Sorry, you need to enable JavaScript to visit this website.

REDAT: ACCENT-INVARIANT REPRESENTATION FOR END-TO-END ASR BY DOMAIN ADVERSARIAL TRAINING WITH RELABELING

Primary tabs

Citation Author(s):
Hu Hu, Xuesong Yang, Zeynab Raeesy, Jinxi Guo, Gokce Keskin, Harish Arsikere, Ariya Rastrow, Andreas Stolcke, Roland Maas
Submitted by:
Hu Hu
Last updated:
21 June 2021 - 5:48pm
Document Type:
Poster
Document Year:
2021
Event:
Presenters Name:
Hu Hu
Paper Code:
2760
Categories:

Abstract 

Abstract: 

Accents mismatching is a critical problem for end-to-end ASR. This paper aims to address this problem by building an accent-robust RNN-T system with domain adversarial training (DAT). We unveil the magic behind DAT and provide, for the first time, a theoretical guarantee that DAT learns accent-invariant representations. We also prove that performing the gradient reversal in DAT is equivalent to minimizing the Jensen-Shannon divergence between domain output distributions. Motivated by the proof of equivalence, we introduce reDAT, a novel technique based on DAT, which relabels data using either unsupervised clustering or soft labels. Experiments on 23K hours of multi-accent data show that DAT achieves competitive results over accent-specific baselines on both native and non-native English accents but up to 13% relative WER reduction on unseen accents; our reDAT yields further improvements over DAT by 3% and 8% relatively on non-native accents of American and British English.

up
0 users have voted:

Dataset Files

ICASSP2760poster.pdf

(38)

ICASSP2760slides.pdf

(51)