N-best list re-ranking using semantic relatedness and syntactic score: An approach for improving speech recognition accuracy in air traffic control

Abstract

In this paper, we investigate how we can take advantage of the availability of linguistic knowledge, particularly semantic knowledge, in Air Traffic Control (ATC) to reduce the Word Error Rate (WER) of Automatic Speech Recognition (ASR) systems. To facilitate this, we integrate semantic knowledge into post-processing by performing n-best list re-ranking. We first propose a feature called semantic relatedness. We then use the WER-Sensitive Pairwise Perceptron algorithm which is proposed in previous work to combine the semantic relatedness, syntactic score and speech decoder's confidence score features to perform n-best list re-ranking. We evaluate the proposed approach in terms of WER on the well known ATCOSIM Corpus of Non-prompted Clean Air Traffic Control Speech (ATCOSIM) and our own Air Traffic Control Speech Corpus (ATCSC). The evaluation results show that our proposed approach reduces the WER by 0.31% and 1.53% on the ATCOSIM and ATCSC corpora respectively. Our proposed approach also shows 19.93% WER improvement compared with traditional n-gram model on the ATCSC corpus.