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 makalah ini, keteguhan ukuran keyakinan berasaskan posterior dipertingkatkan dengan menggunakan maklumat entropi, yang dikira untuk posterior peringkat unit pertuturan hanya menggunakan hasil pengecaman terbaik, tanpa memerlukan beban pengiraan yang lebih besar daripada kaedah konvensional. Menggunakan kaedah normalisasi yang berbeza, dua ukuran keyakinan entropi berasaskan posterior dicadangkan. Butiran praktikal dibincangkan untuk dua tahap tipikal ukuran keyakinan posterior berasaskan model Markov (HMM) tersembunyi, dan kedua-dua tahap dibandingkan dari segi prestasinya. Eksperimen menunjukkan bahawa maklumat entropi menghasilkan peningkatan ketara dalam ukuran keyakinan berasaskan posterior. Peningkatan mutlak kadar penolakan di luar perbendaharaan kata (OOV) adalah lebih daripada 20% untuk kedua-dua ukuran keyakinan peringkat fonem dan langkah keyakinan peringkat negeri untuk set ujian terbenam kami, tanpa penurunan ketara dalam perbendaharaan kata. ketepatan.
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
Yanqing SUN, Yu ZHOU, Qingwei ZHAO, Pengyuan ZHANG, Fuping PAN, Yonghong YAN, "Enhancing the Robustness of the Posterior-Based Confidence Measures Using Entropy Information for Speech Recognition" in IEICE TRANSACTIONS on Information,
vol. E93-D, no. 9, pp. 2431-2439, September 2010, doi: 10.1587/transinf.E93.D.2431.
Abstract: In this paper, the robustness of the posterior-based confidence measures is improved by utilizing entropy information, which is calculated for speech-unit-level posteriors using only the best recognition result, without requiring a larger computational load than conventional methods. Using different normalization methods, two posterior-based entropy confidence measures are proposed. Practical details are discussed for two typical levels of hidden Markov model (HMM)-based posterior confidence measures, and both levels are compared in terms of their performances. Experiments show that the entropy information results in significant improvements in the posterior-based confidence measures. The absolute improvements of the out-of-vocabulary (OOV) rejection rate are more than 20% for both the phoneme-level confidence measures and the state-level confidence measures for our embedded test sets, without a significant decline of the in-vocabulary accuracy.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E93.D.2431/_p
Salinan
@ARTICLE{e93-d_9_2431,
author={Yanqing SUN, Yu ZHOU, Qingwei ZHAO, Pengyuan ZHANG, Fuping PAN, Yonghong YAN, },
journal={IEICE TRANSACTIONS on Information},
title={Enhancing the Robustness of the Posterior-Based Confidence Measures Using Entropy Information for Speech Recognition},
year={2010},
volume={E93-D},
number={9},
pages={2431-2439},
abstract={In this paper, the robustness of the posterior-based confidence measures is improved by utilizing entropy information, which is calculated for speech-unit-level posteriors using only the best recognition result, without requiring a larger computational load than conventional methods. Using different normalization methods, two posterior-based entropy confidence measures are proposed. Practical details are discussed for two typical levels of hidden Markov model (HMM)-based posterior confidence measures, and both levels are compared in terms of their performances. Experiments show that the entropy information results in significant improvements in the posterior-based confidence measures. The absolute improvements of the out-of-vocabulary (OOV) rejection rate are more than 20% for both the phoneme-level confidence measures and the state-level confidence measures for our embedded test sets, without a significant decline of the in-vocabulary accuracy.},
keywords={},
doi={10.1587/transinf.E93.D.2431},
ISSN={1745-1361},
month={September},}
Salinan
TY - JOUR
TI - Enhancing the Robustness of the Posterior-Based Confidence Measures Using Entropy Information for Speech Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 2431
EP - 2439
AU - Yanqing SUN
AU - Yu ZHOU
AU - Qingwei ZHAO
AU - Pengyuan ZHANG
AU - Fuping PAN
AU - Yonghong YAN
PY - 2010
DO - 10.1587/transinf.E93.D.2431
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E93-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2010
AB - In this paper, the robustness of the posterior-based confidence measures is improved by utilizing entropy information, which is calculated for speech-unit-level posteriors using only the best recognition result, without requiring a larger computational load than conventional methods. Using different normalization methods, two posterior-based entropy confidence measures are proposed. Practical details are discussed for two typical levels of hidden Markov model (HMM)-based posterior confidence measures, and both levels are compared in terms of their performances. Experiments show that the entropy information results in significant improvements in the posterior-based confidence measures. The absolute improvements of the out-of-vocabulary (OOV) rejection rate are more than 20% for both the phoneme-level confidence measures and the state-level confidence measures for our embedded test sets, without a significant decline of the in-vocabulary accuracy.
ER -