WildElder: A Chinese Elderly Speech Dataset from the Wild with Fine-Grained Manual Annotations

2025-10-10

AI Summary

Existing Chinese elderly speech datasets are predominantly collected in controlled environments, lacking diversity from real-world scenarios and insufficient coverage of age-related acoustic characteristics—such as slowed articulation rate and vocal tremor—thereby limiting robustness in elderly speech recognition and speaker analysis. To address this, we introduce WildElder, the first large-scale, real-world Chinese elderly speech dataset curated from publicly available online videos. It features fine-grained transcriptions and multi-dimensional annotations, including speaker age, gender, and accent strength, all rigorously validated by domain experts. By integrating the ecological validity of “wild” data with the high fidelity of expert-curated labels, WildElder significantly increases task difficulty while establishing a new benchmark for elderly speech processing. The dataset and associated preprocessing code are publicly released to foster reproducible research.

Technology Category

• Natural Language Processing: Speech
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Application Domains: Humanities & Computational Social Science

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Elderly speech poses unique challenges for automatic processing due to age-related changes such as slower articulation and vocal tremors. Existing Chinese datasets are mostly recorded in controlled environments, limiting their diversity and real-world applicability. To address this gap, we present WildElder, a Mandarin elderly speech corpus collected from online videos and enriched with fine-grained manual annotations, including transcription, speaker age, gender, and accent strength. Combining the realism of in-the-wild data with expert curation, WildElder enables robust research on automatic speech recognition and speaker profiling. Experimental results reveal both the difficulties of elderly speech recognition and the potential of WildElder as a challenging new benchmark. The dataset and code are available at https://github.com/NKU-HLT/WildElder.

Problem

• Addressing elderly speech recognition challenges from age-related changes
• Overcoming limited diversity in controlled-environment Chinese speech datasets
• Providing fine-grained annotated in-the-wild Mandarin elderly speech data

Innovation

• Collected elderly speech from online videos
• Added fine-grained manual annotations to data
• Combined wild data realism with expert curation