Learning Self-Informed Feature Contribution for Deep Learning-Based Acoustic Modeling

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 30
  • Download : 0
In this paper, we introduce a new feature engineering approach for deep learning-based acoustic modeling, which utilizes input feature contributions. For this purpose, we propose an auxiliary deep neural network (DNN) called a feature contribution network (FCN) whose output layer is composed of sigmoid-based contribution gates. In our framework, the FCN tries to learn element-level discriminative contributions of input features and an acoustic model network (AMN) is trained by gated features generated by element-wise multiplication between contribution gate outputs and input features. In addition, we also propose a regularization method for the FCN, which helps the FCN to activate the minimum number of the gates. The proposed methods were evaluated on the TED-LIUM release 1 corpus. We applied the proposed methods to DNN- and long short-term memory-based AMNs. Experimental results results showed that AMNs with the FCNs consistently improved recognition performance compared with AMN-only frameworks.
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Issue Date
2018-11
Language
English
Article Type
Article
Keywords

SPEECH RECOGNITION; NEURAL-NETWORKS; FEATURE-SELECTION; CLASSIFICATION

Citation

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, v.26, no.11, pp.2204 - 2214

ISSN
2329-9290
DOI
10.1109/TASLP.2018.2858923
URI
http://hdl.handle.net/10203/245868
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0