Transcribe audio
Transcribe audio
TryScriptorium transcribes audio into editable, timestamped text with optional speaker labels. Upload the recording, choose the language, and review the result. It suits interviews, podcasts, lectures and meetings. One file up to 15 minutes and 25 MB is free with no account. Exports include TXT, SRT, VTT, Markdown, JSON and CSV.
Different recordings need different things from a transcript. An interview needs to be quotable, a podcast needs show notes, a lecture needs to be revisable, a meeting needs decisions. Here is how each one actually goes.
Key facts
- Speaker labels (up to 8) can be renamed.
- A built-in editor and full-text search work across your transcripts.
- Languages: 33 plus automatic detection, including Arabic.
- Export: TXT, TXT with timestamps, SRT, VTT, Markdown, JSON (word timings), CSV.
- Free: one file up to 15 minutes and 25 MB with no account, 3 tries a day.
- Privacy: no-account files are deleted after 24 hours; uploads are not used to train models.
Last updated:
Interviews
What you want from an interview transcript is an accurate quote and the timestamp that proves it. Upload the recording, rename the speaker labels to the real names once, then work through it with the audio playing — clicking a line jumps straight to that moment, which is what makes checking a quote quick rather than a search through an hour of tape.
Export timestamped text when you need to cite, or plain text when you are writing up. If the interview is on the record, agree that before you record it: consent rules for recording vary by jurisdiction and are your responsibility, not the software's.
Podcasts
An episode transcript does three jobs: show notes, a searchable archive, and text a search engine can actually read, since audio on its own is invisible to one. Upload the finished episode audio, label the host and guests, then export Markdown for show notes or plain text for the episode page.
If you also publish the episode as a video, export WebVTT so the same transcript becomes captions rather than being written twice.
Lectures
A lecture transcript is for finding things again. The useful move is to transcribe as you go through a term rather than the night before an exam, because searching across everything you have transcribed is what turns a pile of recordings into something you can revise from.
Lecture audio is often the worst case for a microphone: a phone on a desk, a room with hard walls, and a lecturer who moves. Closer is better, always.
Meetings
A recorded call gives you a written record of who said what and what was decided. Upload the recording, label the speakers, and export Markdown for a formatted record.
Everyone on the call should know it is being recorded. That is a legal requirement in many places and a matter of basic courtesy everywhere.
How it works
Add a recording
Drop in a video or audio file, choose one from your device, or record from the microphone. Nothing to install.
It transcribes
Speech becomes timestamped text, with speaker labels when more than one person talks.
Read, fix, export
Click any line to correct it, then download TXT, subtitles, Markdown, JSON or CSV.
Questions
Can it label who is speaking?
How long can a recording be?
Is it suitable for research interviews?
Can I search across all my transcripts?
Which audio quality works best?
Can I get timestamps for every sentence?
How do I transcribe a very long recording?
Should I clean up the audio first?
Where do I see the measured speed?
Want your transcripts saved and longer files? Create a free account · See pricing · All use cases
Try it with your file
No account needed for a first transcript.