Robust feature space adaptation for telephony speech recognition
Robust feature space adaptation for telephony speech recognition
复制标题
用于电话语音识别的鲁棒特征空间自适应
DOI：
10.21437/interspeech.2006-268
复制
发表时间：
2006
期刊：
影响因子：
--
通讯作者：
Xiaodong He
中科院分区：
文献类型：
--
作者：
X. Lei;J. Hamaker;Xiaodong He
说话人自适应是现代语音识别系统的关键。由于计算和多通道模型共享的考虑，在电话语音识别系统中，模型自适应技术的使用受到限制。另一方面，诸如特征空间最大似然线性回归（fMLLR）之类的特征空间自适应方法是适用于电话系统的有效方法。在这项工作中，我们首先描述了有效实现在线fMLLR自适应的技术。然后提出了特征空间最大后验线性回归（fMAPLR）方法，将先验知识引入到特征变换估计中，提高了传统fMLLR方法的鲁棒性。在电话数据上的实验表明，fMAPLR比fMLLR更鲁棒，并且优于fMLLR，特别是当自适应数据非常有限时。
Speaker adaptation is critical for modern speech recognition sys-tems. Due to the computational and multi-channel model sharing considerations, the use of model adaptation techniques is limited in telephony speech recognition systems. On the other hand, feature space adaptation methods such as feature space maximum likelihood linear regression (fMLLR) are efficient approaches suitable for telephony systems. In this work, we first describe techniques for efficient implementation of online fMLLR adaptation. Then feature space maximum a posteriori linear regression (fMAPLR) is proposed to incorporate prior knowledge for the feature transform estimation and improve the robustness of the conventional fMLLR approach. Experiments on telephony data indicate that fMAPLR is significantly more robust than fMLLR, and outperforms fMLLR especially when the adaptation data is very limited.
影响因子：
4.3
作者：
LEGGETTER, CJ;WOODLAND, PC
通讯作者：
WOODLAND, PC