Although the UA-Speech and TORGO databases of control and dysarthric speech are invaluable resources made available to the research community with the objective of developing robust automatic speech recognition systems, they have also been used to validate a considerable number of automatic dysarthric speech classification approaches. Such approaches typically rely on the underlying assumption that recordings from control and dysarthric speakers are collected in the same noiseless environment using the same recording setup. In this paper, we show that this assumption is violated for the UA-Speech and TORGO databases. Using voice activity detection to extract speech and non-speech segments, we show that the majority of state-of-the-art dysarthria classification approaches achieve the same or a considerably better performance when using the non-speech segments of these databases than when using the speech segments. These results demonstrate that such approaches trained and validated on the UA-Speech and TORGO databases are potentially learning characteristics of the recording environment or setup rather than dysarthric speech characteristics. We hope that these results raise awareness in the research community about the importance of the quality of recordings when developing and evaluating automatic dysarthria classification approaches.
Differentiable and Severity-invariant Discrete Tokens for Dysarthric Speech Recognition
The Chinese University of Hong Kong(香港中文大学);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);National Research Council Canada(加拿大国家研究委员会)
DSSCNet: A Transfer Learning Framework for Cross-Corpus Dysarthric Speech Severity Classification
Department of Computer Science and Engineering, Sikkim Manipal Institute of Technology, India(计算机科学与工程系,西基姆曼普尔理工学院,印度);Department of Electronics and Communication Engineering, National Institute of Technology Sikkim, India(电子与通信工程系,西基姆国家理工学院,印度);Signal Analysis and Interpretation Laboratory (SAIL), University of Southern California, Los Angeles, USA(信号分析与解释实验室(SAIL),南加州大学,美国洛杉矶)
Systematic Study of Dysarthric Speech Recognition: Spectral Features and Acoustic Models
Department of Electronics and Communication Engineering, National Institute of Technology Sikkim, India(电子与通信工程系,印度尼特技术学院锡金分校);Department of Information and Communications Engineering, Aalto University, Finland(信息与通信工程系,阿尔托大学,芬兰);Signal Analysis and Interpretation Laboratory (SAIL), University of Southern California, Los Angeles, USA(信号分析与解释实验室(SAIL),美国南加州大学洛杉矶分校)
Cross-Dataset, Age, and Gender Generalization: A Comprehensive Analysis of Fine-Tuning Strategies for Low-Resource Children's ASR
Department of Electronics and Communication Engineering, National Institute of Technology Sikkim, India(印度西西姆国立技术学院电子与通信工程系);Signal Analysis and Interpretation Laboratory, University of Southern California, Los Angeles, USA(美国南加州大学洛杉矶分校信号分析与解释实验室)
Institute of Psychology(心理研究所);University of the Chinese Academy of Sciences(中国科学院大学);Beijing Union University(北京联合大学);Jiangsu University of Science and Technology(江苏科技大学);University of Science and Technology of China(中国科学技术大学)
Benchmarking Commercial Speech Recognition and Multimodal Large Language Models on Dysarthric Speech: Severity-Stratified Baselines and Architecture-Specific Prompting Effects