Contextual Biasing Methods for Improving Rare Word Detection in Automatic Speech Recognition

Contextual Biasing Methods for Improving Rare Word Detection in Automatic Speech Recognition

Stať ve sborníku v databázi WoS či Scopus

In specialized domains like Air Traffic Control (ATC), a notable challenge in porting a deployed Automatic Speech Recognition (ASR) system from one airport to another is the alteration in the set of crucial words that must be ac- curately detected in the new environment. Typically, such words have limited occurrences in training data, making it impractical to retrain the ASR system. This paper explores innovative word-boosting techniques to improve the detec- tion rate of such rare words in the ASR hypotheses for the ATC domain. Two acoustic models are investigated: a hybrid CNN-TDNNF model trained from scratch and a pre-trained wav2vec2-based XLSR model fine-tuned on a common ATC dataset. The word boosting is done in three ways. First, an out-of-vocabulary word addition method is explored. Second, G-boosting is explored, which amends the language model before building the decoding graph. Third, the boosting is performed on the fly during decoding using lattice re-scoring. The results indicate that the G-boosting method performs best and provides an approximately 30-43% relative improvement in recall of the boosted words. Moreover, a relative improve- ment of up to 48% is obtained upon combining G-boosting and lattice-rescoring

In specialized domains like Air Traffic Control (ATC), a notable challenge in porting a deployed Automatic Speech Recognition (ASR) system from one airport to another is the alteration in the set of crucial words that must be ac- curately detected in the new environment. Typically, such words have limited occurrences in training data, making it impractical to retrain the ASR system. This paper explores innovative word-boosting techniques to improve the detec- tion rate of such rare words in the ASR hypotheses for the ATC domain. Two acoustic models are investigated: a hybrid CNN-TDNNF model trained from scratch and a pre-trained wav2vec2-based XLSR model fine-tuned on a common ATC dataset. The word boosting is done in three ways. First, an out-of-vocabulary word addition method is explored. Second, G-boosting is explored, which amends the language model before building the decoding graph. Third, the boosting is performed on the fly during decoding using lattice re-scoring. The results indicate that the G-boosting method performs best and provides an approximately 30-43% relative improvement in recall of the boosted words. Moreover, a relative improve- ment of up to 48% is obtained upon combining G-boosting and lattice-rescoring

Automatic speech recognition, air traffic control, domain adaptation, contextual biasing, rare word recognition

Automatic speech recognition, air traffic control, domain adaptation, contextual biasing, rare word recognition

BHATTACHARJEE, M.; NIGMATULINA, I.; PRASAD, A.; RANGAPPA, P.; MADIKERI, S.; MOTLÍČEK, P.; HELMKE, H.; KLEINERT, M.

2025

14.04.2024

IEEE Signal Processing Society

Seoul

979-8-3503-4485-1

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

12652

12656

5

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10447465

bhattacharjee_icassp2024_Contextual_Biasing_spoluautorka Amrutha