We show the transcript here and email you the text, SRT and VTT, so you still have them tomorrow.
Free: 2 recordings a month, up to 1 hour each. Your first needs no account; after that, sign in for the rest. Need more or longer? Hour packs from $1.99, paid once.
2 free a month · then from $1.99
Prices
Pay once · never expires · no subscription
Hour pack
1 hour
$1.99 once
about $1.99 an hour
Hour pack
5 hours
$4.99 once
about $1 an hour
No subscription. Pay once for the hours you need; they never expire.
Students with a .edu or .ac.uk email: 5 hours for $3.99. Sign in with your school email.
More options15 hours, 40 hours, a monthly plan and the API
Hour pack
15 hours
$9.99 once
about 67¢ an hour
Semester packBest value
40 hours
$24.99 once
about 62¢ an hour
20 hours every month · $7.99/mo
For weekly shows and lectures. Cancel anytime.
Transcribing from code? The transcription API has its own hours.
No account needed to buy a pack. Pay with your email, and we send you a link to sign in with your hours waiting.
Every pack: recordings up to 3 hours, no limit on how many, Word and Notion export, notes with timestamps. A 47-minute lecture uses 47 minutes.
How do I transcribe a focus group with speaker labels?
Record the session, with everyone’s consent, on one recorder in the middle of the table (or the Zoom or Teams recording for an online group), then upload the file here: MP3, M4A, WAV, MP4 and most other audio or video formats, up to 600MB. The transcript labels each voice Speaker A, B, C and so on, starts a new paragraph at every change of speaker, and comes with .txt, .srt and .vtt files. It cannot know names: map the labels to participant codes afterwards. Free: two recordings a month, each up to 1 hour. With Slice Transcribe hour packs (1 hour for $1.99, 5 hours for $4.99, 15 hours for $9.99 or 40 hours for $24.99), paid once with no subscription, sessions can run up to 3 hours, every turn gets a timestamp, you name the speakers once and the names go into the Word file and every download, and you can move a paragraph to the right speaker. A session uses its length rounded up to the minute. It is an automatic transcript, not a human one: with many voices, two similar voices can share a label and crosstalk loses words, so check the quotes you will use against the recording.
Free, or hour packs
| Feature | Free | Hour packs |
|---|---|---|
| Price | Free | From $1.99, paid once |
| Longest recording | 1 hour | 3 hours |
| How many | 2 recordings a month | As many as your hours cover |
| How hours are counted | Each recording counts as one of the month’s free ones | Each recording’s length, rounded up to the minute: a 47-minute lecture uses 47 minutes |
| Do they run out? | They reset on the 1st of each month | Never: the hours do not expire |
| Account | Not for your first one; then a free account | A free Slice account |
| Downloads | .txt, .srt, .vtt | Word (.docx), Copy for Notion, notes with timestamps, plus .txt, .srt, .vtt |
| Speaker labels | Yes, when more than one person talks | Yes, when more than one person talks |
| How long the text is kept | 30 days | On your account until you delete it |
| Audio kept | 7 days, then deleted | 7 days, then deleted |
Paid once by card through Stripe. No subscription, nothing renews, and the hours never expire. Any pack can be bought as a gift: you get a link to send, and the hours go to whoever redeems it. More on Slice Transcribe.
What you get
- Every voice labelled. Speaker A, Speaker B, Speaker C, with a new paragraph whenever the speaker changes, in the text and in the subtitle files.
- Timings. The free .srt and .vtt time every line. With an hour pack the notes and Word file start each turn with its time, like [41:05], so a quote can be traced back to the moment.
- Names instead of letters, with an hour pack: name each voice once, from a line each one said, and every download uses the names.
- Corrections, with an hour pack: fix a word, or move a paragraph to the right speaker, and the downloads follow.
- Any language it can detect. Accuracy is best in English.
How far to trust the speaker labels
A focus group is one of the hardest recordings for automatic speaker labels: many voices, short replies, people agreeing over each other. AssemblyAI, which does the speech-to-text, lists the same three weak points in its own documentation:
- Similar voices can end up under one label.
- Crosstalk, two people at once, lowers accuracy.
- Short contributions. Someone who only says “yes, same” may be folded into another speaker. AssemblyAI says each speaker is best heard for about 30 seconds without interruption.
So treat the labels as a first pass. Longer, uninterrupted answers are where they are most reliable; quick back-and-forth needs a listen. Check every quote you will publish or code against the recording.
Recording a session that transcribes well
- One good recorder or phone flat in the middle of the table, away from laptop fans and the air conditioning. For a long table, a second recorder at the other end, uploaded separately.
- Ask people to speak one at a time for the questions that matter. A moderator saying the participant code before a reply (“P3, go ahead”) makes matching labels to people quick.
- Have a note-taker log who speaks first in each answer, with the time. Those few lines turn Speaker D into P5 in seconds.
- Draw a seating plan with codes, not names.
From transcript to analysis
Download the Word or .txt file and work on your copy: replace the labels with participant codes, take out names and anything identifying, then import it into your analysis tool or code it by hand. If you are working to an ethics protocol, the same steps as for research interviews apply: consent that covers a transcription service, anonymising before analysis, and deleting our copy once you have yours.
How to transcribe a focus group
- 1Record with consentOne recorder in the middle of the table, or the meeting recording for an online group.
- 2Upload the sessionAudio or video up to 600MB. Free up to 1 hour; 3 hours with an hour pack.
- 3Map the speakersVoices come back as Speaker A, B, C. Match them to participant codes using your seating plan.
- 4Check and exportRead the quotes you will use against the recording, then download Word, .txt, .srt or .vtt.
Questions
Can it tell eight people apart?
Is it verbatim?
How long can the recording be?
Can I name the speakers?
How do I handle consent and anonymity?
What about an online focus group on Zoom or Teams?
For other recordings
- Voice memo to text: iPhone Voice Memos and Android recorder files.
- Transcribe a meeting recording: Zoom, Teams and Google Meet recordings.
- WhatsApp voice note to text: A saved voice note, .opus or .ogg.
The same tool, for other files
Audio to text · MP3 to text · Video to text · Audio to SRT · Transcribe a podcast · Your transcripts
More Free Tools
Shorts by Slice
From the team behind these free tools
One video. A week of posts.
Upload an episode and get back the moments worth posting, cut and captioned.
- Reads the whole episode, then picks the excerpts worth posting
- Captions burned in and framed vertical, ready to post
- Your first video (we search the first hour) and first 2 clips are free, no card