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

facebooktwittermailshare

DEEP FACTORIZATION FOR SPEECH SIGNAL

Abstract: 

Various informative factors mixed in speech signals, leading to great difficulty when decoding any of the factors. An intuitive idea is to factorize each speech frame into individual informative factors, though it turns out to be highly difficult. Recently, we found that speaker traits, which were assumed to be long-term distributional properties, are actually short-time patterns, and can be learned by a carefully designed deep neural network (DNN). This discovery motivated a cascade deep factorization (CDF) framework that will be presented in this paper. The proposed framework infers speech factors in a sequential way, where factors previously inferred are used as conditional variables when inferring other factors. We will show that this approach can effectively factorize speech signals, and using these factors, the original speech spectrum can be recovered with a high accuracy. This factorization and reconstruction approach provides potential values for many speech processing tasks, e.g., speaker recognition and emotion recognition, as will be demonstrated in the paper.

up
0 users have voted:

Paper Details

Authors:
Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng
Submitted On:
20 April 2018 - 7:42am
Short Link:
Type:
Poster
Event:
Presenter's Name:
Miao Zhang
Paper Code:
2969
Document Year:
2018
Cite

Document Files

180417-deepFactor-LLT.pptx

(56 downloads)

Subscribe

[1] Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng, "DEEP FACTORIZATION FOR SPEECH SIGNAL", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3101. Accessed: Aug. 19, 2018.
@article{3101-18,
url = {http://sigport.org/3101},
author = {Lantian Li; Dong Wang; Yixiang Chen; Ying Shi; Zhiyuan Tang; Thomas Fang Zheng },
publisher = {IEEE SigPort},
title = {DEEP FACTORIZATION FOR SPEECH SIGNAL},
year = {2018} }
TY - EJOUR
T1 - DEEP FACTORIZATION FOR SPEECH SIGNAL
AU - Lantian Li; Dong Wang; Yixiang Chen; Ying Shi; Zhiyuan Tang; Thomas Fang Zheng
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3101
ER -
Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng. (2018). DEEP FACTORIZATION FOR SPEECH SIGNAL. IEEE SigPort. http://sigport.org/3101
Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng, 2018. DEEP FACTORIZATION FOR SPEECH SIGNAL. Available at: http://sigport.org/3101.
Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng. (2018). "DEEP FACTORIZATION FOR SPEECH SIGNAL." Web.
1. Lantian Li, Dong Wang, Yixiang Chen, Ying Shi, Zhiyuan Tang, Thomas Fang Zheng. DEEP FACTORIZATION FOR SPEECH SIGNAL [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3101