A compact model for speaker-adaptive training
A compact model for speaker-adaptive training
复制标题
DOI：
10.1109/icslp.1996.607807
复制
发表时间：
1996-10
期刊：
影响因子：
--
通讯作者：
T. Anastasakos;J. McDonough;R. Schwartz;J. Makhoul
中科院分区：
文献类型：
--
作者：
T. Anastasakos;J. McDonough;R. Schwartz;J. Makhoul
我们制定了一个新的方法来估计参数的连续密度HSPER非特定人（SI）连续语音识别。它的动机是这样一个事实，即在SI声学模型的变化是由于语音的变化和训练人口的扬声器之间的变化，这是独立的语音信号的信息内容。这两个变化源是解耦的，所提出的方法联合消除扬声器间的变化，并估计HMM参数的SI声学模型。我们比较所提出的训练算法的监督适应的上下文中的共同SI训练范例。我们表明，所提出的声学模型更有效地适应测试扬声器，从而实现显着的整体字错误率降低19%和25%，分别为20K和5K词汇任务。
We formulate a novel approach to estimating the parameters of continuous density HMMs for speaker-independent (SI) continuous speech recognition. It is motivated by the fact that variability in SI acoustic models is attributed to both phonetic variation and variation among the speakers of the training population, that is independent of the information content of the speech signal. These two variation sources are decoupled and the proposed method jointly annihilates the inter-speaker variation and estimates the HMM parameters of the SI acoustic models. We compare the proposed training algorithm to the common SI training paradigm within the context of supervised adaptation. We show that the proposed acoustic models are more efficiently adapted to the test speakers, thus achieving significant overall word error rate reductions of 19% and 25% for 20K and 5K vocabulary tasks respectively.