Verbatim transcription

Verbatim transcription is a written record that keeps everything a speaker said, word for word. It keeps fillers such as "um" and "uh," repeated words, false starts, self-corrections and, in many styles, notes on laughter, pauses and coughs. The same style goes by several names. "Full verbatim," "true verbatim" and "strict verbatim" are used interchangeably by different services, so check how a provider defines its labels before you order one.

Verbatim keeps the delivery, not just the words

The defining rule is that the transcript should match the audio as closely as possible, including the parts a reader would normally skip. A verbatim transcript of "so I, I think we should, uh, wait" keeps the repeated "I," the filler and the hesitation. Researchers use this detail when they study how people speak, because hesitation, emphasis and restarts can carry meaning. A transcript that removes them has already made an editing decision.

Verbatim takes far longer than the audio

The price of that detail is time. In a 2024 study of qualitative interviews, Wollin-Giering and colleagues report that verbatim transcription took an average of six hours for each hour of interview material in the research projects of Gläser and Laudel ¹. That is a figure from one body of research, not a fixed rate, but it shows why projects choose verbatim only when they need it. Automatic transcription can produce a first verbatim-style draft quickly, and a person then checks it against the audio.

A concrete example

Here is one answer from an interview, transcribed three ways. The speaker said: "Um, so, I, I think we should, uh, wait until March."

The verbatim line keeps every sound. The clean line keeps the meaning and the wording. The edited line changes some of the phrasing as well. Each is correct for a different purpose.

Why it matters for dictation

Dictation tools aim at the clean style: they usually drop fillers and insert punctuation, so the text reads as a finished sentence. A verbatim transcript is the opposite. When you check a dictation tool, ask whether it writes fillers. Fillers are a problem in a formal letter and often acceptable in a meeting record, so the answer decides how much editing you will do. For typing, a verbatim job is slower than the same words typed from a script, since the typist must catch every sound, including the ones a speaker barely notices.

Verbatim also matters for scoring. A word error rate (WER) counts every word in the reference, so a transcript that keeps fillers and a tool that removes them will score differently even when both are accurate. Compare scores only when the two used the same style. The people who produce these transcripts are transcriptionists, and the lighter style is covered in clean verbatim.

Related terms

• Clean verbatim: the edited style that removes fillers and false starts.
• Timestamp: a time marker that many verbatim transcripts add to each speaker turn.
• Speaker diarization: labeling who said each part of the recording.

Guides that cover it

• Verbatim vs clean verbatim transcription
• What does a transcript look like?
• How to transcribe an interview