Table of Contents Author Guidelines Submit a Manuscript
Advances in Multimedia
Volume 2018, Article ID 3748141, 13 pages
Research Article

Research on Objective Evaluation of Recording Audio Restoration Based on Deep Learning Network

1Key Laboratory of Media Audio & Video, Communication University of China, Beijing, China
2College of Science and Technology, Communication University of China, Beijing, China
3Advertising School, Communication University of China, Beijing, China

Correspondence should be addressed to Cong Jin; nc.ude.cuc@3260gnocnij

Received 13 March 2018; Accepted 13 June 2018; Published 18 September 2018

Academic Editor: Zechao Li

Copyright © 2018 Cong Jin et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.


There are serious distortion problems in the history audio and video data. In view of the characteristics of audio data repair, the intelligent technology of audio evaluation is explored. As the traditional audio subjective evaluation method requires a large number of personal to audition and evaluation, the tester’s subjective sense of hearing deviation and sample space data limited the impact of the accuracy of the experiment. Based on the deep learning network, this paper designs an objective quality evaluation system for historical audio and video data and evaluates the performance of the system and the audio signal quality from the perspective of feature extraction and network parameter selection. Experiments show that the system has good performance in this experiment; the predictive results and subjective evaluation of the correlation and dispersion indicators are good, up to 0.91 and 0.19.