Synthesis of Non-Linguistic Utterances for Sound Design Support Using a Genetic Algorithm

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

This paper is clear and well written, and covers an important topic. 

 

My main concern is the fact the paper talks about being designed for robotics, but there’s actually no robots or application of the sounds to robots. It could be directly about sounds for any software system. I think this needs to at least be addressed and explained how there is the contrast between starting the paper in the abstract describing social robots, yet they are not really connected to the main work.

 

 

 

 

There’s been a big increase in work on sound and robotics in the last 5 years that is missing from the related work. There are some papers that are similar and omitted. The related work seems a few years out of date really.

 

Directly related are:

 

• Short emotion tagged phrases, omitting seems like a big miss

Savery, Richard, Amit Rogel, and Gil Weinberg. "Emotion musical prosody for robotic groups and entitativity." 2021 30th IEEE International Conference on Robot & Human Interactive Communication (RO-MAN). IEEE, 2021.

 

• Audio generation

Robinson, Frederic Anthony, Mari Velonaki, and Oliver Bown. "Smooth operator: Tuning robot perception through artificial movement sound." Proceedings of the 2021 ACM/IEEE international conference on human-robot interaction. 2021

 

• Sound in long-term real interactions

Pelikan, Hannah RM, and Malte F. Jung. "Designing robot sound-in-interaction: The case of autonomous public transport shuttle buses." Proceedings of the 2023 ACM/IEEE International Conference on Human-Robot Interaction. 2023.

 

Author Response

Thank you very much for taking the time to review my work. Your comments and suggestions are much appreciated and I believe they will help me greatly to improve the paper. Regarding the first comment about addressing the lack of follow up in the paper after suggesting the application of this method to social robots, I decided to add the following paragraph in the introduction to explain it more clearly:

"This work presents a robust method of automatic emotionally expressive sound generation that is intended to be used in social robots. While the experiments conducted in this study did not involve physical robots, the generated sounds are specifically designed for application to social robots with the goal of improving their interactive capabilities. This intention is represented in the work by the use of sounds sampled from social robots to help infer the valence and arousal of the generated sounds. Although the method and the experiment can be extended to almost any other use case for which sounds are being developed, the methodology and findings presented here lay the groundwork for future applications where these sounds can be dynamically utilized by social robots in real-world settings. Application of this method to social robots would enable them to effectively convey emotions through sound."

Thank you also for highlighting three important and interesting references that I should have considered. After reviewing them all, I decided that I should mention the work of Savery et al. from the first mentioned reference "Emotion musical prosody for robotic groups and entitativity." as it highlights a promising application and clear need for these sounds for social robots. I did not include mention of the other two references in the end as I felt that they might be slightly out of the scope that I am considering. Here is what I included from the first reference:

"Savery et al. explored the effect of musical prosody on interactions between humans and groups of robots. They introduced the concept of entitativity, meaning the perception of the group as being a single entity, and found that alterations of prosodic features of musical sounds increased both the likeability and trust of the robots as well as the entitativity. The results of this study suggest that NLU's can also improve interactions between humans and groups of robots."

Reviewer 2 Report

Comments and Suggestions for Authors

The paper proposes a method of sound generation using a genetic algorithm. The sounds were tested in an experiment where subjects rated the perceived valences and arousal. One of the authors previous work developed a model capable of inferring valence/arousal from NLU. This previous work is the random forest tree used here. 

 

The paper is well written, the work is well presented and properly described. The statistical analysis of the results are thorough as well as the subsequent discussion. 

 

I have some reservations in discussing the application to social robotics because the results are solely on human experiments. Although the authors mention related work on NLU for social robotics. Other than that, the paper is good. 

 

Author Response

Thank you very much for taking the time to review my work and for the comments you have provided.