Glossary
Filler words
Filler words are sounds and words such as um, uh, like and you know that speakers produce while they plan what to say next, without adding content. Many dictation tools can remove them automatically, and research on transcripts found they can be deleted without compromising meaning.
Updated October 16, 20264 cited facts, each traced to its source
Also called disfluencies, um and uh.
Filler words are the sounds and short words that fill a gap in speech, such as um, uh, "like" and "you know". Linguists group them with other disfluencies, such as false starts, repeated words and self-corrections. They carry no new content, but they are not random noise either.
Um and uh announce a pause before the next word
Herbert Clark and Jean Fox Tree proposed that speakers use uh and um to announce that they are about to pause. In their account, speakers use these sounds to announce what they expect to be a minor delay after uh, or a major delay after um 1. The sounds are therefore planning signals. A speaker who says "um" is telling the listener that the next thought will take longer to arrive.
Dictation tools differ in whether they keep fillers
A verbatim transcript keeps every filler, while a clean transcript removes them. Dictation tools are not all the same on this point. Microsoft says Fluid dictation in Windows voice typing, on Copilot+ PCs, automatically corrects grammar, punctuation and filler words as you speak 2. Tools that do not offer this clean-up keep the words you said.
Check the setting in your own tool before you judge its output. If your text comes out with "um" in it, the tool is likely transcribing verbatim. If it comes out clean, the tool has removed the fillers for you, and you may need to check that it did not also drop a word you meant to keep.
Removing fillers does not usually change the meaning
Readability studies cited by Lease and Johnson in 2006 found that fillers and speech repairs can be deleted from transcripts without compromising meaning 3. That is why clean-up is a safe default for most dictated drafts. It removes clutter that a reader does not need, and it leaves the content in place.
Clean-up does have limits. A filler sometimes marks a real hesitation, and a removed hesitation can make a sentence sound more certain than the speaker was. For legal, medical or formal text, read the cleaned version against the audio before you rely on it.
Filler words affect how dictation is scored
The same rule matters when accuracy is measured. For the Open ASR Leaderboard's normalized transcripts, punctuation and casing are removed, along with disfluencies such as fillers, repetitions and repairs, before word error rate is computed 4. A tool that writes "um" where the reference text has none would be counted as making errors. When you compare dictation tools, check whether a benchmark counted fillers at all, because the answer changes the score.
How to dictate without filling the gaps
Most people who want to reduce fillers learn to pause silently instead. A short silence reads as a break, and it costs nothing. Compose the sentence before you speak it, then say it in one even phrase. Our guide to how to dictate describes the method, and the question on how to stop saying um when dictating covers the practice steps. Our question on how AI removes filler words explains the clean-up step in more detail.
Related terms
- Dictation: speaking so that software writes your words as you say them.
- Homophone: a word that sounds like another word, a different kind of error in dictated text.
- Oral composition: working out a text by speaking, which is where most fillers come from.
Guides that cover it
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
Related terms
- Dictation and Voice Typing: How to Talk Instead of TypeThe topic this term belongs to
- Automatic punctuationAutomatic punctuation is when speech software infers commas, periods and question marks from your speech and inserts them without a spoken command. It can be switched on or off, and it works only in supported languages.
- DictaphoneA Dictaphone is a machine that records a speaker's voice so that another person can play it back and type it later. The name began as a brand, and it became the everyday word for dictation machines.
- DictationDictation is speaking your words so that software writes them as text, usually into the app that has the cursor. Phones and computers now include it, and people also call it voice dictation or voice typing.
- Dictation foot pedalA dictation foot pedal is a footswitch that controls playback of a recording, so a typist can start, pause and rewind audio without using their hands. It is the traditional control for transcribing dictation.
- Digital dictationDigital dictation is recording speech as digital audio, then sending the file to a typist or to speech recognition software for transcription. It replaced tape with files that can be encrypted, uploaded or routed for processing.
- HomophoneA homophone is a word pronounced the same as another word but different in meaning, spelling or both, such as to, too and two. Dictation software has to choose the right spelling from context, so homophones are one of the most common errors in dictated text.
- Noise-canceling microphoneA noise-canceling microphone is designed to filter ambient noise, so it picks up mainly the speaker's voice. It can help dictation in noisy rooms, but it does not remove all noise, and it is different from the active noise cancellation in headphones.
- Oral compositionOral composition is working out and shaping a text by speaking it, whether to a dictation tool, to a recorder or silently in thought. Research on letters finds that planning takes most of the time, however the text is produced, so speaking mainly changes how the words get onto the page.
- Push-to-talkPush-to-talk is a dictation mode in which you hold a key while you speak and release it to stop, so the software writes only the speech made while the key is down. Some tools also offer tap-to-start and tap-to-stop.
- ShorthandShorthand is an abbreviated symbolic writing method that increases the speed of writing compared with longhand. Trained writers used it to take dictation before recording machines, and it still underpins court reporting.
- Spoken punctuationSpoken punctuation is saying the name of a punctuation mark, such as comma or question mark, so that dictation software writes that mark into the text. It is the manual alternative to automatic punctuation.
- Voice commandA voice command is a spoken instruction that the software carries out, such as delete or stop listening, rather than writing the words as text. Dictation tools must separate commands from dictated words, and the rules differ by product.
Pages that use this term
- Dictation and Voice Typing: How to Talk Instead of TypeTopic
- AI Dictation: How AI Voice Typing WorksGuide
- Is Speech to Text AI? What Kind of AI Voice Typing UsesGuide
- AI Cleanup for Dictation: Fillers, Punctuation, RewritesGuide
- How to Dictate Emails Faster Than You Type ThemGuide
- How to Dictate: Techniques for Clean, Usable TextGuide
- Voice Access Commands: The Windows 11 Command ListGuide
- AI Voice Notes: From Rambling to Summaries and To-DosGuide
- Verbatim vs Clean Verbatim TranscriptionGuide
- Does Dictation Change How You Write?Guide
- Fluid Dictation, Copilot+ PCs and Windows on ArmGuide
- How do you calculate your speaking speed in words per minute?Question
- How do you dictate text with Voice Access?Question
- How do you dictate with AI?Question
- How do you stop saying um when you dictate?Question
- How do you turn off AI rewriting in a dictation app?Question
- How do you turn voice notes into text with AI?Question
- How does AI remove filler words from dictation?Question
- What is fluid dictation in Windows 11?Question
- What is the Open ASR Leaderboard?Question
- Why does dictated text sound different from typed text?Question
- Why does dictating feel so awkward at first?Question
- HomophoneGlossary
- Non-native speechGlossary