Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

PathBench: Speech Intelligibility Benchmark for Automatic Pathological Speech Assessment

About

Automatic speech intelligibility assessment is crucial for monitoring speech disorders and therapy efficacy. However, existing methods are difficult to compare: research is fragmented across private datasets with inconsistent protocols. We introduce PathBench, a unified benchmark for pathological speech assessment using public datasets. We compare reference-free, reference-text, and reference-audio methods across three protocols (Matched Content, Extended, and Full) representing how a linguist (controlled stimuli) versus machine learning specialist (maximum data) would approach the same data. We establish benchmark baselines across six datasets, enabling systematic evaluation of future methodological advances, and introduce Dual-ASR Articulatory Precision (DArtP), achieving the highest average correlation among reference-free methods.

Bence Mark Halpern, Thomas Tienkamp, Defne Abur, Tomoki Toda• 2026

Related benchmarks

TaskDatasetResultRank
Pathological speech intelligibility assessmentCOPAS Word
PCC0.6
36
Pathological speech intelligibility assessmentCOPAS Sentence
PCC0.36
36
Pathological speech intelligibility assessmentUASpeech Word
PCC0.98
24
Pathological speech intelligibility assessmentTORGO Sentence
PCC0.92
24
Pathological speech intelligibility assessmentNeuroVoz Sentence
PCC0.79
24
Pathological speech intelligibility assessmentEasyCall Word
PCC0.71
24
Pathological speech intelligibility assessmentTORGO Word
PCC0.57
24
Pathological speech intelligibility assessmentEasyCall Sentence
PCC0.62
24
Pathological speech intelligibility assessmentYouTube (YT) Sentence
PCC0.78
10
Showing 9 of 9 rows

Other info

Follow for update