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

facebooktwittermailshare

Robust Spoken Language Understanding with unsupervised ASR-error adaptation

Abstract: 

Robustness to errors produced by automatic speech recognition (ASR) is essential for Spoken Language Understanding (SLU). Traditional robust SLU typically needs ASR hypotheses with semantic annotations for training. However, semantic annotation is very expensive, and the corresponding ASR system may change frequently. Here, we propose a novel unsupervised ASR-error adaptation method, obviating the need of annotated ASR hypotheses. It only requires semantically annotated transcripts for the slot-tagging task and the transcripts paired with hypotheses for an input sentence reconstruction task. In this method, feature encoders which share part of the parameters are exploited to enforce the tasks in a similar feature space. Therefore, the transcript side slot-tagging model can be transferred to ASR hypotheses side easily. Experiments show that the proposed approach can yield significant improvement over strong baselines, and achieve performance very close to the oracle system.

up
0 users have voted:

Paper Details

Authors:
Su Zhu, Ouyu Lan, Kai Yu
Submitted On:
19 April 2018 - 3:58pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Su Zhu
Paper Code:
HLT-P3.1
Document Year:
2018
Cite

Document Files

zhu-icassp18-poster.pdf

(124 downloads)

Subscribe

[1] Su Zhu, Ouyu Lan, Kai Yu, "Robust Spoken Language Understanding with unsupervised ASR-error adaptation", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3016. Accessed: Oct. 16, 2018.
@article{3016-18,
url = {http://sigport.org/3016},
author = {Su Zhu; Ouyu Lan; Kai Yu },
publisher = {IEEE SigPort},
title = {Robust Spoken Language Understanding with unsupervised ASR-error adaptation},
year = {2018} }
TY - EJOUR
T1 - Robust Spoken Language Understanding with unsupervised ASR-error adaptation
AU - Su Zhu; Ouyu Lan; Kai Yu
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3016
ER -
Su Zhu, Ouyu Lan, Kai Yu. (2018). Robust Spoken Language Understanding with unsupervised ASR-error adaptation. IEEE SigPort. http://sigport.org/3016
Su Zhu, Ouyu Lan, Kai Yu, 2018. Robust Spoken Language Understanding with unsupervised ASR-error adaptation. Available at: http://sigport.org/3016.
Su Zhu, Ouyu Lan, Kai Yu. (2018). "Robust Spoken Language Understanding with unsupervised ASR-error adaptation." Web.
1. Su Zhu, Ouyu Lan, Kai Yu. Robust Spoken Language Understanding with unsupervised ASR-error adaptation [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3016