up to 30 minutes on the free plan, no minute limit on Pro — a long interview fits, if the file does
For journalists
Transcribe an interview for a journalist
Transcribe an interview for a journalist: timecodes on every line, speakers named, and a Word file to quote from.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps
In short
Upload the interview and the transcript comes back with a timecode on every line, the speakers named on request and a Word file on Pro. One file is up to 30 minutes, and the free plan is $0 with 3 files a day.
The case
Why a quote needs a timecode
An interview is a record of speech, and a quote from it has to survive being read a day later. The timecode is what saves the quote: it points at the place in the recording where the words were said, so the sentence can be checked against the audio in a second.
What a journalist gets from the transcript
DOCX with the transcript, a Pro feature
$0: 3 files a day. Pro is $10 a month paid yearly ($120 a year)
How to transcribe an interview
- 1
Upload the recording of the interview
The file from the recorder, the phone or the call. One file is up to 30 minutes, so most interviews fit as they are, without a cut.
- 2
Split the speakers
A journalist and a source are two voices. Switch on the split by speaker and the transcript names them, which is what makes a quote attributable.
- 3
Export the file you have to send
DOCX with the transcript for the editor, SRT for the film, CSV for the archive. Every format keeps the timecodes, so the audio stays reachable.
Tips
Check the first minute: names are the first thing to go wrong.
Keep the original recording; the transcript is derived from it.
Export DOCX when the piece has to go to an editor.
Questions this page answers
How do I know a quote is correct?
Read it against the recording: every line carries the time it starts and ends, so the sentence in the transcript can be checked against the audio in a second.
Can I correct a name in the transcript?
Yes. The editor in the app lets you fix a word and save the change; the export files are made from the corrected text.
Do you publish an accuracy number?
No, and that is on purpose. A score without the test set behind it means nothing, so ask for the evaluation report and the sample set and we will send both.
Where does the audio of the interview go?
A Russian or mixed file is processed by our own engine on infrastructure outside the United States. A pure English file may go to a cloud speech provider in the United States when that fallback is on.
Reading: 6 minutesUpdated Transcribsi editorial team
Drop a recording — the text arrives in a minute
Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.
3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps