Beyond "Hey Siri": towards Inclusive Speech Technology
Featured in the University of Groningen news about our work on speech technology for lesser-resourced languages, accessibility, and sarcasm detection.
Mar 24, 2025
AMuSeD: An Attentive Deep Neural Network for Multimodal Sarcasm Detection Incorporating Bi-modal Data Augmentation
A novel approach to multimodal sarcasm detection that combines text and audio data augmentation with attention mechanisms.
Dec 13, 2024
Sarcasm Detection and Synthesis
Research on multimodal approaches to detect and synthesize sarcastic speech, attracting international media coverage.
May 20, 2024
Improving Sarcasm Detection from Speech and Text through Attention-based Fusion Exploiting the Interplay of Emotions and Sentiments
We enhance sarcasm detection by combining speech and text data with emotion recognition and sentiment analysis, using attention-based fusion to capture the characteristic mismatch between emotional tone and semantic content.
May 13, 2024