Peer-reviewed research
Community-Supported Shared Infrastructure in Support of Speech Accessibility
Mark Hasegawa-Johnson et al., Journal of Speech, Language, and Hearing Research, 2024 · DOI 10.1044/2024_JSLHR-24-00122
Read the originalRead on October 16, 2026
- What it is
- The peer-reviewed description of the Speech Accessibility Project's data collection, with a first speech recognition experiment on its 2023-10-05 data package of people with Parkinson's disease. It compares an off-the-shelf model with one fine-tuned on the project's recordings.
- How much weight it carries
- Peer-reviewed in an ASHA journal and open access in PubMed Central. The baseline is an open-source wav2vec 2.0 model, not a commercial dictation product, and the test covers one condition (Parkinson's disease) in a first data release.
Facts taken from this source
- A baseline speech recognizer with a 3.4 percent word error rate on typical speakers had a 36.3 percent word error rate on speech from people with Parkinson's disease; fine-tuning on Speech Accessibility Project data reduced it to 23.7 percent.
“A baseline ASR, with a word error rate of 3.4% for typical speakers, transcribes test speech with a word error rate of 36.3%. Fine-tuning reduces the word error rate to 23.7%.” (Abstract, Results)
Pages that rest on this source
- Speech to Text for Accessibility, RSI and DisabilitiesTopic
- Word Error Rate (WER): How to Calculate and Read ItGuide
- How Accurate Is Speech Recognition? The Real NumbersGuide
- Speech Recognition for Stuttering and DysarthriaGuide
- Can voice recognition understand dysarthric speech?Question
- Is speech recognition less accurate for some people?Question
- What is a good word error rate?Question
- DysarthriaGlossary