Turnwise / NVivo, ATLAS.ti and MAXQDA
Transcripts for NVivo, ATLAS.ti, MAXQDA and Dedoose
Turnwise exports checked interview transcripts in the shapes each analysis package expects, with timestamps that link back to the recording and speaker names intact.
Choose an export for your software
Each analysis package expects transcripts in a slightly different shape if you want the text linked to the recording. Turnwise has a preset for each, so timestamps and speaker names arrive where the software looks for them.
| Software | Turnwise export | What you get |
|---|---|---|
| NVivo | NVivo (.txt) | Tab-separated rows with Timespan, Content and Speaker columns. |
| ATLAS.ti | ATLAS.ti (.vtt) | WebVTT transcript with speaker names, linkable to the audio. |
| MAXQDA | MAXQDA (.docx) | Word document with f4-style timestamps such as #00:01:23-4#. |
| Dedoose | Dedoose (.docx) | Word document with speaker names; timestamps optional. |
Whichever you choose, the export uses your corrected text, your speaker names and the style you pick (clean or verbatim).
NVivo transcription import
Export with the NVivo preset. The file is plain text with one row per turn: the time span, the words, and the speaker. In NVivo, import the audio file, open it, choose Import Transcript, and map the columns to Timespan, Content and Speaker. Each row then plays from the right point in the recording.
For a whole project, Turnwise also writes Participants.csv, one row per participant with the details you recorded (age group, site, role and so on). NVivo imports it as a classification sheet, so you can compare groups in your queries.
ATLAS.ti
Export with the ATLAS.ti preset to get a WebVTT (.vtt) file with speaker names. Add the recording to your ATLAS.ti project and import the .vtt as its transcript; ATLAS.ti links each segment to its place in the audio.
MAXQDA
Export with the MAXQDA preset to get a Word document with timestamps in the f4/f5 format (#00:01:23-4#) that MAXQDA recognises. Import it with MAXQDA's transcript import and assign the recording; the timestamps become links to the audio.
The project export also includes Transcripts and variables.csv, one row per transcript starting with Document group and Document name, which MAXQDA imports as document variables.
Dedoose
Export with the Dedoose preset to get a Word document with speaker names, ready to upload and excerpt. You can include timestamps, and codes and memos from Turnwise if you want them in the document.
Exporting a whole project
Exporting a project writes one folder containing:
- every transcript in the format you chose;
- Participants.csv, for NVivo case classifications;
- Transcripts and variables.csv, for MAXQDA document variables;
- Codebook.csv, your codes with descriptions and how many turns each is applied to;
- Methods statement.txt, describing the models, versions and settings used for the study.
Codes and memos
Codes and memos you add in Turnwise are included in the Word, Dedoose, plain-text, CSV and JSON exports. They are not carried into NVivo, ATLAS.ti or MAXQDA as codes in those programs, so if you plan to code there, it is usually simplest to do your coding there too.
Other formats
- Word (.docx) for reading, sharing and appendices.
- Plain text (.txt), which opens anywhere.
- CSV, one row per turn with start, end and speaker, for Excel, R or Python.
- SRT and WebVTT subtitles for video players and editors.
- Turnwise JSON, a documented open format with every word's timing and confidence, speakers, checked turns, codes, memos and how the transcript was made. Useful for archiving a study's data and for your own scripts.
Mac with Apple silicon (M1 or newer), macOS 14 or later. Desktop licenses are currently free for pilot users and students: write to [email protected].