Approaches to Improving Recognition of Underrepresented Named Entities in Hybrid ASR Systems

TL;DR

This paper introduces multiple complementary methods, including graphemic lexicons and neural language models, to significantly improve recognition of underrepresented named entities in hybrid ASR systems without harming overall accuracy.

Contribution

It proposes novel techniques such as graphemic lexicons and embedding enrichment to enhance underrepresented NE recognition in hybrid ASR systems.

Findings

Up to 42% relative improvement in NE recognition

Effective handling of OOV and rare words in hybrid ASR

Neural LM rescoring boosts NE detection accuracy

Abstract

In this paper, we present a series of complementary approaches to improve the recognition of underrepresented named entities (NE) in hybrid ASR systems without compromising overall word error rate performance. The underrepresented words correspond to rare or out-of-vocabulary (OOV) words in the training data, and thereby can't be modeled reliably. We begin with graphemic lexicon which allows to drop the necessity of phonetic models in hybrid ASR. We study it under different settings and demonstrate its effectiveness in dealing with underrepresented NEs. Next, we study the impact of neural language model (LM) with letter-based features derived to handle infrequent words. After that, we attempt to enrich representations of underrepresented NEs in pretrained neural LM by borrowing the embedding representations of rich-represented words. This let us gain significant performance improvement on underrepresented NE recognition. Finally, we boost the likelihood scores of utterances containing NEs in the word lattices rescored by neural LMs and gain further performance improvement. The combination of the aforementioned approaches improves NE recognition by up to 42% relatively.