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

facebooktwittermailshare

Tuplemax Loss for Language Identification

Abstract: 

In many scenarios of a language identification task, the user will specify a small set of languages which he/she can speak instead of a large set of all possible languages. We want to model such prior knowledge into the way we train our neural networks, by replacing the commonly used softmax loss function with a novel loss function named tuplemax loss. As a matter of fact, a typical language identification system launched in North America has about 95% users who could speak no more than two languages. Using the tuplemax loss, our system achieved a 2.33% error rate, which is a relative 39.4% improvement over the 3.85% error rate of standard softmax loss method.

up
0 users have voted:

Paper Details

Authors:
Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno
Submitted On:
24 April 2019 - 11:03am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Quan Wang
Paper Code:
3793
Document Year:
2019
Cite

Document Files

poster

(27)

Subscribe

[1] Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno, "Tuplemax Loss for Language Identification", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3896. Accessed: Jun. 16, 2019.
@article{3896-19,
url = {http://sigport.org/3896},
author = {Li Wan; Prashant Sridhar; Yang Yu; Quan Wang; Ignacio Lopez Moreno },
publisher = {IEEE SigPort},
title = {Tuplemax Loss for Language Identification},
year = {2019} }
TY - EJOUR
T1 - Tuplemax Loss for Language Identification
AU - Li Wan; Prashant Sridhar; Yang Yu; Quan Wang; Ignacio Lopez Moreno
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3896
ER -
Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno. (2019). Tuplemax Loss for Language Identification. IEEE SigPort. http://sigport.org/3896
Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno, 2019. Tuplemax Loss for Language Identification. Available at: http://sigport.org/3896.
Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno. (2019). "Tuplemax Loss for Language Identification." Web.
1. Li Wan, Prashant Sridhar, Yang Yu, Quan Wang, Ignacio Lopez Moreno. Tuplemax Loss for Language Identification [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3896