Pepys
12,438,517minutes transcribed

Qualitative research transcription, ready for analysis

Upload an in-depth interview or focus group – get a speaker-labeled transcript plus the themes, verbatim quotes, and a Q&A breakdown your synthesis already needs.

or paste a link
InstagramTikTokYouTubeFacebookSpotifyApple Podcasts

60 min free · no card required · we never train on your audio

PodcasterJournalistContent creatorResearcherStudent
Trusted by 100,000+ creators, podcasters, journalists & researchers

How do you transcribe sessions?

To transcribe a research interview, upload the recording or paste its link and Pepys returns a speaker-labeled, timestamped transcript in minutes – plus AI-drafted themes, verbatim quotes, and a moderator/participant Q&A breakdown ready for coding. It's pay-as-you-go with no subscription, and credits never expire.

Made for research teams

The recording is never the deliverable. Between fieldwork and the readout sits the slowest part of the job: transcribing every in-depth interview and focus group, keeping moderator and participant straight, hunting back through the audio for the one verbatim that proves a theme, and turning a folder of recordings into something you can actually code. That work is sitting inside audio you already captured – it just needs to become accurate, attributable text.

In practice, qualitative research transcription is where the coding frame lives or dies. Word-level timestamps mean a quote you flag stays welded to its moment in the recording, so a teammate can jump to the audio and confirm tone before it lands in the findings. Speaker labels keep the probe distinct from the response, full-text search pulls every mention of a code across a study in seconds, and a JSON export drops the clean transcript straight into your CAQDAS or coding sheet instead of a retyped paste.

  • Coded themes for synthesis

    Recurring themes surfaced from each transcript so coding starts from a draft instead of a blank screen.

  • Verbatim quotes, attributed

    Participant quotes pulled and tied to a timestamp, ready to drop straight into the readout deck.

  • A searchable transcript bank

    Every interview becomes full-text searchable, so you can find the one line that proves a finding in seconds.

  • Moderator and participant, separated

    Speaker labels keep the moderator's questions distinct from the participant's answers across the whole session.

Built in, not bolted on

What participants said, the quotes that back it, and a question-by-question recap

Every sessionis analyzed automatically the moment it’s transcribed. Here’s a real sample, run through it.

idi-p07-budgeting-churn.mp3AI analysis, built in
AI analysis

Why She Churned: A Budgeting App That Judged Instead of Coached

In this in-depth interview, a former user explains why she abandoned a budgeting app despite a strong first week. The onboarding hooked her by revealing where her money actually went, but the daily over-budget warnings made her feel judged and guilty rather than supported. She contrasts it with a fitness app she has kept for two years because it celebrates small wins. Price was not the driver – she would happily pay for a tool that made her feel in control. Her core insight is that the product is a great mirror and a terrible coach.

Themes

Churn driven by emotion, not featuresPunishing feedback vs. encouraging feedbackStrong onboarding undercut by the daily experienceWillingness to pay decoupled from priceMoney as an emotional, not arithmetic, problem

Notable quotes

  • The app treated it like a math problem, and the part I needed help with was the guilt, not the arithmetic.
  • I would make it encouraging instead of punishing.
  • I cancelled because it made me feel worse, not because of the price.
  • I'd tell them it's a great mirror and a terrible coach.

Q&A

Walk me through the day you decided to stop using it.

There was no single day – it was a slow fade. She would open the app, see red over-budget warnings, feel judged, and close it again, until she stopped opening it and eventually deleted it while cleaning up her home screen one Sunday.

Before the warnings turned you off, was there a moment early on where it actually worked?

The first week worked well. Connecting her bank account revealed where her money actually went and surfaced how much she spent on takeout, which hooked her. It was the daily nagging afterward that wore her down.

If you could have changed one thing to keep yourself as a user, what would it be?

Make the feedback encouraging instead of punishing – tell her she is under budget, not only scream when she is over. She drew a direct contrast with a fitness app she has stayed with for two years precisely because it celebrates small wins.

Did the price play into leaving?

No. The app cost around nine dollars a month, and she says she would happily pay that for something that made her feel in control. She cancelled because it made her feel worse, describing it as paying a subscription to be scolded.

Clean, speaker-labeled, click-to-seek

0:00 / 2:34

Ask, don’t scrub

Ask the transcript anything.

An hour-long recording? Don’t skim it – ask. Every answer stays grounded in your transcript and cites the exact timestamp, so you can jump to the moment and check it yourself.

idi-p07-budgeting-churn.mp3Ask AI

What actually drove her to churn – was it the price?

No, price wasn't the driver. She says the app cost around nine dollars a month and she'd happily pay that for something that made her feel in control – she cancelled because it made her feel worse, like she was paying a subscription to be scolded.

Cited1:53

Did the product ever work for her, or was it bad from the start?

The first week worked well – connecting her bank account showed her where her money actually went, including how much she spent on takeout, and that hooked her. What wore her down was the daily over-budget warnings afterward, which she describes as the app being a great mirror and a terrible coach.

Cited1:022:18
Ask anything about this transcript…

Grounded in your transcript – if the answer isn’t in the audio, it says so instead of guessing.

Who said what

Speaker labels that survive cross-talk

Automatic speaker diarization. Two people, four people, cross-talk and interruptions – interviews, panels, messy meetings. Pepys keeps each voice on its own line instead of blurring them into one, so you never rewind to figure out who was talking.

Reporter

So the festival nearly didn't happen this year–

Mara Okonkwo

–it almost didn't. We lost the venue three weeks out.

Reporter

Three weeks? How do you even start to–

Mara Okonkwo

You call everyone you know. The whole town pitched in.

Reporter

And that's how it ended up in the park.

Record in any language – 99+ detected automatically

Works with the platforms you live in.

Paste a link from YouTube, TikTok, Instagram, Facebook, Spotify, or Apple Podcasts – or drop in any audio or video file. We transcribe it once, then you export it however your workflow needs.

  • YouTubeYouTube
  • TikTokTikTok
  • InstagramInstagram
  • FacebookFacebook
  • SpotifySpotify
  • Apple PodcastsApple Podcasts
  • or any file

Export to any format

  • TXT
  • Markdown
  • DOCX
  • PDF
  • SRT
  • VTT
  • JSON

Most useful for research teams: Transcript (DOCX) · TXT · SRT · VTT · PDF · JSON

Timestamps, speaker labels, and subtitle timing carry through to every export.

How qualitative research transcription works

Upload or paste a link

Drop your session or paste its link – any audio or video, in any language.

Get your transcript

A clean, speaker-labeled transcript with AI notes tuned to your format, ready in minutes.

Edit and export

Fix anything inline, then export to SRT, VTT, TXT, DOCX, PDF, or JSON.

Why research teams pick Pepys

  • No subscription – pay per session, and credits never expire, so a study that spans quarters never races a billing clock.

  • Themes and verbatim quotes are drafted for you, not a separate paste into another tool.

  • We never train on your audio – participant recordings stay yours, which matters for consent and ethics review.

  • Speaker labels keep moderator and participant cleanly apart, the way coding software expects.

What research teams say

  • I work across three languages and it detected each one correctly without me changing a single setting. The timestamps line up to the word – exactly what my research needs.
    Priya N.Linguistics PhD candidate · Trustpilot
  • Every user interview comes back as a clean, searchable transcript I can tag and quote directly in my reports. Synthesis used to be the slowest part of my week and now it's an afternoon. The speaker labels alone are worth it for me.
    Sofia L.UX researcher · G2
  • multilingual focus groups, transcribed and translated into one working language so i can compare responses side by side. used to wait days on a vendor for this – now its same-day.
    Erica B.Erica B.Market researcher · Product Hunt

Qualitative research transcription – questions, answered

How do I transcribe a research interview or focus group?

Upload the recording or paste its link, and Pepys returns a speaker-labeled, timestamped transcript in minutes – along with AI-drafted themes, verbatim quotes, and a Q&A breakdown you can take straight into coding.

Can it separate the moderator from participants?

Yes. Speaker diarization labels each voice, so a moderator-led interview or a multi-person focus group comes back attributed rather than as one block of text. You can rename "Speaker 1" to the participant ID and it updates everywhere.

Do you train models on our participant recordings?

No. We never train on your audio. Recordings and transcripts stay yours, which keeps you on the right side of participant consent and your ethics or IRB obligations.

Can it handle multilingual fieldwork?

Yes. It auto-detects the spoken language across 99+ languages, so interviews run in different languages come back transcribed in the original – and you can deliver a translated version with timing preserved for cross-market comparison.

What can I export for analysis?

A DOCX or PDF transcript, plain text, SRT and VTT for clip review, and JSON for piping into your coding or CAQDAS workflow. Timestamps stay attached so every quote keeps its source.

How accurate is it with accents, crosstalk, and filler?

It handles a range of accents and overlapping speech, and timestamps line up to the word. Anything it gets wrong you can fix inline in the editor before you export, so the verbatim you quote is the verbatim that was said.

Is there a subscription, and what does it cost?

No subscription. Pepys is pay-as-you-go – buy a block of minutes, use them across a whole study whenever you need, and the credits never expire. You can start free with 60 minutes, no card.

More industries

Turn your next round of fieldwork into coded transcripts and quotes – and pay only for the sessions you run.

Pay as you go – credits never expire, nothing to cancel. Or start free with 60 minutes, no card.