1) discriminative model training
区分性模型训练
1.
Considering the characteristics of such features and differences between audio types,a hierarchical discrimination algorithm was proposed based on discriminative model training and feature filtration,which trained GMMs(Gaussian mixture models) in each layer and selected the feature subset resulting in maximal separability for them.
根据该特征参数的特点以及各类音频之间的差异,给出了一种区分性模型训练和特征筛选相结合的多级二分类音频分类方法,即为各级建立GMM(Gaussian mixture model)模型的同时挑选出使当前模型区分程度达到最大的特征子集。
2) discriminative training
区分性训练
1.
First,discriminative training on the HMM (hidden Markov model) based tone models is proposed.
提出了2种解决汉语语音识别中声调问题的方法:利用区分性方法对基于隐马尔可夫模型(HMM)的声调模型进行训练;提出将区分性训练的声调模型加入大词汇量连续语音识别系统的最优方法,该方法根据最小音子错误的训练准则以及利用扩展Baum-Welch算法区分性训练与模型相关的概率权重,对声学模型以及声调模型概率进行加权。
2.
Discriminative Training has become one of the standard configuration methodsfor the state-of-the-art acoustic modeling and parameters optimization.
区分性训练已经成为自动语音识别中声学模型训练和参数优化的标配方法之一。
3.
Hidden Markov model(HMM) and dependency-tree hidden Markov model(DT-HMM) improvements through discriminative training were proven theoretically possible,so a DT-HMM model with complete parameters was derived and proven to be consistent with the HMM model.
区分性训练可以很好地弥补由于训练样本的缺乏对识别系统所造成的影响,能够提高非特定人手语识别的识别率。
3) model training
模型训练
1.
The latest development in the areas of the feature extraction, model training and classification is reviewed and the trend and rubs are also discussed.
详细介绍了说话人识别的基本原理,从特征提取、模型训练和分类等几个方面就近年的主要研究情况进行综述和评价,并在此基础上探讨了研究难点和发展前景。
4) training model
训练模型
1.
This paper analyses the statistics of body slam in sanda and discusses the importanc of body slam and related factors,and proposes a training model.
对散打竞技场上运用下潜抱摔进行量化分析,探讨其在比赛中的重要性及影响其得分的相关因素,并依此构建了其训练模型。
5) Mixed Training Model(MTM)
混合训练模型
6) Audio Training Pattern
语音模型训练
补充资料：连续性与非连续性(见间断性与不间断性)
连续性与非连续性(见间断性与不间断性)
continuity and discontinuity
11an父ux泊g四f“山。麻以角g、.连续性与非连续性(c。nt，n琳t:nuity一)_见间断性与不间断性。and diseo红ti-
说明：补充资料仅用于学习参考，请勿用于其它任何用途。
参考词条