Contextual Language Model Adaptation for Conversational Agents

TL;DR

This paper introduces a DNN-based method for adapting language models in conversational agents using contextual information, leading to notable improvements in speech recognition accuracy and named entity recognition.

Contribution

The paper presents a novel framework for context-dependent language model adaptation using neural networks, enhancing ASR performance in conversational agents.

Findings

3% relative WER improvement with 1-pass decoding

6% relative WER improvement with 2-pass decoding

Up to 15% improvement in named entity recognition

Abstract

Statistical language models (LM) play a key role in Automatic Speech Recognition (ASR) systems used by conversational agents. These ASR systems should provide a high accuracy under a variety of speaking styles, domains, vocabulary and argots. In this paper, we present a DNN-based method to adapt the LM to each user-agent interaction based on generalized contextual information, by predicting an optimal, context-dependent set of LM interpolation weights. We show that this framework for contextual adaptation provides accuracy improvements under different possible mixture LM partitions that are relevant for both (1) Goal-oriented conversational agents where it's natural to partition the data by the requested application and for (2) Non-goal oriented conversational agents where the data can be partitioned using topic labels that come from predictions of a topic classifier. We obtain a relative WER improvement of 3% with a 1-pass decoding strategy and 6% in a 2-pass decoding framework, over an unadapted model. We also show up to a 15% relative improvement in recognizing named entities which is of significant value for conversational ASR systems.