Institution profile

Otto von Guericke University Magdeburg

Academic institutioneurope · de
Official website
Research library106linked papers
Opportunities0open roles
Selected work

Representative Papers

TidyVoice: A Curated Multilingual Dataset for Speaker Verification Derived from Common Voice

Jan 22, 2026

This work addresses the scarcity of large-scale multilingual read-speech datasets with reliable speaker labels, which has hindered progress in tasks such as anti-spoofing and speaker verification. We present the first systematic resolution of speaker label heterogeneity in Mozilla Common Voice, introducing TidySpeech—a high-quality dataset comprising over 210,000 monolingual and 4,500 multilingual speakers. We further define standardized evaluation protocols for monolingual (Tidy-M) and multilingual (Tidy-X) scenarios. A ResNet-based model fine-tuned on Tidy-M achieves an equal error rate (EER) of 0.35% and demonstrates significantly improved generalization on the unseen conversational dataset CANDOR. Both the dataset and models are publicly released, establishing the first large-scale open benchmark for cross-lingual speaker verification.

1 citationsRead paper

IM HERE: Interaction Model for Human Effort Based Robot Engagement

Jun 02, 2025Conference on Cognitive and Computational Aspects of Situation Management

Existing engagement models are typically vague and lack generalizability, hindering natural human–machine–social interaction. This paper proposes the first unified dynamic engagement modeling framework that jointly encodes subjective perception and objective behavioral states, supporting human–human, human–robot, and robot–robot interaction scenarios. Centered on “cognitive effort,” the framework quantifies attention allocation, identifies four fundamental engagement states, and enables automatic misalignment detection and socially norm-constrained behavior generation. By integrating social behavior analysis with autonomous system control techniques, our approach achieves, for the first time, cross-agent, interpretable, and intervenable engagement state modeling. Experimental results demonstrate significant improvements in robots’ social goal achievement and interaction naturalness. The framework establishes both theoretical foundations and technical infrastructure for the societal integration of embodied intelligence.

1 citationsRead paper
Recent publications

Latest Papers