Fact · Transcription
In a 2017 comparison of machine and vendor transcripts on telephone speech, the automatic recognizer tended to confuse filled pauses such as uh with backchannels such as uhhuh, and humans tended not to make that error.
Checked against its source on October 17, 2026
In the source's own words
“the automatic recognizer tends to confuse filled pauses ("uh") and backchannel acknowledgments ("uhhuh"). Humans tend not to make this error”
Comparing Human and Machine Errors in Conversational Speech Transcription, Andreas Stolcke and Jasha Droppo, Proceedings of Interspeech 2017, ISCA, 2017, Abstract.
Open the original- What it measures
- The paper compares the most frequent error types of a recognizer and a commercial transcription vendor on conversational telephone speech.
- What it does not show
- Conversational telephone speech from 2017. Recognizers have improved since then, so the specific confusion may be smaller now.