The Generation of Articulatory Animations Based on Keypoint Detection and Motion Transfer Combined with Image Style Transfer

Round 1

Reviewer 1 Report

This manuscript was focusing on  automatic key-points detection and style transfer, the topic looks interesting. My comments are as follows:

1) Both motivations and contributions are unclear in Abstract and Introduction, please refine them.

2) In the related work section, more comments on state-of-the-arts are suggested.

3) High-quality figures are strongly suggested to better demonstrate the proposed method and experimental results.

4) More baseline methods and evaluation metrics should be included to make the experiments more sufficient.

5) Separate discussion section should be considered to discuss the limitations of the proposed method and future directions.

This manuscript was focusing on  automatic key-points detection and style transfer, the topic looks interesting. My comments are as follows:

1) Both motivations and contributions are unclear in Abstract and Introduction, please refine them.

2) In the related work section, more comments on state-of-the-arts are suggested.

3) High-quality figures are strongly suggested to better demonstrate the proposed method and experimental results.

4) More baseline methods and evaluation metrics should be included to make the experiments more sufficient.

5) Separate discussion section should be considered to discuss the limitations of the proposed method and future directions.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript is devoted to the task of creating an animated video demonstrating articulation when pronouncing words and sentences. The created video is aimed to be used in the process of learning a foreign language. The authors proposed a way to automatically create animation from an MRI articulatory video with help of neural networks.

 

1) The Introduction is very short and does not reveal the motivation and purpose of the study. The sentence “For Chinese, it’s very difficult to pronounce…” is vague. It is not clear what words and in what language is difficult to pronounce. So it is recommended to improve the description of motivation, focusing on what exactly problem is being studied and on the importance of the results.

2) The work contribution must be explicitly written in the Introduction. Section 2.2 “Main innovation…” is close to the work contribution. But I recommend revisiting it to highlight your achievements more.

3) The manuscript does not report any comparison with existing solutions. This is a significant flaw of the study.

4) Given that only 6 volunteers participated in the evaluation of the created videos, the values presented in Table 1 (close to 50%) have no statistical value. It is not clear what the authors intend to prove with them.

5) The declared innovation “Considering the variations in stress and unstressed syllables, we adjust the duration of movements to create different animations for the same phoneme with different stress patterns” is not demonstrated or proven in the manuscript. 

6) The abbreviations should be revealed in the first place where they are mentioned (for example, EMA in the line 34, HMM in the line 48 and more). Also the appropriate references should be added to the list of networks in the line 95 and models in the line 103.

7) In the line 90 it is written “each channel in the figure”. What “figure” is mentioned here?

8) The Figure number is absent in the line 257. The Figure number is incorrect in line 286. Also “formula (1)” looks mistakenly in the line 291.

There are misprints in some places, commas are sometimes mixed with dots. Please check and correct them. For example, check the sentences in the lines 61-62, 139, 242, 245 and more.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

When I have checked the paper, the author is most satisfied the paper for requirements. However, it needs to check english grammar and journal structure. English grammar and structure were improving than previous version.

When I have checked the paper, the author is most satisfied the paper for requirements. However, it needs to check english grammar and journal structure. English grammar and structure were improving than previous version.

Reviewer 2 Report

The text of the manuscript has been improved. My remarks are properly reflected in the text and answered in the author’s reply.