Peer-reviewed research
Reviewing Speech Input with Audio: Differences between Blind and Sighted Users
J. Hong et al., ACM Transactions on Accessible Computing, 2020 · DOI 10.1145/3382039
Read the originalRead on October 16, 2026
- What it is
- A lab study of 12 blind and 12 sighted participants that combined interviews about speech input with a task in which people had to find recognition errors in their dictated text using audio only. It compares how well each group identified errors and how much each group worried about them.
- How much weight it carries
- Peer-reviewed in a leading accessibility journal. The sample is small, the task used a fixed set of dictated texts, and the speech engine is older than current systems. The finding that blind participants did no better than sighted participants is a result, not a hypothesis that held.
Facts taken from this source
- The 2020 study by Hong and colleagues compared 12 blind participants with 12 sighted participants.
“we conducted a lab study with 12 blind and 12 sighted participants” (Abstract)
- In the 2020 study's speech input task, blind participants identified only 40 percent of speech recognition errors.
“blind participants were able to identify only 40% of ASR errors” (Abstract)
- The 2020 study found that blind users were more highly concerned about speech recognition errors than sighted users.
“blind users were more highly concerned” (Abstract)