importance 5/10ai assistedSpeech Recognition and Synthesisassistive speech technologystage: research
Problem
People with dysarthric speech (discontinuous, uncontrolled volume, slow or explosive pronunciation) are poorly served by standard automatic speech recognition.
Why it matters
Reliable ASR can restore communication and independence for speakers with dysarthria.
What the AI did
This is a scoping survey (1990-2022) documenting how deep learning methods such as CNNs, DNNs and RNNs have been applied to acoustic, lexical and language models to improve dysarthric-speech ASR.
Results
not stated in source
Impact
The survey reports deep learning was widely applied in the 2010s to improve ASR performance, but cites no concrete benchmark figures.
Record
Venue
Electronics
Published
2023
Work type
article
Authors
Zhaopeng Qian, Kejing Xiao
Citations
43 (OpenAlex, at import time)
References
85
Model or method
convolutional neural networks, deep neural networks, recurrent neural networks
dysarthric speech, automatic speech recognition, deep learning, survey, CNN, RNN
Importance rating
5/ 10
real but narrower advance
ASR for dysarthric speakers could restore communication and independence for a disability community, but the abstract is a scoping review citing no scale or benchmark figures.
Basis: inferred.
Judged from the source text, which gives no explicit scale.
Stated scale: not stated in source. Method and limits.