Peer-reviewed research
Early Deletion of Fillers in Processing Conversational Speech
Matthew Lease and Mark Johnson, Proceedings of NAACL HLT 2006, Short Papers, 2006
Read the originalRead on October 16, 2026
- What it is
- A computational linguistics paper from Brown University that tests whether deleting filler words such as "you know" and "like" from transcripts helps a parser process conversational speech. Its abstract reviews readability studies of disfluencies, meaning fillers and speech repairs.
- How much weight it carries
- Peer-reviewed conference paper. It is about machine parsing of transcripts, not dictation software or the reader's experience, and the readability findings it cites come from earlier work (Jones et al., 2003) that we did not read directly.
Facts taken from this source
- Readability studies cited by Lease and Johnson in 2006 found that fillers and speech repairs can be deleted from transcripts without compromising meaning.
“may be deleted from transcripts without compromising meaning” (Abstract)
Pages that rest on this source
- Speech to Text AI: Dictation, Assistants and Voice AgentsTopic
- AI Cleanup for Dictation: Fillers, Punctuation, RewritesGuide
- Verbatim vs Clean Verbatim TranscriptionGuide
- How do you stop saying um when you dictate?Question
- How does AI remove filler words from dictation?Question
- Why does dictated text sound different from typed text?Question
- Filler wordsGlossary