… μ is also a Gaussian distribution with mean γ and variance 2 … And if the conjugate prior for
the mean μ is substituted into Eq… ], MLLR adaptation was firstly proposed under the framework …

… model will integrate into speaker dependent model. Last, we estimate the adapted MLLR
transformation matrix set by weighting function with recognition result, the present MLLR matrix, …

… MLLR uses the technique of regression trees. It calculates the … one in MLLR adaptation,
and use MPE to make further adjustment to acoustic models which have been adapted by MLLR. …

… MLLR) framework, this paper proposes two fast speaker adaptation approaches,which are
called Speaker Adaptation … ( SATD) and Speaker Adaptation using Shared Block Diagonal …

… In this paper, a new model adaptation using vector Taylor series is presented, which adapts
the mean … 本文分别做了MLLR 均值自适应算法(MLLRMean)[6],MLLR 均值和方差自适应算法(…

… 我們使用Aurora2 語料庫做實驗, 在複合情境的訓練模式下, 我們得到最佳的辨識錯誤率為
4.37%, 比傳統強健性參數求取方法MVA (Mean subtraction, Variance normalization, and ARMA …

… 改造, 其目 的是得到说话人自适应(Speaker Adapted,SA) 码本以 提升识别性能.基于模型变换的
… 广泛使用的技术[12]包括最大 似然线性回归(MLLR)[13],分类自适应训练(CAT)和约 束最大似然…

… and the mean compensation bias, and fits the covariance matrix using a variance adaptive …
In the second system, we extract feature over variable sized windows of three or five times the …

… MLLR(Top1)為傳統只取用第一名辨識結果詞序列來做MLLR調適; MLLR(CM)為引入信心度評估
的MLLR… “Mean and Variance Adaptation within the MLLR Framework,” Computer Speech …

… Seide and Lin-Shan Lee; “Rapid speaker adaptation using a priori knowledge by
eigenspace analysis of MLLR parameters”. Acoustics, Speech, and Signal Processing, 2001. …