Clinical Annotations for Automatic Stuttering Severity Assessment
This work addresses the need for reliable clinical expertise in stuttering assessment, but it is incremental as it builds on existing datasets and standards.
The paper tackled the problem of assessing stuttering severity by enhancing the FluencyBank dataset with a new clinical annotation scheme, resulting in multi-modal annotations and a test set with expert consensus for model evaluation.
Stuttering is a complex disorder that requires specialized expertise for effective assessment and treatment. This paper presents an effort to enhance the FluencyBank dataset with a new stuttering annotation scheme based on established clinical standards. To achieve high-quality annotations, we hired expert clinicians to label the data, ensuring that the resulting annotations mirror real-world clinical expertise. The annotations are multi-modal, incorporating audiovisual features for the detection and classification of stuttering moments, secondary behaviors, and tension scores. In addition to individual annotations, we additionally provide a test set with highly reliable annotations based on expert consensus for assessing individual annotators and machine learning models. Our experiments and analysis illustrate the complexity of this task that necessitates extensive clinical expertise for valid training and evaluation of stuttering assessment models.