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

facebooktwittermailshare

CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION

Abstract: 

The greedy decoding method used in the conventional sequence-to-sequence models is prone to producing a model with a compounding
of errors, mainly because it makes inferences in a fixed order, regardless of whether or not the model’s previous guesses are correct.
We propose a non-sequential greedy decoding method that generalizes the greedy decoding schemes proposed in the past. The proposed
method determines not only which token to consider, but also which position in the output sequence to infer at each inference step.
Specifically, it allows the model to consider easy parts first, helping the model infer hard parts more easily later by providing more information. We study a grapheme-to-phoneme conversion task with a fully convolutional encoder-decoder model that embeds the proposed
decoding method. Experiment results show that our model shows better performance than that of the state-of-the-art model in
terms of both phoneme error rate and word error rate.

up
0 users have voted:

Paper Details

Authors:
Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park
Submitted On:
13 April 2018 - 12:22am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Moon-jung Chae
Paper Code:
MLSP-P3.1
Document Year:
2018
Cite

Document Files

NSGD_poster_at_ICASSP2018_v1.1.pdf

(189)

Subscribe

[1] Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park, "CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2586. Accessed: Aug. 25, 2019.
@article{2586-18,
url = {http://sigport.org/2586},
author = {Moon-jung Chae; Kyubyong Park; Jinhyun Bang; Soobin Suh; Jonghyuk Park; Namju Kim; Jonghun Park },
publisher = {IEEE SigPort},
title = {CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION},
year = {2018} }
TY - EJOUR
T1 - CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION
AU - Moon-jung Chae; Kyubyong Park; Jinhyun Bang; Soobin Suh; Jonghyuk Park; Namju Kim; Jonghun Park
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2586
ER -
Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park. (2018). CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION. IEEE SigPort. http://sigport.org/2586
Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park, 2018. CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION. Available at: http://sigport.org/2586.
Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park. (2018). "CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION." Web.
1. Moon-jung Chae, Kyubyong Park, Jinhyun Bang, Soobin Suh, Jonghyuk Park, Namju Kim, Jonghun Park. CONVOLUTIONAL SEQUENCE TO SEQUENCE MODEL WITH NON-SEQUENTIAL GREEDY DECODING FOR GRAPHEME TO PHONEME CONVERSION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2586