计算机系统应用
計算機繫統應用
계산궤계통응용
APPLICATIONS OF THE COMPUTER SYSTEMS
2015年
4期
32-37
,共6页
张凤仪%夏秀渝%冉国敬%何礼%叶于林
張鳳儀%夏秀渝%冉國敬%何禮%葉于林
장봉의%하수투%염국경%하례%협우림
说话人识别%语音增强%方位信息%时频掩蔽%MFCC参数
說話人識彆%語音增彊%方位信息%時頻掩蔽%MFCC參數
설화인식별%어음증강%방위신식%시빈엄폐%MFCC삼수
speaker recognition%speech enhancement%sound source azimuth information%time-frequency masking%MFCC
针对多声源干扰环境下说话人识别系统性能急剧下降的问题,提出一种提取目标语音的前端处理方法,该方法依据独立语音时频域的近似稀疏性,基于目标语音方位信息采用非线性时频掩蔽方法提取目标语音。建立了基于梅尔倒谱系数(MFCC)的高斯混合模型(GMM)说话人识别系统。仿真实验证明,该方法能有效提取目标语音,提高说话人识别系统的鲁棒性。该文多声源干扰仿真实验条件下,说话人识别系统的识别率平均提高了25%左右。
針對多聲源榦擾環境下說話人識彆繫統性能急劇下降的問題,提齣一種提取目標語音的前耑處理方法,該方法依據獨立語音時頻域的近似稀疏性,基于目標語音方位信息採用非線性時頻掩蔽方法提取目標語音。建立瞭基于梅爾倒譜繫數(MFCC)的高斯混閤模型(GMM)說話人識彆繫統。倣真實驗證明,該方法能有效提取目標語音,提高說話人識彆繫統的魯棒性。該文多聲源榦擾倣真實驗條件下,說話人識彆繫統的識彆率平均提高瞭25%左右。
침대다성원간우배경하설화인식별계통성능급극하강적문제,제출일충제취목표어음적전단처리방법,해방법의거독립어음시빈역적근사희소성,기우목표어음방위신식채용비선성시빈엄폐방법제취목표어음。건립료기우매이도보계수(MFCC)적고사혼합모형(GMM)설화인식별계통。방진실험증명,해방법능유효제취목표어음,제고설화인식별계통적로봉성。해문다성원간우방진실험조건하,설화인식별계통적식별솔평균제고료25%좌우。
The Speaker Recognition System is significantly affected by the Multi-Sound sources problem. In order to overcome this problem, a target sound extraction algorithm named time-frequency masking is proposed. The proposed algorithm is based on the sound source azimuth information and the approximate sparse nature of sound. A Mel-frequency cepstral coefficient (MFCC) based Gaussian mixture model (GMM) speaker recognition system is presented to improve the recognition robustness. The proposed algorithm has been tested on the simulated data through a number of experiments which shows the efficiency and robustness of the proposed algorithm. In the Multi-Sound sources environment, the recognition rate of the proposed algorithm can be improved by about 25%.