Interactive Feature Fusion for End-to-End Noise-Robust Speech Recognition
TL
TL;DR
An interactive feature fusion network (IFF-Net) is proposed for noise-robust speech recognition to learn complementary information from the enhanced feature and original noisy feature to complement some missing information in the over-suppressed enhanced feature.
Abstract
Speech enhancement (SE) aims to suppress the additive noise from noisy speech signals to improve the speech’s perceptual quality and intelligibility. However, the over-suppression phenomenon in the enhanced speech might degrade the performance of downstream automatic speech recognition (ASR) task due to the missing latent information. To alleviate such problem, we propose an interactive feature fusion network (IFF-Net) for noise-robust speech recognition to learn complementary information from the enhanced feature and original noisy feature. Experimental results show that the proposed method achieves absolute word error rate (WER) reduction of 4.1% over the best baseline on RATS Channel-A corpus. Our further analysis indicates that the proposed IFF-Net can complement some missing information in the over-suppressed enhanced feature.
References28 items
1
Multitask-based joint learning approach to robust ASR for radio communication speech
2
Learning Disentangled Feature Representations for Speech Enhancement Via Adversarial Training
3
Interactive Speech and Noise Modeling for Speech Enhancement
4
Gated Recurrent Fusion With Joint Training Framework for Robust End-to-End Speech Recognition
5