Transcribe a podcast episode to text
Transcribe a podcast episode to text: upload the audio of the episode and get timecodes, speaker labels and six export formats.
Upload the audio file of the episode and the transcript comes back with a timecode on every line, the speakers named on request and six export formats. One file is up to 30 minutes, and the free plan is $0 with 3 files a day.
How to transcribe a podcast episode
- 1
Take the audio of the episode
The episode file is the audio of the show. Download it from the feed or from the platform that hosts it, and keep the original file: there is no need to convert it first.
- 2
Upload it and set the language
Drop it on the widget above. The file may be up to 30 minutes at most. Choose 99+ languages, or leave the detection on, and split the speakers if the show has more than one voice.
- 3
Cut the parts you need
Every line has the time it starts and ends, so the quotes for a show note can be found in the transcript instead of in the audio. Export to TXT, Markdown, SRT, VTT, CSV or DOCX.
What you get
- TXT
Plain text for an email, a note or a chat
- SRT
Subtitles for video editors and platforms
- VTT
Subtitles for a browser and a website
- CSV
A table: time, speaker, text
- DOCX
A Word document with speakers and times
- Markdown
Text with headings for notes
What a podcast transcript contains
- How an episode carries its audio
the feed points at the audio file with an enclosure element: a URL, the size in bytes and the MIME type
- Timecodes and speakers
the start and the end time of every segment, speakers named on request
- Export
TXT, Markdown, SRT, VTT, CSV and DOCX (DOCX is a Pro feature)
What an episode file is, and why the feed matters
An episode is a file, and the feed is a list of files. The feed of a show points at the audio of every episode with an enclosure element: an address, the size in bytes and the type of the file. That is where the file to transcribe comes from — the same way a listener gets it.
Tips
Keep the original episode file; the transcript is derived from it.
Split by speaker when the show has two microphones.
Name the file after the episode number, not after the title.
Questions this page answers
Does the feed have to be changed for this?
No. Nothing is written back into the feed and nothing is published for you. You upload the audio file yourself, and the transcript stays in your account until you export it.
How long can one episode be?
One file is up to 30 minutes, on every plan. A longer episode has to be split into parts before the upload; the service does not cut files for you.
Can a transcript of a show be searched?
Yes. The transcript is text with one line per segment, so it can be searched in the app and read in the editor. The CSV export gives the same lines with the time and the speaker for a spreadsheet.
What about the accuracy of the transcription?
We publish no single accuracy number, because a number without the test set behind it means nothing. Ask for the evaluation report and the sample set and we will send both.
Sources
- RRSS 2.0 Specification — the enclosure elementrssboard.org · checked 22 October 2026
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.