Multilingual Stutter Event Detection for English, German, and Mandarin Speech

2026-03-27

AI Summary

This study addresses the challenge of automatic stuttering event detection in multilingual settings by proposing a language-agnostic, multi-label detection approach. By integrating annotated data from four speech corpora spanning English, German, and Mandarin Chinese, the method leverages multi-task learning and multilingual speech modeling for joint training. This work provides the first empirical validation that stuttering exhibits shared acoustic and prosodic characteristics across languages. The resulting cross-lingual model not only overcomes the limitations of monolingual systems but also achieves performance on par with or superior to state-of-the-art language-specific methods, substantially enhancing the generalization, robustness, and universality of stuttering detection.

Abstract

This paper presents a multi-label stuttering detection system trained on multi-corpus, multilingual data in English, German, and Mandarin.By leveraging annotated stuttering data from three languages and four corpora, the model captures language-independent characteristics of stuttering, enabling robust detection across linguistic contexts. Experimental results demonstrate that multilingual training achieves performance comparable to and, in some cases, even exceeds that of previous systems. These findings suggest that stuttering exhibits cross-linguistic consistency, which supports the development of language-agnostic detection systems. Our work demonstrates the feasibility and advantages of using multilingual data to improve generalizability and reliability in automated stuttering detection.

Problem

• stutter detection
• multilingual
• speech disfluency
• cross-linguistic
• automatic speech analysis

Innovation

• multilingual stutter detection
• language-independent features
• multi-corpus training
• cross-linguistic consistency
• automated stuttering detection