Fact · Typing and writing speed
The study found significantly lower quality for streaming speech recognition, the kind used for live events.
Checked against its source on October 16, 2026
In the source's own words
“We also measured a significant lower quality for streaming ASR, which is used for live events”
Measuring the Accuracy of Automatic Speech Recognition Solutions, Korbinian Kuhn et al., arXiv (cs.CL), preprint, 2024, Abstract.
Open the original- What it measures
- Streaming recognition shows words while the speaker is still talking, which is what live captions need.
- What it does not show
- The abstract reports the finding without a headline percentage. Check the full paper for the numbers before quoting them.