Speaker adaptation using multiple reference speakers

Abstract

We introduce a new technique for using the speech of multiple reference speakers as a basis for speaker adaptation in large vocabulary continuous speech recognition. In contrast to other methods that use a pooled reference model, this technique normalizes the training speech from multiple reference speakers to a single common feature space before pooling it. The normalized and pooled speech can then be treated as if it came from a single reference speaker for training the reference hidden Markov model (HMM). Our usual probabilistic spectrum transformation can be applied to the reference HMM to model a new (target) speaker. In this paper, we describe our baseline (single reference speaker) speaker-adaptation system and give current performance results from a recent formal evaluation of the system. We also describe our proposal for adapting from multiple reference speakers and report on recent preliminary experimental results in support of the proposed technique.