Models and Approaches for Comprehension of Dysarthric Speech Using Natural Language Processing: Systematic Review.
Benard Alaka, Bernard Shibwabo
PMID 37889538WHAT IT FOUND
Most dysarthric speech research focused on intelligibility, not actual meaning extraction.
For speech-language pathologists, this review indicates current tools may still miss listener familiarity, topic knowledge, and context needed to understand patients.
Key findings
01The review included 30 studies from 834 search results, selected for meaning extraction from dysarthric speech rather than intelligibility alone.
02Most included studies leaned toward intelligibility, and actual meaning extraction was minimal.
03Hybrid approaches that combined speech features, speech patterns, and semantic knowledge were among those trying to go beyond intelligibility.
STILL TO COME
How it was doneWhat they foundWhat it means for SLPs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The review used a strict definition of comprehension, so studies that only measured intelligibility were excluded. It did not cover other possible meaning-extraction techniques, such as computer vision. Seven of the 30 included studies were judged to have somewhat concerning risk of bias. The review synthesized published models and approaches rather than testing an intervention in patients.
Declared interests
The authors declared no conflicts of interest.
The easy way to misread this
Do not read this review as proof that natural language processing tools now improve dysarthric speech comprehension. The included studies mostly leaned toward intelligibility, and the review's main finding was that actual meaning extraction was minimal.