...之用语者正规化 (Speaker Normalization)在特征参数撷取 (feature extraction) 阶段，或声学模型训练 (acoustic model training) 阶段，设法消除语者间变异性代表技术：Vocal Tract Length Normalization (VTLN), Speaker Adaptive Training (SAT)语者调适 (S...
基于8个网页-相关网页
From the Angle of controlling the training, an adjustable acoustic rhythm model was established for approach training of long jumpers.
从控制训练的角度出发，为跳远运动员建立一种可随时调整的音响助跑节奏模式。
Four typical discriminative training criterions and some updating methods of acoustic model parameters are introduced, then, they are defined in a unified framework.
再经过对各种区分性训练准则的目标函数和参数更新算法进行推导和比较，将它们统一地纳入到一套训练框架体系之中。
应用推荐