Session

Speech Technologies 2

Day / Time: 24.03.2026, 16:00-18:20
Type: Vortrag (strukturierte Sitzung)

Session chair

Contributions

16:00 - 16:20
Towards a General Framework for Pathologic Voice Rating Across Different Domains
Karolin Krüger, Marten J. Finck, Steffen Paschen, Gerhard Schmidt
16:20 - 16:40
Robustness of Speech Quality Models to Pause Manipulation: Implications for Data Preprocessing
Robert Spang, Harshvardhan Ram, Kirill Shchegelskiy⁩, Tuğçe Koçak Büyüktaş, Wafaa Wardah, Sebastian Möller
16:40 - 17:00
Assessing the Reliability of Crowdsourced Speech Data for Objective Quality Prediction
Kirill Shchegelskiy⁩, Muni Won, Robert Spang, Sebastian Möller
17:20 - 17:40
Azimuth-Conditioned Dual-Encoder Architectures for Binaural Speaker Separation: Exploring Encoder Sharing Strategies
Ruben Johnson Robert Jeremiah, Peyman Goli, Steven van de Par
17:40 - 18:00
Investigating Text-to-Audio Latent Diffusion Model for Industrial Audio Synthesis
Muhammad Afaq Khattak, Syed Mir Izhar Ali Shah, Zain Ali, Saichand Gourishetti, Andrew McLeod
18:00 - 18:20
Avatar animation and control for speech reading cues studies
Antoine Bourachot, Bernhard U. Seeber