Structural speaker adaptation using maximum a posteriori approach and a Gaussian distributions merging technique

The aim of speaker adaptation techniques is to enhance the speaker-independent acoustic models to bring their recognition accuracy as close as possible to the one obtained with speaker-dependent models. Recently, a technique based on hierarchical structure and the maximum a posteriori criterion was proposed (SMAP). In this paper, like in SMAP, we assume that the acoustic models parameters are organized in a tree containing all the Gaussian distributions. Each node in that tree represents a cluster of Gaussian distributions sharing a common affine transformation representing the mismatch between training and test conditions. To estimate this affine transformation, we propose a new technique based on merging Gaussians and the standard MAP adaptation. This new technique is very fast and allows a good unsupervised adaptation for both means and variances even with small amount adaptation data. This adaptation strategy has shown a significant performance improvement in a large vocabulary speech recognition task, alone and combined with the MLLR adaptation.