Free tool
Focus group transcription
Drop in a session recording for free focus group transcription: the whole discussion comes back as text with a time on every paragraph, so you can match comments to participants from your notes and draw out the themes.
- Free
- Times through the discussion
- Nothing is stored
Drop your focus group recording here, or Choose your focus group recording
MP3, M4A, WAV, AAC, FLAC, OGG or WebM. Up to 30 minutes of each file, 3 free files a day. The language is detected automatically.
Running groups often? NoteOwl transcribes each session on your Mac with unlimited transcription and labels the speakers, ready for you to name.
How it works
How do you transcribe a focus group?
Drop the session recording here. The converter cuts it at pauses, transcribes it piece by piece and writes the discussion out in timed paragraphs. It doesn’t name the speakers, so match each comment to a participant using the note-taker’s log and the times.
- 1
Add the session
The recorder’s MP3 or WAV from the room, or the MP4 or M4A an online session saved.
- 2
Let it run
The first 30 minutes of each file are transcribed in order, with the time at the start of each paragraph.
- 3
Attribute and save
Download the Word file, then add the moderator’s and participants’ codes from your notes.
Why are focus groups so hard to transcribe?
Several people share one microphone, and they don’t take turns politely. Participants agree over each other, laugh, start side conversations and trail off, and the quiet ones often sit farthest from the recorder. Crosstalk is where any transcript loses the most words.
Overlapping voices
When two people speak at once, the transcript often keeps only the louder one. Note crosstalk during the session so you know where to listen again.
Quiet participants
A soft voice at the far end of the table may come back incomplete. A second recorder at that end helps.
“Mm-hm” around the table
Murmured agreement carries meaning in a group but may not make it into the text. Notes on who nodded fill the gap.
Many similar voices
Without labels, telling eight people apart from text alone is guesswork. Introductions and a seating plan make it possible.
How do you record a focus group for a clean transcript?
Put an omnidirectional recorder in the middle of the table, with a second one at the far end as a backup. Ask everyone to say their first name the first time they speak, and have the moderator repeat any comment the room may have missed.
- 1
Start with introductions
Each participant says their first name and one fact about themselves. Those opening lines are your key to every voice that follows.
- 2
Keep a speaker log
The note-taker writes the time and the first few words whenever the speaker changes. It’s what turns a transcript into attributed data.
- 3
Online, use the call’s own recording
For a group held on Zoom or Teams, transcribe the recording the call app saved, not a microphone near the laptop speaker.
How do you tell participants apart without speaker labels?
With the speaker log and the times. Every paragraph starts with its time, so line it up with the note-taker’s entries, then type a code at the start of each turn: M for the moderator, P1, P2 and so on for participants.
On a Mac, NoteOwl handles this step with speaker diarization: it labels each speaker in an imported recording, and you give each label a name once.
- 12:40 · P3: “I stopped using it when…”
- Where it lands in the transcriptThe paragraph at 12:40 that opens with those words
- 15:05 · M: next question
- Where it lands in the transcriptA new topic: start a new section in your analysis
- 15:20 · P1 and P5 talk over each other
- Where it lands in the transcriptA paragraph to check by ear before you attribute it
| Speaker log entry | Where it lands in the transcript |
|---|---|
| 12:40 · P3: “I stopped using it when…” | The paragraph at 12:40 that opens with those words |
| 15:05 · M: next question | A new topic: start a new section in your analysis |
| 15:20 · P1 and P5 talk over each other | A paragraph to check by ear before you attribute it |
What do you do with a focus group transcript?
Sort it by the discussion guide. Use each of the moderator’s questions as a heading, gather what every participant said under it, then compare across groups. The times let you go back to the audio for tone, which text alone loses.
Themes
Tag each comment with the themes it touches, and count how many participants raised each one, not just how many comments.
Quotes
Copy the strongest comments word for word with their times, so anyone can check them against the recording.
Group differences
Set one question’s answers from every group side by side to see where groups agreed and where they split.
The report
Findings first, then the evidence: a few quotes per theme, each marked with its group and participant code.
Want each participant labelled automatically on your Mac?
NoteOwl transcribes focus group recordings on your Mac with unlimited transcription and labels each speaker, so you name the voices once instead of attributing every line by hand. Sessions stay on your Mac, searchable across a whole study. See how researchers use it.
FAQ
Questions about focus group transcription
Is this focus group transcription free?
Yes, with no sign-up. You can transcribe 3 session recordings a day, up to 30 minutes of each, and download Word, text or subtitles.
How long a session can it transcribe?
The first 30 minutes of each file. Many groups run longer, so split the recording and transcribe the parts one after another, noting where each began, since every part’s times start at zero. NoteOwl on a Mac transcribes a whole session at once.
Does it label the moderator and each participant?
No. The free converter writes one timed transcript with no speaker names. Add codes from your speaker log, or use NoteOwl on a Mac, which labels each speaker in the recording.
Can I transcribe an online focus group?
Yes. Download the session’s recording from Zoom, Teams or whichever platform you used, then drop the MP4 or M4A here. Only the sound is used.
Can it handle a group that speaks two languages?
Yes. The language is identified roughly every 30 seconds, so a discussion that moves between languages is written in each one as it goes. Within a single half-minute piece, the main language is the one written.
Is the session recording kept?
No. Your browser sends short pieces of the sound over a secure connection to a cloud AI transcription service, which transcribes and discards them. NoteOwl keeps neither the recording nor the transcript.
How accurate is it when people talk over each other?
One clear voice at a time comes back close to word for word. Where people talk at once, expect gaps or one voice winning, so check those stretches by ear, using the times to find them.
Keep exploring
More audio to text tools
- Audio to text converterTranscribe audio to text free: drop in an MP3, M4A, WAV or other recording and get an AI transcript with times, as text, SRT, VTT or Word; nothing is stored
- Interview transcriptionTranscribe a recorded interview free in your browser, with times to find each quote; text, SRT, VTT or Word; nothing is stored
- Lecture transcriptionTranscribe a recorded lecture or class free in your browser, with times, ready to turn into study notes; nothing is stored
- Podcast transcriptMake a podcast episode transcript free in your browser, with times for show notes and SRT subtitles; nothing is stored
- Voice memo to textTurn a phone voice memo into text free in your browser, from an iPhone or Android recorder; nothing is stored
Related
- Speaker recognitionTranscripts that say who said what: speakers separated automatically, and a voice named once is recognised in later meetings, on your Mac
- AI note taker for researchersUser interviews transcribed speaker by speaker, past recordings imported, searchable and askable across a study; nothing uploaded
Group discussions into notes, privately on your Mac
Free to download. Runs entirely on your Mac.
macOS 26 or later · Apple Silicon (M1 or newer) · 16 GB memory minimum, 24 GB recommended
About NoteOwl
NoteOwl is private AI meeting notes and memory for the Mac. It captures meetings without a bot joining the call, labels every speaker, writes summaries and action items, and builds a memory of every conversation that you can chat with — fully offline, because the speech recognition and the AI run on the Mac itself. Dictation into any app is free; Pro is a one-time purchase.
NoteOwl pricing: what’s free and what Pro adds
Last updated