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

facebooktwittermailshare

A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification

Abstract: 

A novel learnable dictionary encoding layer is proposed in this paper for end-to-end language identification. It is inline with the conventional GMM i-vector approach both theoretically and practically. We imitate the mechanism of traditional GMM training and Supervector encoding procedure on the top of CNN. The proposed layer can accumulate high-order statistics from variable-length input sequence and generate an utterance level fixed-dimensional vector representation. Unlike the conventional methods, our new approach provides an end-to-end learning framework, where the inherent dictionary are learned directly from the loss function. The dictionaries and the encoding representation for the classifier are learned jointly. The representation is orderless and therefore appropriate for language identification. We conducted a preliminary experiment on NIST LRE07 closed-set task, and the results reveal that our proposed dictionary encoding layer achieves significant error reduction comparing with the simple average pooling.

up
0 users have voted:

Paper Details

Authors:
Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li
Submitted On:
13 April 2018 - 9:37am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Weicheng Kai
Paper Code:
3889
Document Year:
2018
Cite

Document Files

poster_weichcai_icassp2018_lde.pdf

(131)

Subscribe

[1] Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li, "A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2701. Accessed: Jun. 16, 2019.
@article{2701-18,
url = {http://sigport.org/2701},
author = {Weicheng Cai; Zexin Cai; Xiang Zhang; Xiaoqi Wang; Ming Li },
publisher = {IEEE SigPort},
title = {A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification},
year = {2018} }
TY - EJOUR
T1 - A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification
AU - Weicheng Cai; Zexin Cai; Xiang Zhang; Xiaoqi Wang; Ming Li
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2701
ER -
Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li. (2018). A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification. IEEE SigPort. http://sigport.org/2701
Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li, 2018. A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification. Available at: http://sigport.org/2701.
Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li. (2018). "A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification." Web.
1. Weicheng Cai, Zexin Cai, Xiang Zhang, Xiaoqi Wang, Ming Li. A Novel Learnable Dictionary Encoding Layer for End-to-End Language Identification [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2701