Speech enhancement based on audible noise suppression

被引：116
作者：
Tsoukalas, DE
Mourjopoulos, JN
Kokkinakis, G
机构：
[1] Wire Communications Laboratory, University of Fatras
来源：
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING | 1997年 / 5卷 / 06期
关键词：
D O I：
10.1109/89.641296
中图分类号：
O42 [声学];
学科分类号：
070206 ; 082403 ;
摘要：
A novel speech enhancement technique is presented based on the definition of the psychoacoustically derived quantity of audible noise spectrum and its subsequent suppression using optimal nonlinear filtering of the short-time spectral amplitude (STSA) envelope, The filter operates with sparse spectral estimates obtained from the STSA, and, when these parameters are accurately known, significant intelligibility gains, up to 40%, result in the processed speech signal, These parameters can be also estimated from noisy data, resulting into smaller but significant intelligibility gains.
引用
收藏
页码：497 / 514
页数：18
相关论文
共 32 条
  • [1] BENIOT MG, 1991, SAMICPUCL001
  • [2] BEROUTI M, 1979, APR P IEEE INT C AC, P208
  • [3] SUPPRESSION OF ACOUSTIC NOISE IN SPEECH USING SPECTRAL SUBTRACTION
    BOLL, SF
    [J]. IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1979, 27 (02): : 113 - 120
  • [4] BRANDENBURG K, 1994, J AUDIO ENG SOC, V42, P780
  • [5] SPEECH ENHANCEMENT BASED CONCEPTUALLY ON AUDITORY EVIDENCE
    CHENG, YM
    OSHAUGHNESSY, D
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1991, 39 (09) : 1943 - 1954
  • [6] *DEF RES AG, 1992, NOISEX 92 CD ROMS NA
  • [7] DONE J, 1979, P IEEE ICASSP, P228
  • [8] SPEECH ENHANCEMENT USING A MINIMUM MEAN-SQUARE ERROR SHORT-TIME SPECTRAL AMPLITUDE ESTIMATOR
    EPHRAIM, Y
    MALAH, D
    [J]. IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1984, 32 (06): : 1109 - 1121
  • [9] FRAZIER H, P IEEE ICASSP, V251, P76
  • [10] Gradshteyn I. S., 1994, TABLES INTEGRALS SER
1234