Using Articulatory Information for Speaker Adaptation

Abstract

Articulatory Features (AF) have proven beneficial for Automatic Speech Recognition (ASR) in noisy environments, for hyper-articulated speech or in multi-lingual settings. A stream setup can combine standard sub-phone Gaussian Mixture Models with feature GMMs; the weights assigned to each feature stream such as VOICED or BILABIAL could intuitively be used… (More)

7 Figures and Tables

Topics

  • Presentations referencing similar topics