For university students
Transcribe interviews without typing every word.
Upload a recording to transcribe interviews automatically, with speaker labels and a timestamp on every speaker turn. Correct the draft while you listen, then download it in Word and spend your time on the analysis.
7 free hours of audio a month · No card needed
Free hours need a confirmed university email from Denmark, Germany, Austria, Switzerland, Sweden or Norway. Otherwise, hours cost €2 each.
Try an example
Interview about student life
English · Two speakers · Fictional excerpt
Interviewer 00:12
What helps when an assignment starts to feel overwhelming?
Student 00:18
It makes a real difference that we can talk it through in our study group. Then I’m not the only one trying to find a way forward.
Interviewer 00:31
Can you tell me about a specific time when that helped?
For students at the University of Copenhagen, LMU Munich, ETH Zurich, the University of Vienna, Uppsala University and the University of Oslo, and many more in six countries.
See it in action
Six short steps, from consent and recording to transcription, analysis and sharing. Pick a step below the video to jump straight to it.
Jump to a step
How it works
Let the tool write the first draft. You listen through, fix mistakes and pick the quotes you want to use.
Step 1
Sign up and upload your recording
Create your account and upload the interview as mp3, m4a, wav or mp4. There is nothing to install.
Step 2
The AI transcribes it with speaker labels
You get a draft with speakers and timestamps. How long it takes depends on the length of the recording. Choose a cleaned-up or word-for-word style.
Step 3
Edit and export to Word
Correct words, rename speakers and listen to the audio as you read. Download a .docx with timestamps, ready for your appendix and your analysis.
Recorded on your phone or a voice recorder? See how interview transcription works, from file to finished text.
Pricing
No catch. Here is who gets what, and why.
Students at universities in Denmark, Germany, Austria, Switzerland, Sweden and Norway get 7 hours of audio free every month. Sign up with your university email and confirm it with a code we send you. That’s all it takes: no card, no trial and no subscription that starts after your deadline.
Everyone else can sign up too and buy hours: €2 per hour of audio, or Spirelight Plus at €14 a month with 10 hours included. Hours you buy separately never expire; Plus hours renew each month.
Spirelight Student is made by Spire Light, a small Danish startup with roots at Roskilde University. We build analysis tools for companies, and the student edition is our way of giving something back. We have typed out interviews word by word late at night ourselves, and we would rather you didn’t have to.
- 7 free hours a month with a confirmed university email
- €2 per hour, or €14 a month for 10 hours with Spirelight Plus
- No card for the free hours, no trial and no watermarks
Features
Transcription, recording, coding and analysis live together in one project. This is what it looks like.
Transcribe
Audio and video to text, with speaker labels
Upload audio or video and get a clean transcript with the speakers separated and a timestamp on every speaker turn. Choose a cleaned-up or word-for-word style to suit your method. Want the details? Read how interview transcription works, from file formats to proofreading.
- mp3, m4a, wav and mp4, including recordings several hours long
- Speaker labels you can rename across the whole document
- Built-in player, so you can listen while you correct
Interview 03 · Emma W.
Saved just now · 42:18 · English
Interviewer
00:12
Could you start by telling me what your work involves day to day?
Emma W.
00:18
I’m a project manager in the council’s customer services team. I’ve been working on digital services for about eighteen months.
Interviewer
00:31
And which tools do you use when you test new services with residents?
Record
Record interviews and video calls with consent built in
No recording yet? Start a call from your library and send the link to your participant. They sign a consent form in the browser before recording starts, and when the call ends, one click turns it into a transcript. Consent templates are in English and Danish.
- Video and audio in the browser, no extra software
- Digital consent forms with a copy for the participant
- One click from finished call to transcript
Code
Qualitative coding right in the transcript
Build a colour-coded codebook for the project, select a passage and apply a code with a keyboard shortcut. The code overview gathers every coded excerpt from all your interviews into one table.
- One codebook across all interviews and documents
- Code overview with filter, search and CSV export
- A count for each code, so you can follow your themes
We had to build trust with residents before anyone would test anything. The biggest barrier was the old case system, so we started prototyping in a simple shared spreadsheet instead.
Codebook
- Trust12
- Barriers7
- Tools5
Select a passage and press a shortcut to code it.
Analyse
AI summaries and chat with your sources
Ask a question and get answers drawn from every interview and document in the project. Each statement cites the passage it comes from, so you can check everything. Summarise a whole interview in one click.
- Answers with citations to your own data
- An AI summary for each interview
- PDF and Word files can be used as sources
1Interview 01 · 21:38
2Interview 03 · 04:11
Hand in
Export to Word with speakers and timestamps
Download the transcript as a .docx with speakers and timestamps, ready for your appendix. Share the project with your group and your supervisor, and work in one up-to-date document instead of emailing files back and forth.
- Word, plain text or SRT, with or without notes
- Share as viewer, commenter or editor
- Remove access at any time
Share “Council services study”
- SEditor
Supervisor
a.taylor@uu.se
- PCommenter
Priya (group)
priya.n@student.uu.se
- LViewer
Link
Anyone with the link
Privacy
Interview recordings can contain names, workplaces and personal stories. Here is exactly where they are processed and stored, so you can check it against your university’s rules.
Processing and storage
Transcription in the EU, storage and AI in the US
Transcription runs on AssemblyAI’s servers in the EU. Your files, transcripts and account are stored with Google Firebase in the US. If you use AI summaries or chat, the text of your transcripts and documents is processed by OpenAI in the US. We don’t offer a data processing agreement (DPA).
Consent
Consent forms built in
Send a digital consent form before you record. Your participant signs in the browser, gets their own copy and can withdraw consent later. The templates are in English and Danish.
Control
You decide what happens to your data
Your recordings are not sold and are not part of any commercial data strategy. Delete files at any time; deleted files can be restored for 30 days.
Read the terms and privacy policy, and follow your university’s rules on consent and research data before you upload.
Questions and answers
Is it really free? What’s the catch?
For students, yes. Confirm a university email from Denmark, Germany, Austria, Switzerland, Sweden or Norway and you get 7 hours of audio free every month, with no card and no trial. That is enough for most theses and dissertations. Need more, or studying somewhere else? Buy hours at €2 each, or take Spirelight Plus at €14 a month with 10 hours included. The only catch: free hours reset each month and don’t roll over.
Who gets the free hours?
Anyone can create an account. Students at universities and colleges in Denmark, Germany, Austria, Switzerland, Sweden and Norway get 7 free hours a month once they confirm their university email with the code we send to it, for example at the University of Copenhagen, LMU Munich, the University of Vienna, ETH Zurich, Uppsala University or the University of Oslo. If your university isn’t recognised, write to us and we’ll look into adding it. With any other email you can buy hours or a subscription.
Which file formats can I upload?
Common audio formats including mp3, m4a and wav, plus mp4 video. That covers recordings from iPhone, Android, voice recorders and most meeting tools. You can also add PDF and Word files to a project, so your literature and notes sit next to your interviews.
How accurate is it? Do I still need to proofread?
Accuracy depends mostly on the recording. Clear audio with one person speaking at a time gives the best draft, while background noise, overlapping speech, strong accents and specialist terms cause more errors. Treat the transcript as a first draft: listen through it in the built-in player, check names and terms, and verify every quote you plan to use.
Where is my data processed and stored?
Transcription runs on AssemblyAI’s servers in the EU. Your files, transcripts and account are stored with Google Firebase in the US. If you use AI summaries or chat, the text of your transcripts and documents is processed by OpenAI in the US. We don’t offer a data processing agreement (DPA). Whether you may upload interviews with personal data depends on your university’s rules, so if you are unsure, ask your supervisor or your university’s data protection officer. Our terms and privacy policy have the details.
Is Spirelight in English, and which languages can it transcribe?
Yes, it is in English: the website, sign-up, billing, emails, the editor, the help page and the terms. Consent form templates are in English and Danish. English audio transcribes well, and you can also transcribe German, Danish, Swedish, Norwegian and many other languages, or let Spirelight detect the language for you.
How do I transcribe an interview?
First decide whether you need a word-for-word (verbatim) transcript or a cleaned-up one; your method or your supervisor usually settles that. Then upload the recording, and Spirelight drafts the transcript with speaker labels and timestamps. Go through the text while you listen, correct mistakes, rename the speakers and export to Word for your appendix.
How long does it take to transcribe an interview?
Typing out an hour of interview by hand usually takes several hours. With Spirelight the draft is ready after a processing time that depends on the length of the recording, so most of your time goes into listening through and correcting it. Plan for that, especially if you need a word-for-word transcript.
Get started
Stop typing out interviews by hand
Your next step is simple: sign up with your university email and upload your recording. You get a draft you can correct and download in Word.
7 free hours a month with an eligible university email · No card · No trial