For doctoral research
Transcription for a dissertation, and the years after it
Upload the round, get speaker-labeled transcripts you correct in place, and keep every interview searchable through analysis, writing, and the viva. Twenty audio hours a month is $12.
P14 · Wave 2
Did anyone at the department ever raise it with you directly?
Nobody says it out loud. You learn it from who gets asked back.
Informal gatekeepingBy the second year you stop needing to be told.
Open the sample workspaceNo account needed.
The arithmetic of a doctoral round
Twenty interviews at an hour each is twenty audio hours. At ninety minutes it is thirty, and a longitudinal design with a second wave doubles whichever number you started from. Typing those yourself runs at roughly four hours of work per hour of audio, which is where the months go.
Every route below covers the same twenty hours. The gap between them is wide enough that it is worth knowing which one you are choosing before the fieldwork starts.
Phonotheca Solo
20 audio hours a month, archive and search included
$12 a month
Local Whisper
noScribe free or MacWhisper at €65 once, on your own machine
€0 to €65
NVivo Transcription credits
20 hours at €34.77, on top of a licence
€695
Human transcription
20 hours at Rev's published $1.99 an audio minute
$2,388
Vendor prices as published in August 2026, each in the currency its vendor lists.
What the workspace does after the transcript arrives
Correct the passages you quote
Fix a word, split a turn where an interjection landed in the wrong speaker, and rename a participant once so the change carries into every quote and export from that recording. Corrections happen in place rather than in a downloaded copy that then diverges.
Find the line you half-remember
Search runs across every interview at once, filtered by project, speaker, or tag. The retrieval failure that eats an afternoon, knowing someone said it and not which transcript, is the thing this is built to end.
Leave with the whole project
Export as Markdown, CSV, or a REFI-QDA .qdpx archive that opens in NVivo, ATLAS.ti, MAXQDA, and Dedoose. Tags travel as codes and coded passages map onto the turns they cover.
The quote you know exists
Analysis chapters are written out of half-remembered lines. Someone described the thing precisely, in one of thirty recordings, eighteen months ago. Search covers the whole corpus at once and returns the passage with its speaker and timestamp, so the citation is settled at the moment you find it.
Tags collect the same theme across every interview, which is the shape a coding chapter needs long before it becomes formal codes. Saved searches keep the questions you return to.
See the whole-study workflow3 passages across 3 interviews
01:05:03 · 1:22:05
Nobody says it out loud. You learn it from who gets asked back.
00:38:41 · 1:11:26
There was never a rule about it, more like who gets asked back to the reading group.
00:52:10 · 1:03:55
My supervisor called it fit, which really meant who gets asked into things.
Nobody says it out loud. You learn it from who gets asked back.
P14 · 01:05:03 · Wave 2 interviews
Copy with citationWhat the ethics application asks
Committees ask a fixed set of questions. Who processes the recordings, which countries they sit in, how long anything is kept, who else can see the material, whether a data processing agreement is signed, and whether the audio is used to train models.
Those are answered in writing, in that order, with the sub-processor list and the countries named, on a page you can link from the application itself. Audio is processed in the EU and the transcription service holds no copy once the transcript returns.
Read the data governance statementPriced for a fieldwork season
Subscribe while interviews are arriving, stop when they are not. The archive stays.
Free
Enough to see a corpus take shape.
$0forever
3 audio hours, once
- Unlimited projects, interviews, and tags
- Full-text search across every transcript
- Markdown, CSV, and REFI-QDA export
Solo
A study's worth of interviews, every month.
$12per month
20 audio hours a month
- Hours reset on your renewal date
- Everything in Free, never gated
Pro
RecommendedFor a fieldwork season or a whole cohort.
$29per month
50 audio hours a month
- Hours reset on your renewal date
- Everything in Solo
Prices are in USD, tax added at checkout. Compare the plans
Common questions from doctoral researchers
- How much does transcribing a dissertation cost?
- Twenty interviews of an hour is twenty audio hours, which is $12 on Solo. The same round is about €695 in NVivo transcription credits and about $2,388 at Rev's published human rate. Local Whisper through noScribe is free if you have the machine time and will fix speaker labels yourself.
- My supervisor says transcribing by hand is part of the analysis.
- That position has a real methodological basis, and its own rebuttal sits inside it. The value claimed for manual transcription is close listening, and close listening survives correcting a machine transcript against the audio, which is what most people do anyway on the passages that matter. What hand-typing does not produce is a corpus you can search. Audio cannot be searched, and a Word file per interview is searched by opening files one at a time.
- Can I use it for only part of the year?
- Yes, and it suits how doctoral funding actually arrives. Subscribe for the fieldwork months when interviews are coming in, then stop. Hours pause, and the archive, search, tags, and export keep working, so the corpus stays usable through the writing years without a subscription running against it.
- What do I tell my ethics committee?
- The data governance statement is written for exactly that form. It names the processor and the countries, states retention, confirms a signed DPA and that nothing trains models, and gives wording you can paste into a consent form or an application. It is at /data-governance.
- My university gives me NVivo for free. Why pay for this?
- Then keep NVivo for the coding. Site licences cover the licence and rarely cover transcription, which NVivo meters separately at €34.77 an audio hour. Transcribe here, export the project as a REFI-QDA archive, and code it in the software your department supports.
- Will the transcripts be accurate enough to quote?
- Accurate enough to work from, and every quoted passage should still be checked against the audio, which is a click from the line. On a hard two-speaker recording, a 1983 oral history cassette with rapid turns and a third voice in the room, speakers held with roughly 95% of turns correctly attributed. Short interjections absorbed into the wrong turn were the one failure mode, and those are fixed by splitting the segment.
- What about interviews that are not in English?
- Set the spoken language on each recording rather than leaving it to detection. Detection misreads a strong accent as another language, and when that happens the features that language does not support drop silently, speaker labels among them. The accents and languages page lists what to expect per language.
- How do I get quotes into the thesis with citations?
- Copy a passage and the speaker, source recording, and timestamp come with it. An appendix of transcripts exports as Markdown or CSV.
Keep reading
Start with the pilot interview
Three audio hours are free, and the archive stays open whether or not you subscribe.