This task introduces a new modality in the AER challenge by incorporating audio. The objective is to integrate text and voice signals to determine the emotion conveyed in each segment, based on five of Ekman's six basic emotions: anger, disgust, fear, joy, and sadness, as well as a neutral emotion.
Publication
Pan et al. (2024). Overview of EmoSPeech at IberLEF 2024: Multimodal Speech-text Emotion Recognition in Spanish. Procesamiento del Lenguaje Natural, Revista, 73: 359-368.
Language
Spanish
NLP topic
Dataset
Year
2024
Publication link
Task results
| System | MacroF1 Sort ascending |
|---|---|
| BSCUPC | 0.8669 |
| THAUUPM | 0.8248 |
| CogniCIC | 0.7123 |
| ITST | 0.6876 |
| UNEDUNIOVI | 0.6709 |
