Skip to main content

Answer · Research and NVivo

How do I import interview transcripts into NVivo?

Export each interview as a Word file where every paragraph starts with the speaker's name followed by a colon, then import it into NVivo as a document. NVivo can then auto-code by speaker, creating one code or case per person. Our NVivo format does this for you, with a timestamp in front: "[00:04:12] Interviewer: ...".

Try it on your own file

Free for 3 files a day, up to 30 minutes each. Speaker labels, timestamps, and exports for Word, SRT and NVivo.

Drop audio or video here

MP3, MP4, M4A, WAV, MOV, WEBM, OGG · up to 5 hours

Why imports get messy

A common complaint: transcripts exported from online tools import into NVivo with speaker labels and timestamps that NVivo doesn't recognise. The usual cause is layout. Some exports put the speaker name on its own line, put timestamps in a table, or wrap one turn across several paragraphs.

NVivo's auto-coding by speaker works on paragraphs. It looks for a speaker name at the start of each paragraph, followed by a colon, and assigns that paragraph to the speaker. If the name is on a separate line, or if one turn spans several paragraphs, the coding breaks.

Document or media transcript?

There are two general ways to bring a transcript into NVivo. Importing it as a document is the simplest: you get the text, can code it, and auto-code by speaker. This is what our NVivo format is made for.

The other way is to import the audio or video as a media file and attach a transcript to it, so text rows are linked to time spans in the recording. NVivo expects a specific timestamp layout for that, which differs between versions. If you need it, check NVivo's own help for your version before you transcribe 40 interviews.

The details

One speaker turn per paragraph

Every paragraph should be one person's turn, starting with their name and a colon.

Use the same name every time

"Interviewer" in one file and "I" in another become two different speakers in NVivo. Rename speakers before export.

Anonymise before import

Replace participant names with codes such as P07 in the transcript editor, so the codes, not the names, become NVivo cases.

Timestamps help you go back to the audio

A [hh:mm:ss] at the start of each paragraph lets you find the moment in the recording when checking a quote.

MAXQDA has its own format

For MAXQDA, use the MAXQDA export, which writes timestamps as #hh:mm:ss-0# that MAXQDA recognises.

Step by step

  1. Transcribe and fix speakers

    Upload the interview, rename Speaker 1 and Speaker 2 to Interviewer and the participant code, and fix any lines at speaker changes.

  2. Export NVivo format

    Choose the NVivo export. You get a DOCX where each paragraph begins "[hh:mm:ss] Speaker: ".

  3. Import as a document

    In NVivo, import the DOCX as a file or document into your project, as you would any Word file.

  4. Auto-code by speaker

    Run NVivo's auto-coding and choose to code by speaker name. Check one file to make sure every turn landed on the right person, then do the rest.

Common mistakes

  • Exporting as plain TXT when you want auto-coding by speaker. Use DOCX in the NVivo format.
  • Editing transcripts heavily inside NVivo after coding has started. Fix speakers and words before import.
  • Leaving speaker names as Speaker 1 and Speaker 2 across 20 interviews; they will merge into the same two NVivo speakers.

Questions people also ask

Why doesn't NVivo recognise my speaker labels?

Usually because the speaker name is on its own line, isn't followed by a colon, or one turn is split across paragraphs. Each paragraph needs to start with the name and a colon. Our NVivo export follows that layout.

Will the timestamps get in the way of auto-coding?

The NVivo format puts the timestamp before the speaker name. If your NVivo version doesn't pick up speakers with a timestamp in front, export DOCX without timestamps for auto-coding and keep the timestamped copy for checking quotes.

Can I link the transcript to the audio in NVivo?

NVivo supports transcripts attached to media files, but it expects its own timestamp layout. Check NVivo's help for your version. Our export is designed for document import and auto-coding.

Does this work for focus groups?

Yes. Rename each participant before export. With many voices, check the speaker changes, especially where people talk over each other.

How much does it cost for 40 hours of interviews?

The 30-hour pack is $19.99 and the 10-hour pack $9.99, both one-time and never expiring, so 40 hours costs $29.98. Or Unlimited for $19.99 for one month.

Related questions