2026 IEEE International Conference on Acoustics, Speech, and Signal Processing

Die 51. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2026) findet vom 4. bis 8. Mai 2026 in Barcelona, Spanien statt. Das Fraunhofer IDMT wird dort aktuelle Forschungsergebnisse zum Thema Speech Deepfake Detection.

In this work, we introduce a multi-task transformer for speech deepfake detection, capable of predicting formant trajectories and voicing patterns over time, ultimately classifying speech as real or fake, and highlighting whether its decisions rely more on voiced or unvoiced regions. Building on a prior speaker-formant transformer architecture, we streamline the model with an improved input segmentation strategy, redesign the decoding process, and integrate built-in explainability. Compared to the baseline, our model requires fewer parameters, trains faster, and provides better interpretability, without sacrificing prediction performance.

The poster will be presented by Luca Cuccovillo on May 7, 2026 at 16:30.