Internal Language Model Estimation based Adaptive Language Model Fusion for Domain Adaptation

TL;DR

This paper introduces ILME-ADA, an adaptive language model fusion method that improves domain-specific speech recognition performance with minimal impact on general domain accuracy by estimating internal language models.

Contribution

The paper proposes a novel ILME-ADA approach for adaptive language model fusion that effectively balances domain adaptation and general performance in speech recognition.

Findings

Significantly improves target domain speech recognition accuracy.

Maintains minimal performance degradation on general domain.

Effective with both RNN-T and LAS frameworks.

Abstract

ASR model deployment environment is ever-changing, and the incoming speech can be switched across different domains during a session. This brings a challenge for effective domain adaptation when only target domain text data is available, and our objective is to obtain obviously improved performance on the target domain while the performance on the general domain is less undermined. In this paper, we propose an adaptive LM fusion approach called internal language model estimation based adaptive domain adaptation (ILME-ADA). To realize such an ILME-ADA, an interpolated log-likelihood score is calculated based on the maximum of the scores from the internal LM and the external LM (ELM) respectively. We demonstrate the efficacy of the proposed ILME-ADA method with both RNN-T and LAS modeling frameworks employing neural network and n-gram LMs as ELMs respectively on two domain specific (target) test sets. The proposed method can achieve significantly better performance on the target test sets while it gets minimal performance degradation on the general test set, compared with both shallow and ILME-based LM fusion methods.