CONF Hernandez_INTERSPEECH_2026/IDIAP Adapting Self-Supervised Speech Representations for Cross-Lingual Dysarthria Detection in Parkinson's Disease Hernandez, Abner Yeo, Eunjung Choi, Kwanghee Li, Chin-Jou Yue, Zhengjun Das, Rohan Kumar Rusz, Jan Magimai-Doss, Mathew Orozco-Arroyave, Juan Rafael Arias-Vergara, Tomás Maier, Andreas Nöth, Elmar Mortensen, David R. Harwath, David Pérez-Toro, Paula Andrea cross-lingual domain adaptation Oral Diadochokinesis (DDK) Parkinson’s disease self-supervised learning ysarthria detection Proceedings of Interspeech 2026 10.21437/Interspeech.2026-773 doi The limited availability of dysarthric speech data makes cross-lingual detection an important but challenging problem. A key difficulty is that speech representations often encode language-dependent structure that can confound dysarthria detection. We propose a representation-level language shift (LS) that aligns source-language self-supervised speech representations with the target-language distribution using centroid-based vector adaptation estimated from healthy control speech. We evaluate the approach on oral diadochokinesis recordings from Parkinson's disease speech datasets in Czech, German, and Spanish under both cross-lingual and multilingual settings. LS substantially improves sensitivity and F1 in cross-lingual settings, while yielding smaller but consistent gains in multilingual settings. Representation analysis further shows that LS reduces language identity in the embedding space, supporting the interpretation that LS removes language-dependent structure.