Next Article in Journal
Degradation Products on Byzantine Glasses from Northern Tunisia
Next Article in Special Issue
Multimodal Few-Shot Learning for Gait Recognition
Previous Article in Journal
Morphological and Chemical Characterization of Titanium and Zirconia Dental Implants with Different Macro- and Micro-Structure
Previous Article in Special Issue
Induction Motor Fault Classification Based on FCBF-PSO Feature Selection Method
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 10
Issue 21
10.3390/app10217522
Review Report
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Robust Deep Speaker Recognition: Learning Latent Representation with Joint Angular Margin Loss

Appl. Sci. 2020, 10(21), 7522; https://doi.org/10.3390/app10217522
by Labib Chowdhury 1, Hasib Zunair 2 and Nabeel Mohammed 1,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Sci. 2020, 10(21), 7522; https://doi.org/10.3390/app10217522
Submission received: 5 October 2020 / Revised: 18 October 2020 / Accepted: 21 October 2020 / Published: 26 October 2020
(This article belongs to the Special Issue Advances in Pattern Analysis for Identity Recognition and Verification)

Round 1

Reviewer 1 Report

The article is interesting. The process consisting of defining the research problem, proposing its solution and presenting the obtained results seems to be carried out correctly. The review of previous work is brief but sufficient. Therefore, this aspect of the paper is evaluated positively, because this is not a review article. On the other hand, I would like to make it clear that the main contribution of the authors is the skilful combination of previously known formulas. From this perspective, they demonstrated knowledge of the available methods and the ability to increase the efficiency of mechanisms by combining them. However, as an article in the Applied Sciences area, it is acceptable.

Certainly the paper requires significant improvement formatting!

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper raises a problem connected with a speaker recognition. The issue is very popular and important especially from viewpoint of man recognition in different aspect of investigations. Authors proposed new approach. They proposed three models AF-SincNet, Ensemble-SincNet and ALL-SincNet. Next these models were compared with other different models in this area. Authors obtained promising results which were presented in a form of diagrams and tables.

Generally the paper is good organised. However it includes several mistakes - scientific and editorial.

The paper has to be written in passive voice. Conclusion part is too short. This should include more information about obtained results which are very extensive but not always in my opinion expected by authors. Proposed models aren't better than other in each case (e.g. a table 3, figure 3). Why? How to select an appropriate model?

Line 208 - 16 khz - better 16 kHz;

Author should use a template of Applied Sciences journal (e.g. locations of table captions);

Table 4 - suddenly AM-Softmax appears. Why?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Dear Authors

Thank you for responses. The explanations are good and doubtless a scientific level of this paper was corrected.
Taking into account an editorial side of the paper you can use a template in MS Word. Then you will notice differences between your paper and the template.

Good luck

A reviewer

Back to TopTop