Interview transcription with AI
Upload a recording and usually within a few minutes you get a transcript with separated speakers and timestamps. Correct it while listening, pseudonymise names and places, and download it as Word, text, subtitles or with MAXQDA timestamps.
Transcribe 30 minutes free. No card needed.
How it works
- 1Upload
MP3, M4A, WAV, FLAC, OGG or WebM, in English or German.
- 2Get the transcript
Speakers are separated and every passage gets a timestamp. We email you when it is ready.
- 3Check and pseudonymise
Correct it while listening, name the speakers, replace names and places.
- 4Download or analyse
Word with line numbers, text, SRT, VTT, CSV or MAXQDA timestamps. Or code it in Themera.
What the tool does
Automatic speaker separation. Rename to I, P1, P2, merge speakers or reassign single passages.
Every passage has a timestamp. One click plays it in the recording.
Fix a misheard name or term across the whole transcript at once.
Find, review and replace names, places, organisations and contact details, including inflected German names.
For your appendix, with speaker labels and optional timestamps.
Timestamps as #hh:mm:ss-x# at paragraph ends, subtitle files and a spreadsheet with a speaker column.
One click once you have checked the transcript. The transcript stays.
If you want, Themera codes the transcript with verbatim evidence, leaving the interviewer’s questions out.
Use cases
Transcribe interviews for a dissertation or thesis: time budget, transcription conventions, an appendix with line numbers and a sentence for your methods chapter.
Transcribe focus groups with automatic speaker separation, then assign every passage to the right participant, mark the moderator and export to Word, CSV or MAXQDA.
Recording sources
Transcribe a Zoom recording: where the M4A audio file is saved, what to expect from cloud recordings, and how to get a speaker-labelled transcript for your interview.
Transcribe a Microsoft Teams recording: where OneDrive and SharePoint keep it, how to turn the MP4 into audio, and when your own transcript beats the Teams one.
Transcribe voice memos from an iPhone or Android phone: save the recording as a file, upload it and get text with speakers and timestamps you can edit and export.
Formats
Export and privacy
Prepare a transcript for MAXQDA: timestamps as #hh:mm:ss-x# at the end of each paragraph, speaker labels, importing with the audio file, and working with VTT or SRT.
Where Themera stores and transcribes recordings, how to delete audio and pseudonymise transcripts, and a consent checklist to work through before you upload.
What automatic transcription cannot do
- Be error-free. Check every transcript against the recording before you quote or analyse it.
- Capture crosstalk fully: when people overlap, usually only one voice is transcribed.
- Rescue poor recordings. Noise, echo and distance from the microphone noticeably lower accuracy.
- Mark pauses, stress and intonation as conversation analysis requires. That remains manual work.
Frequently asked questions
What does transcription cost?
The free account includes 30 transcription minutes, no card needed. More minutes come with the one-off Thesis plan and the Solo and Pro subscriptions; current allowances are on the pricing section.
Which files can I upload?
MP3, M4A, WAV, FLAC, OGG and WebM up to 50 MB per file. Convert video to audio first, as explained under audio to text.
Which languages are supported?
English and German. You choose the language of the recording when you upload.
How accurate is it?
With a clear recording it reads well, but it is never error-free. Plan a correction pass with the recording; the editor is built for exactly that.
Where are my recordings processed?
Stored in the EU, transcribed by Mistral AI in Paris, not used for training. Details under GDPR and transcription.
Further reading
Transcribe 30 minutes free
No card needed. Upload a recording, check the transcript, download it as Word, text or for MAXQDA.