Transcribe an interview for research
Transcribe an interview for research: timecodes on every line, speakers named and a CSV row for every segment.
Upload the recording of the interview and the transcript comes back with a timecode on every line and the speakers named, so every quote can be traced to the minute it was said. One file is up to 30 minutes; Pro is $10 a month paid yearly.
Why the transcript is material, not a summary
An interview is a mess of half sentences, long pauses and two people finishing each other's thoughts. What a study needs from it is not a summary but a record: who said what, in which order, and where in the recording. A transcript with timecodes and speaker names is that record, and it is the material a qualitative analysis starts from rather than the result of it.
What the interview gives the study
- Bulk work
a REST API and an MCP server with 4 tools; the API plan is billed on the audio you send
- The allowance
Pro is $10 a month paid yearly ($120 a year) with no minute limit of audio a month
- Export for the analysis
CSV: one row per segment, with the time and the speaker
How to transcribe an interview
- 1
Collect the recording of the interview
Whatever the recorder of the field work produced is the file. Keep it as it is: no conversion, no cut, and the same handling for every participant.
- 2
Upload it and name the voices
Drop the file on the widget above, up to 30 minutes. Switch on the speaker split, because the interviewer and the participant are two different roles in the data.
- 3
Code the transcript, not the memory
Every line has the time it starts and ends. Find the passage you are coding, write the timecode next to it, and the excerpt points back to the recording.
- 4
Move the excerpts into the analysis
CSV gives one row per segment with the time and the speaker, which is what a spreadsheet wants. DOCX is on Pro; TXT and Markdown are free.
Tips
Name the file with the code of the participant.
Write the timecode next to every excerpt you quote.
Keep the split on: the roles are part of the data.
Questions this page answers
How do I keep the participant anonymous?
The transcript holds the words that were spoken, so a name said in the interview stays in the text. What the service adds is the time of every line and, on request, the speaker label. The code that ties a quote to a participant is yours to keep.
Is there a bulk option for many interviews?
There is a REST API and an MCP server with 4 tools, and the API plan is billed on the audio you send. The Pro plan carries no minute limit of audio a month.
How do the 99+ languages work together?
Athe recording language is detected automatically. Set 99+ languages yourself, or leave the detection on and let the file decide.
What is kept about the participant?
The file and the transcript stay in your account. Nothing is published for you, the audio is never used to train models for other customers and never sold.
Does the service touch the ethics approval?
No. The approval is your document and your institution is your business. We only take the file you upload and give back the text of what was said in it.
Reading: 7 minutesUpdated Transcribsi editorial team
Drop a recording — the text arrives in a minute
Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.