… (Discriminative Training)法則來加以改 進.故本文著重於探討以最小化音素錯誤(Minimum Phone
Error Training… 對於模型參數訓練時的貢獻,可以引入所謂的I-Smoothing 技術[12],其公式如下: …

… Minimum phone error and I-smoothing for improved discriminative training. In: Proceedings
… A post-processing system to yield reduced word error rates: recognizer output voting error …

… the use MPE (Minimum Phone Error Discriminative Training) to … such as adjusting I-smoothing
factors or phone lattices to … calculation, all based on a better use of n-best recognition (3.3.…

… The I-smoothing technique introduces model’s prior to … The transform matrix is trained by
minimum phone error followed by … -space minimum phone error has larger improvements and …

… Minimum phoneme error based heteroscedastic linear discriminant analysis for speech …
Minimum phone error and Ismoothing for improved discriminative training. In: Proceedings of …

… 使用了最小化音素錯誤 (Minimum Phone Error, MPE)鑑別式訓練… 化音素錯誤線性迴歸(Minimum
Phone Error Linear Regression, … 次迭代的鑑別式訓練(Large Scale Discriminative Training),…

… In doing so, the error-correcting capability of the acoustic models can be enhanced. The …
模型 (Four-gram phone LM) 且不會退化小於3 連音素語言模型(Tri-gram phone LM),取代傳 統…

… Minimum phone error and ismoothing for improved discriminative training. In: Proceedings
of the 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing. …

適合雙語混合語言辨識的最小音素錯誤訓練= Minimum Phone Error Training for Code-Mixed
Bilingual … Minimum Phone Error Training for Code-Mixed Bilingual Speech Recognition …

… Abstract: We propose a novel discriminative learning approach for … the minimum decision
value for positive samples and the maximum decision value for negative samples. The learning …