**6. Conclusions**

In this study, we have proposed the first end-to-end deep learning approach to stress recognition based on ECG and RESP data. Our protocol involved collecting ECG and RESP data and recording subjective stress scores while the subjects conducted alternating stressor and relaxation tasks. Using this multiple dataset, our proposed DeepER Net performed better than conventional machine learning models that require the extraction of handcrafted features. By visualizing the network's activation, we found that its neurons were being activated by unique and specific patterns. In conclusion, we believe that our proposed DeepER Net will be of benefit to people who suffer from stress in the workplace.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1424-8220/19/13/3021/ s1. Figure S1: Loss and accuracy information of the proposed DeepER Net during training, Table S1: The structure of the network [12] and its training condition, Table S2: The structure of the network [5] and its training condition.

**Author Contributions:** W.S. conceived and design this study. W.S. and N.K. performed these experiments; S.K. analyzed the collected data; W.S. and N.K. wrote the paper; C.L. and S.-M.P. revised this paper.

**Funding:** This research was supported by the Ministry of Science and ICT (MSIT), Korea, under the ICT Consilience Creative program (IITP-2019-2011-1-00783) supervised by the Institute for Information and communications Technology Promotion (IITP), the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science and ICT (NRF-2017R1A5A1015596), and the Technology Innovation Program (or Industrial Strategic Technology Development Program, 20001841, Development of System for Intelligent ContextAware Wearable Service based on Machine Learning) funded By the Ministry of Trade, Industry and Energy (MOTIE, Korea).

**Conflicts of Interest:** The authors declare no conflict of interest.

**Ethical Statements:** This study was approved by the Ethics Committee of POSTECH (PIRB-2019-E001).
