and Mel-Cepstral Coefficients[J].IEEE Transactions on Bio-medical Engineering,2011,58(2):370-379. [3] SMYTH A.Clinical grading system for submucous cleft palate[J].British Journal of Oral & Maxillofacial Surgery,2014,52(3):275-276. [4] VILLAFUERTE GONZALEZ R,et al.Acoustic analysis ofvoice in children with cleft palate and velopharyngeal insufficiency[J].International Journal of Pediatric Otorhinolaryngology,2015,79(7):1073-1076. [5] MAIER A,HONIG F,HACKER C,et al.Automatic evaluation of characteristic speech disorders in children with cleft lip and palate[C]∥Conference of the International Speech Communication Association.Brisbance,Australia,2008:270-278. [6] ARROYAVE J R O,BONILLA J F V,et al.Automatic detection of hypernasality in children[C]∥International Conference on Interplay Between Natural and Artificial Computation:New Challenges on Bioinspired Applications.Spain,2011:167-174. [7] HE L,ZHANG J,LIU Q,et al.Automatic Evaluation of Hypernasality and Consonant Misarticulation in Cleft Palate Speech[J].IEEE Signal Processing Letters,2014,21(10):1298-1301. [8] KUMARI V S R,DEVARAKONDA D K.A Wavelet Based Denoising of Speech Signal[J].International Journal of Enginee-ring Trends & Technology,2013,5(2):107-115. [9] CHEN Z,WANG S,YIN F.A Time Delay Estimation Method Based on Wavelet Transform and Speech Envelope for Distributed Microphone Arrays[J].Advances in Electrical & Computer Engineering,2013,13(3):39-44. [10] MALLAT S G.A Theory for Multiresolution Signal Decomposition:The Wavelet Representation[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,1989,1(7):674-693. [11] 成礼智,王红霞.小波的理论与应用[M].北京:科学出版社,2004. [12] 刘明才.小波分析及其应用(第2版)[M].北京:清华大学出版社,2013. [13] ZHAO L.Speech Signal Processing[M].Beijing:China Machine Press,2012. [14] DAVE N.Feature Extraction Methods LPC,PLP and MFCC In Speech Recognition[J].Ijaret Org,2013,1(6):1-5. [15] ALI Z,ABBAS A W,THASLEEMA A W,et al.Database development and automatic speech recognition of isolated Pashto spoken digits using MFCC and K-NN[J].International Journal of Speech Technology,2015,18(2):271-275. [16] AARON M,GANESH B,RATNADEEP R.Automatic SpeechRecognition and Verification using LPC,MFCC and SVM[J].International Journal of Computer Applications,2015,127(8):47-52. |
No related articles found! |
|