SER-SAM Alignment Dataset and Analysis Scripts

This repository contains the resources required to reproduce the analyses reported in the paper "Do Emotion Mappings Matter? Evaluating Representation and Interaction Context in SER Alignment with Self-Reported Affect", accepted at the XXVI International Conference on Human-Computer Interaction (INTERACCIÓN 2026). The repository includes the derived datasets used in the study, comprising the outputs from the Speech Emotion Recognition (SER) pipeline and the anonymised questionnaire responses used in the analyses. To facilitate reproducibility, it also provides the Python notebooks and analysis scripts used to answer the research questions and generate the reported results. Original audio recordings are not included in the repository due to ethical and privacy considerations. Only derived data and analysis artefacts necessary to reproduce the published findings are publicly shared.