Edited Verbatim Guidelines
What should I leave out of an "edited verbatim" transcript?
We use an edited (or “clean”) verbatim format. That means these words should be excluded:
- “um”s, “uh”, and other such nonverbal utterances
- “you know”
- excessive uses of “like”
- “all right” and “okay”, unless these are answers to questions
- “yes”, “yeah” and “right”, unless these are answers to questions
“That’s all. Right. Okay. Let’s get started, okay?”
Correct transcription: “That’s all. Let’s get started.” The instances of “right” and “okay” have been removed.
“Um, I don’t think so. I mean like okay she said okay, but I don’t think she wants to go, you know?”
Correct transcription: “I don’t think so. She said okay, but I don’t think she wants to go.”
What do I do if I can't make out a word?
Anything you’re not able to make out is transcribed as [Inaudible 0:00:00]. The time represents the time in the audio where the inaudible occurred. Please use this notation if you are less than 95% sure about a word.
If you are almost positive about a word but want to be safe, put [Inaudible 0:00:00 sounds like:]. For example: [Inaudible 0:04:32 sounds like: pistachios]. This is used if you are at least 85% sure the word is correct.
Also, we appreciate if you give your transcript a second look after you’ve transcribed it. Sometimes a word you couldn’t make out originally will become clear after you’ve listened to the entire file. Many of our transcriptionists listen the second time at 150% speed in order to speed the process up.
What are the standard notations?
- Laughing is transcribed as [Laughter] when more than one person is laughing.
- If only one person is laughing and you know which speaker it is, use [Laughs].
- Overlapping speech is transcribed as [Crosstalk].
- Anything you’re not able to make out is transcribed as [Inaudible 0:00:00]. The time represents the time in the audio where the inaudible occurred.
- If you are almost positive about a word but want to be safe, put [Inaudible 0:00:00 sounds like:]. For example: [Inaudible 0:04:32 sounds like: pistachios].
- If you’re unsure how something is spelled, such as someone’s name, but are sure you have heard the word correctly, place [ph] after the word. For example: Jane Doe [ph].
- All files should end with [End of Audio].
What if multiple people are speaking at once?
As we work a lot with focus groups, a lot of our files include crosstalk (multiple people speaking at once). Our clients usually understand that there is just no way to get every single word spoken in a focus group. Because we don’t want transcripts full of [Crosstalk], if two or more people are speaking and you can make out at least one, just transcribe the one person who is speaking the most pertinent information.
If two or more people are interrupting each other in a short period in an edited verbatim transcript, transcribe both as if there was no interruption. This is only when it’s one or two words spoken before each interruption.
- Julie: “And that’s when I told her—”
- Alice: “Yeah, I—”
- Julie: ” I would.”
- Alice: “did, too.”
Instead transcribe this as:
- Julie: “And that’s when I told her I would.”
- Alice: “I did, too.”
What if multiple people say the exact same thing at once?
In an edited verbatim transcript, if more than one person says the same thing at roughly the same time, you can transcribe it just once and use “Multiple:” as the speaker ID tag.
- Moderator: Did you like this movie?
- Multiple: Yes.
You can also use this method if the words aren’t exactly the same, such as multiple people using different variations of “Yes”, “Yeah” and “Yep.” That can just be transcribed as “Multiple: Yes.” Use your discretion in this situation and make sure that combining the speakers’ words into one word or phrase does not alter the intent of the speakers.
- Moderator: Did you like the movie?
- Respondent: Yes.
- Respondent: Eh.
These responses are not similar enough to be combined, as “Eh” is not as positive as “Yes”.
How do I deal with grammatical errors in a transcript?
For the most part, grammatical errors are kept as is in both edited verbatim and strict verbatim transcripts.
Respondent: “I seen him go in there.”
Grammatically, this should be “I saw him go in there”, but we would leave it as is because it is as the speaker said it.
Do I use "wanna" and "gonna"?
Not in an edited verbatim transcript. Transcribe “going to” instead of “gonna“, and “want to” instead of “wanna“. Most other shortened words should also be transcribed as the full word, such as “because” instead of “’cause” or “‘cuz“.