Improving ASR Error Correction Using N-Best Hypotheses

In the field of Automatic Speech Recognition (ASR), Grammatical Error Correction (GEC) can be used to correct errors in recognition results of ASR systems and whereby it further reduces the word error rate (WER). Most conventional GEC approaches make correction merely based on a single recognition hypothesis from the ASR system, i.e., the best path in the decoding lattice. However, with such limited information, it's common that GEC fails to detect errors in a sentence or even makes wrong corrections. In this paper we propose two novel GEC approaches utilizing N-best hypotheses instead, that's, N-best plain approach and N-best alignment approach. In this way, our novel approaches can access more information than conventional approaches with 1-best-only hypothesis, and thus more accurate corrections can be made. Experiments on an industrial-scale internal dataset and the open-source dataset AISHELL-1 demonstrate the effectiveness of N-best approaches: 1) both N-best plain and N-best alignment approaches achieve remarkable WER reduction. N-best alignment approach achieves the best performance on both datasets, 12.54% WERR on an internal dataset and 27.71% WERR on AISHELL-1. 2) They also outperform the baseline 1-best GEC approach by a large margin.

1. Introduction

In recent years, with the improvement of GPU computing power and the development of the sequence-to-sequence (seq2seq) models, e.g., transformer [1] and ConvS2S model [2], seq2seq-based grammatical error correction (GEC) has attracted more and more attention. In the field of Automatic Speech Recognition (ASR), GEC can be used to correct errors in ASR transcribed results. GEC for ASR can be modeled as a machine translation task where seq2seq models take decoder hypotheses as inputs and ground truth transcriptions as target outputs, and seq2seq models learn to translate decoder hypotheses to error-free transcriptions. Many works have shown that GEC can further reduce the word error rate (WER) of ASR output remarkably [3]–[5].