Transcribe interviews, podcasts, lectures, voice notes and other recordings into timestamped text, review them against the original audio, then export or build useful transcript-grounded outputs.
110 free minutes Private upload TXT · SRT · VTT
Try the real product
Start with your own file here.
This is the same upload, transcription, and Studio flow used by the product—not a separate demo.
Upload filesPrivate direct upload
01
A transcription workspace for interviews, podcasts and voice recordings
Audio transcription is most useful when the result stays connected to the recording. 0ZARO Scribe converts supported audio into timestamped transcript segments and keeps the player beside the text, so you can listen, seek and correct without switching applications.
The workflow fits interviews, research recordings, podcasts, meetings, lectures, voice notes and client calls that need a usable written record rather than a one-time text dump.
Accepted audio formats include MP3, M4A, WAV, WebM audio, OGG and FLAC.
Non-MP3 sources are normalized locally for the transcription path.
Timestamps stay attached to transcript segments.
Search jumps you from a phrase back to the relevant part of the recording.
02
Keep corrections, highlights and speaker naming inside the transcript
A long audio recording usually needs review. Studio lets you edit transcript segments, keep a revision trail, undo changes and mark important moments with persistent bookmarks or highlights.
When the transcription engine provides a real speaker label, you can assign a durable human-readable name to that label. The product does not invent speaker separation where the upstream transcript did not establish it.
Inline segment editing with saved history.
Bookmarks and highlights survive page reloads.
Speaker aliases apply only to existing provider speaker labels.
Playback speed and transcript follow controls support review of long recordings.
03
Create useful outputs without losing transcript provenance
The saved transcript becomes the source for AI actions such as summaries, chapters, key points, quotes, translation and article drafting. Those outputs can be persisted with the transcript instead of disappearing when the page is closed.
Saved outputs remember the transcript update time they were generated from. When transcript text changes later, the interface can show that an older saved artifact needs refresh. This makes the workspace more reliable than copying text into disconnected tools.
Ground AI actions in the saved transcript identity.
Persist successful AI outputs for later review.
Detect stale saved intelligence after transcript edits.
Copy or export the transcript separately from AI-generated material.
04
Control storage and minutes before the recording becomes a hidden cost
The free account starts with 110 transcription minutes. Duration is measured on the server before a job reservation is created, which means the balance is not based on a user-supplied duration. Active reservations and completed usage are tracked separately.
For privacy and storage control, you can keep the original audio, delete the source while retaining the transcript, or delete the entire recording record. There is also an automatic source-deletion option for users who only want the text after a successful transcription.
110-minute starting transcription balance.
Server-authoritative media duration before minute reservation.
Source-only deletion for text-first workflows.
Full recording deletion for permanent cleanup.
Useful questions
Before you upload
Can I transcribe MP3 and M4A files?
Yes. MP3 and M4A are supported, along with WAV, OGG, FLAC and WebM audio. Other supported media can be normalized locally before the private upload.
Can I search inside the transcript?
Yes. Studio filters transcript segments by text and available speaker labels, and each timestamp can seek the media player to that moment.
Do AI summaries replace the original transcript?
No. AI outputs are separate artifacts. The source transcript remains the primary editable record, and saved AI outputs can be refreshed after transcript edits.
Can I delete the audio but keep the text?
Yes. Source deletion removes the stored media while keeping the transcript and exports. Studio then shows an explicit source-deleted state instead of a broken player.
How much can I transcribe for free?
A newly created account receives a 110-minute transcription balance. The settings area shows available, used and actively reserved minutes.