Transcription
Verbatim transcription: the 3 types explained
Ordering a transcript, you will usually be asked to choose between verbatim and non-verbatim transcription, and verbatim itself comes in several flavours. In this guide we explain the three types of verbatim transcription and help you decide which one fits your project.
- Verbatim. Every word transcribed literally, including false starts, fillers and stutters.
- Three flavours. Clean verbatim (lightly tidied), smooth verbatim (readable but complete) and true verbatim (everything, incl. non-speech sounds).
- When it matters. Research, legal transcripts and focus groups, where tone and context carry meaning.
- Faster. Automatic software delivers a literal first draft in minutes.
What is verbatim transcription?
A verbatim transcript captures every spoken word in the recording and puts it into text. That includes false starts, grammatical errors, interruptions and stutters. It is the most complete form of transcription: 100% true to the recording. These verbal cues reveal the context of the conversation, how something was said, not just what.
Written-out speech looks strange to anyone used to normal written language, because people stutter, repeat themselves and jump between thoughts. That is exactly the difference between verbatim and non-verbatim (edited) transcription, which cleans all of this up for readability.
The three types of verbatim transcription
1. Clean verbatim
Clean verbatim (also called intelligent verbatim) removes the noise that adds no meaning: filler words like “uhm”, stutters and false starts are dropped, and obvious slips are lightly tidied. What remains is a complete but comfortable-to-read record of what was said. This is the most popular choice for interviews, podcasts and meetings.
2. Smooth verbatim
Smooth verbatim sits in the middle: repetitions and fillers stay out, but self-corrections, interruptions and colloquial language are kept as spoken. You keep a feel for the dynamics of the conversation without wading through every “uh”. A good fit for journalism and qualitative research where flow matters.
3. True verbatim
True (or full) verbatim captures absolutely everything: fillers, stutters, repetitions, laughter, pauses and non-speech sounds like “mm-hmm (affirmative)”. It is the most labour-intensive form and the standard for legal proceedings, behavioral research and any analysis where the way something is said is data in itself.
Which type should you choose?
- Legal & compliance. True verbatim: courts typically require a complete record, including hesitations.
- Research. True or smooth verbatim, depending on whether non-verbal cues are part of your analysis.
- Journalism & content. Clean verbatim: accurate quotes without the clutter.
- Meetings & notes. Clean verbatim, or an edited (non-verbatim) transcript if readability is all that counts.
How to get your verbatim transcript
Upload your recording
Create a free account and upload your audio or video file; the first 10 minutes are free.
Pick your service
Choose automatic transcription (a literal draft in minutes, 90%+ accuracy with clear audio) or professional transcription, where you specify clean read or verbatim and our experts deliver 99%+ accuracy.
Edit and export
Fine-tune the transcript in the online editor, with audio linked to text, then export as Word, text or JSON with optional timestamps and speaker labels.
Get your verbatim transcript
Create a free account and transcribe your first recording in minutes.
Try for freeFrequently asked questions
Do you offer clean read and verbatim transcription?
When should I choose verbatim?
How do automatic transcription services work?
What is the turnaround time?
Top reads
Accessibility in higher education: how universities can implement inclusive and compliant digital learning
9 Mar 2026 5 minute read
Data protection in transcription software: what you need to know
23 Feb 2026 4 minute read
Data sovereignty in transcription and subtitling: a strategic risk factor
23 Feb 2026 4 minute read