The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dalam surat ini, kami mencadangkan kaedah penyamaan histogram baharu untuk mengimbangi ketidakpadanan akustik yang disebabkan terutamanya oleh kerosakan bunyi aditif dan herotan saluran dalam pengecaman pertuturan. Kaedah yang dicadangkan menggunakan fungsi pengedaran kumulatif ujian (CDF) yang dipertingkatkan dengan lebih tepat melicinkan CDF ujian berasaskan statistik pesanan konvensional dengan penggunaan fungsi tetingkap untuk pampasan ciri yang mantap. Eksperimen pada rangka kerja AURORA 2 mengesahkan bahawa kaedah yang dicadangkan berkesan dalam mengimbangi ciri pengecaman pertuturan dengan mengurangkan purata ralat relatif sebanyak 13.12% berbanding kaedah penyamaan histogram konvensional berasaskan statistik pesanan dan sebanyak 58.02% berbanding ciri berasaskan mel-cepstral untuk tiga set ujian.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Salinan
Youngjoo SUH, Hoirin KIM, Munchurl KIM, "Histogram Equalization Utilizing Window-Based Smoothed CDF Estimation for Feature Compensation" in IEICE TRANSACTIONS on Information,
vol. E91-D, no. 8, pp. 2199-2202, August 2008, doi: 10.1093/ietisy/e91-d.8.2199.
Abstract: In this letter, we propose a new histogram equalization method to compensate for acoustic mismatches mainly caused by corruption of additive noise and channel distortion in speech recognition. The proposed method employs an improved test cumulative distribution function (CDF) by more accurately smoothing the conventional order statistics-based test CDF with the use of window functions for robust feature compensation. Experiments on the AURORA 2 framework confirmed that the proposed method is effective in compensating speech recognition features by reducing the averaged relative error by 13.12% over the order statistics-based conventional histogram equalization method and by 58.02% over the mel-cepstral-based features for the three test sets.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e91-d.8.2199/_p
Salinan
@ARTICLE{e91-d_8_2199,
author={Youngjoo SUH, Hoirin KIM, Munchurl KIM, },
journal={IEICE TRANSACTIONS on Information},
title={Histogram Equalization Utilizing Window-Based Smoothed CDF Estimation for Feature Compensation},
year={2008},
volume={E91-D},
number={8},
pages={2199-2202},
abstract={In this letter, we propose a new histogram equalization method to compensate for acoustic mismatches mainly caused by corruption of additive noise and channel distortion in speech recognition. The proposed method employs an improved test cumulative distribution function (CDF) by more accurately smoothing the conventional order statistics-based test CDF with the use of window functions for robust feature compensation. Experiments on the AURORA 2 framework confirmed that the proposed method is effective in compensating speech recognition features by reducing the averaged relative error by 13.12% over the order statistics-based conventional histogram equalization method and by 58.02% over the mel-cepstral-based features for the three test sets.},
keywords={},
doi={10.1093/ietisy/e91-d.8.2199},
ISSN={1745-1361},
month={August},}
Salinan
TY - JOUR
TI - Histogram Equalization Utilizing Window-Based Smoothed CDF Estimation for Feature Compensation
T2 - IEICE TRANSACTIONS on Information
SP - 2199
EP - 2202
AU - Youngjoo SUH
AU - Hoirin KIM
AU - Munchurl KIM
PY - 2008
DO - 10.1093/ietisy/e91-d.8.2199
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E91-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2008
AB - In this letter, we propose a new histogram equalization method to compensate for acoustic mismatches mainly caused by corruption of additive noise and channel distortion in speech recognition. The proposed method employs an improved test cumulative distribution function (CDF) by more accurately smoothing the conventional order statistics-based test CDF with the use of window functions for robust feature compensation. Experiments on the AURORA 2 framework confirmed that the proposed method is effective in compensating speech recognition features by reducing the averaged relative error by 13.12% over the order statistics-based conventional histogram equalization method and by 58.02% over the mel-cepstral-based features for the three test sets.
ER -