UBM-based incremental speaker adaptation

This paper addresses a novel algorithm of incremental speaker adaptation (ISA) based on universal background model (UBM) for saving storage and real-time processing. This algorithm can be seen as an extension of traditional speaker adaptation. It consists of two steps, adaptation and combination. It not only considers the speaker´s characteristics in limited training data, but also prohibits over-fitting of the updated model. The incremental adaptation algorithm needs little storage and meets the requirement of real-time processing. In order to evaluate the efficiency and effectivity of the proposed approach, a real-time speaker segmentation system for broadcasting news is built. Experiment results demonstrate that our approach yields real time operation and achieves satisfactory performance.