Contextual Biasing with the Knuth-Morris-Pratt Matching Algorithm

TL;DR

This paper introduces a novel algorithm for contextual biasing in speech recognition using the KMP pattern matching algorithm, improving accuracy without extra model parameters.

Contribution

It proposes a KMP-based pattern matching approach for contextual biasing that enhances speech recognition accuracy efficiently on TPUs.

Findings

Significant WER reduction on biasing test sets

Compatible with model-based biasing methods

Efficient memory and computation performance

Abstract

Contextual biasing refers to the problem of biasing the automatic speech recognition (ASR) systems towards rare entities that are relevant to the specific user or application scenarios. We propose algorithms for contextual biasing based on the Knuth-Morris-Pratt algorithm for pattern matching. During beam search, we boost the score of a token extension if it extends matching into a set of biasing phrases. Our method simulates the classical approaches often implemented in the weighted finite state transducer (WFST) framework, but avoids the FST language altogether, with careful considerations on memory footprint and efficiency on tensor processing units (TPUs) by vectorization. Without introducing additional model parameters, our method achieves significant word error rate (WER) reductions on biasing test sets by itself, and yields further performance gain when combined with a model-based biasing method.