A Preliminary Investigation on Frequency Dependant Cues for Human Emotions
Round 1
Reviewer 1 Report
Areas of improvement
- I suggest you can provide the listening duration for 1 frequency sound as such readers can imagine how the ratings were conducted.
- The size of samples can be bigger instead of 30 samples per frequency. Referring to the emotion class - happy, the height of the median line of the box at 417 Hz is almost the same as the entire box at 440 Hz, which means there is no sharp or clear likelihood to show a difference between the two groups. So, the more data you can obtain, the more accurate the result would be.
- To clarify the specific preference on frequency choice, you can put some rules on picking the preferred tone frequency. You have cited the reference that the most preferred tones range from 400 to 750 Hz. However, it is observed that the emotion class - anger is triggered starting from 440 Hz to 528 Hz and your preferred tone is from 210 to 540 Hz. So, what is your underlying principle that includes the anger triggering frequency?
Overall, your paper provides evidence of the results with novel ideas in analyzing frequency range for emotion cues. I particularly appreciate the details and the clarity that you have mentioned and cross-referenced to, where I believe readers can fully understand the significance of your findings and the contribution to the emotion cues approaches.
Author Response
This is the response to the reviewers’ comments on the manuscript (acoustics-1696040) submitted to MDPI acoustics special issues on Human’s Psychological and Physiological Responses to Sound Environment.
We would like to thank the reviewers for their time in giving constructive feedback on our manuscript. In this document, we have addressed each question/comment separately with a detailed response, along with references to the changes made in the revised manuscript. The major changes from the original manuscript are highlighted in red in the revised manuscript for convenience. Please note that due to the advised modifications, there have been a few changes in the figures and section numbers of the revised manuscript, and all the references we have made in our responses correspond to the revised manuscript unless explicitly mentioned otherwise. We sincerely hope that our attempts to improve the manuscript have done justice to the time and effort spent by the reviewers.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The paper is devoted to a few aspects that are associated with the definition of human emotions. Namely, the establishment of relationships between the frequencies of pure tones and reproducible emotional reactions of a person. The authors of the paper are trying to convey the idea that despite great practical potential, the problem of effective emotion recognition has not yet been solved. Therefore, we can say that currently operating fully automated models and methods capable of determining the emotional state of a person with a high degree of probability do not currently exist in a full-fledged form. To create such full-fledged models, it is necessary to perform a deep analysis of acoustic features that could clearly determine a person's emotion at a certain point in time. To solve the set goal of the study (establishing the relationship between the frequencies of pure tones and reproducible human emotions), the authors of the paper conducted online testing for the perception of various types of pure tones in the range from 110 to 963 Hz. This experiment involved 30 people aged 21 to 55 years. The gender distribution among the participants was the same (15 males and 15 females) and the mean age was 27 years. Comparative analysis showed the relationship between certain classes of human emotions and the sound frequency in Hz. However, the paper is not free from a lot of shortcomings. First, it is rather difficult to assess the validity of the main conclusions of the paper and their compliance with the world level, since the authors of the paper briefly described the main problems associated with the recognition of emotions, and the section with related work of the world scientific community is completely absent. Works related to the subject of emotion recognition, both in audio and video modalities, are constantly presented at such conferences as: ICASSP, INTERSPEECH, CVPR and others. Further, as stated in the paper, the average age of people who participated in online testing is 27 years old, if the maximum age is 55 years old, this may indicate that the sample is dominated by a young age group and this may lead to some bias in the perception of different types of pure tones in the specified range. However, the authors of the paper did not take this into account. In section 3, the authors of the paper write: "For that, we only consider primary emotions - Happy, Sad, Anger and Joy", where is Joy Calm? The list of references is mainly represented by rather old works, which once again emphasizes the need for a section with a detailed analysis of previous studies. It seems to me that all the proposed additions and comments will improve this paper. Finally, the style of the paper requires minor revision due to the presence of spelling, punctuation, and typographical errors, such as in section 3.
In general, a paper is only interesting if it is improved.
Author Response
This is the response to the reviewers’ comments on the manuscript (acoustics-1696040) submitted to MDPI acoustics special issues on Human’s Psychological and Physiological Responses to Sound Environment.
We would like to thank the reviewers for their time in giving constructive feedback on our manuscript. In this document, we have addressed each question/comment separately with a detailed response, along with references to the changes made in the revised manuscript. The major changes from the original manuscript are highlighted in red in the revised manuscript for convenience. Please note that due to the advised modifications, there have been a few changes in the figures and section numbers of the revised manuscript, and all the references we have made in our responses correspond to the revised manuscript unless explicitly mentioned otherwise. We sincerely hope that our attempts to improve the manuscript have done justice to the time and effort spent by the reviewers.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors of the paper finalized the paper and made all the necessary changes. In this form, the paper may be useful to many scientific readers who are interested in the topic of automatic detection of human emotions. In general, the paper can be accepted for publication.
