The priority with intelligent verbatim transcription is capturing the meaning of the data.
Verbatim means “word for word.” It’s one of multiple types of transcription that can be used depending on the goal of the document. Another is intelligent verbatim transcription, which takes context into account.
Transcribers who work straight verbatim take the audio and type absolutely everything said. They include every utterance, regardless of the level of importance or usefulness.
Verbatim transcription is indispensable in a legal setting, for example. It’s even handy if you’re trying to settle an argument. Wouldn’t we love a verbatim transcript when we’re trying to figure out who agreed to do the dishes earlier?
For most situations though, straight verbatim transcription isn’t that useful. That’s where intelligent verbatim transcription becomes useful.
What is intelligent verbatim transcription?
Let’s say you’re turning a presentation into a written report. Your CEO gave a talk at a conference with a big PowerPoint component.
She’s an engaging speaker, but she does say “you see” quite a bit. At one point she accidentally clicked too far ahead in her slides, then stopped to apologize and get herself organized. Also, she had a cold and spent about half a minute explaining her scratchy voice at the beginning of the presentation.
This is where “intelligence” comes into the picture.
Elements intelligent verbatim transcription will skip:
- Filler words like ‘um’ ‘so’ and ‘you know’, which act like conversational grease but make your written content cluttered and confusing
- Repeated sentences that might sound great when delivered live, for emphasis, but are unnecessary for reading
- Digressions and other irrelevant or off-topic content which will lessen the impact on the page
Again, the priority with intelligent verbatim transcription is uncovering the meaning of the data. So, the transcriber may also use their best judgement to do things like insert punctuation and correct obvious grammar errors to make the text more readable.
This is partly why human transcription is still needed. When clients come to us for speech data collection and transcription, they’re trying to solve for the edge cases where automatic speech recognition (ASR) still struggles – whether it’s recognizing a greater variety of accents, dealing with background noise, transcribing conversational data, or cutting out examples noted above.
Are you looking for more information about verbatim transcription, or other ways of transcribing content? Shoot us a message. We would love to help you out.
How AI Annotation Will Affect The Future of UX and UI
How AI annotation of text and images is changing the field of UX and UI, and how you can use these tools t...
The 5 Types of Image Annotation You Need to Know
The most common types of image annotation, and a bonus newer development.