Can I edit a transcript after it is finished?

Yes. Every segment opens for editing and carries 4 fields: speaker label, start, end and text. Edits are saved to the transcript, so every format you download afterwards reflects them.

Do I have to export again after editing?

Yes — a file you already downloaded is a snapshot and does not update. Make your corrections first, then export. That order is worth forming a habit around, because re-exporting costs nothing and re-editing an exported file costs everything you already fixed.

Which format should I use for video subtitles?

SubRip (.srt) is the safe default — almost every video editor and platform accepts it. Use WebVTT (.vtt) when the captions are going onto a web page you control, because that is what an HTML video player expects.

What is in the JSON export?

The structured version: segments, times, speakers, and per-word timings when word timestamps were on. This is the one to take if code is going to read it. It is the export to take when another program is going to read the transcript rather than a person — a search index, a summarizer, an editing pipeline.

Can I rename a speaker everywhere at once?

No. Speaker labels are edited on the segment you have open, so renaming across a long recording means working through the segments that carry that label. This is the practical reason to declare the speaker count before you run the job rather than to repair labels afterwards.

Can I get the transcript as a Word document?

Yes — Word (.docx) is in the download menu, alongside PDF (.pdf) and HTML (.html). Pick Word when someone is going to edit or comment on it, and PDF when they are not.

Transcription

Editing and exporting a transcript

Fix the transcript before you export it, then pick the right file: subtitles for video, a document for people, JSON for anything a machine has to read.

Lesson 3 · core · 6 min read

Open Transcription

A transcript is finished when the words are right and it is in the file the next step needs. Those are two separate jobs and they belong in that order — exporting first and correcting afterwards means doing the export twice.

Fix it before you export it

Every segment opens for editing, and each one carries the same 4 fields. Edits are saved against the transcript itself, so a name you correct once is correct in every format you download afterwards.

Field: What it is for
Speaker label: Who this segment belongs to. Retype it to reassign a crossed segment, or to replace the generated label with a person's actual name.
Start: When this segment begins, in seconds. Worth nudging when a caption comes in slightly ahead of the line being spoken.
End: When it ends. The other half of a caption that sits a beat off the audio.
Text: The words themselves. Names, jargon and numbers are where a transcript is wrong, and they are usually wrong the same way every time.

What an open segment lets you change.

Start with the repeated errors
Transcripts are wrong systematically, not randomly. A surname, a product name, a piece of jargon — each is usually wrong the same way every time it appears, so a handful of fixes clears most of the errors in the file.
Then the speakers
Reassign the segments that crossed, and put real names where the generated labels are. This is per segment, so it is worth doing on a transcript you are going to publish and not worth doing on one you are going to skim.
Timings last, and only if they matter
Nudge a start or an end when a caption sits off the speech. In a document the timings are invisible, so this step is skippable for everything except subtitles.

Careful

A file you have already downloaded does not change when you edit the transcript. Correct first, export second — and if you have already sent the wrong version out, re-export rather than patching the file by hand.

Which format goes where

The download menu offers 7 formats, and the choice is entirely about the destination rather than about quality — each one is generated from the same saved transcript.

Format: Where it belongs
Text (.txt): Just the words. The right choice when the transcript is going into something else — a draft, an email, a prompt — and nothing about timing or layout matters.
JSON (.json): The structured version: segments, times, speakers, and per-word timings when word timestamps were on. This is the one to take if code is going to read it.
SubRip (.srt): Subtitles. The format almost every video editor and video platform accepts, and the default answer for captions.
WebVTT (.vtt): Subtitles for the web — what an HTML video player expects. Take this one when the captions are going onto a page you control rather than into an editor.
Word (.docx): A document someone is going to read and mark up. The right export for an interview a human has to work through, not a machine.
PDF (.pdf): A document nobody is going to edit — a record to file, send or attach. Fixed layout, so fix the transcript before you export it.
HTML (.html): A web page you can open in a browser or paste into a CMS with its formatting intact.

Every export format, and where it belongs.

Going into a video: SubRip (.srt) and WebVTT (.vtt). SubRip is the safe default; WebVTT is for a web player.
Going to a person: Word (.docx), PDF (.pdf) and HTML (.html). Word if they will edit it, PDF if they will not.
Going into software: JSON (.json) — the only export that carries the structure, and the only one worth parsing.
Going into a draft: Text (.txt). Nothing but the words, which is the point.

Tip

Exporting is not a commitment. Download SubRip for the edit and Word for the write-up from the same transcript — the file is generated fresh each time from what is saved.

What this does not do

Worth knowing before you plan a workflow around it. AudioTranscribe turns speech into text and labels who was speaking; it does not translate, it does not burn subtitles into a video, and it does not rename a speaker across a whole transcript in one action.

Renaming a speaker is per segment. Declaring the speaker count before the job is the cheaper path — see the previous lesson.
A caption file is a file, not a video. Your editor or player applies it; nothing here writes it onto the picture.
The Link tab is YouTube-only, so anything hosted elsewhere has to be uploaded.

That is the track. If a transcript came out worse than you expected, the fix is almost never in this lesson — go back to accuracy and the speaker count, because those two decide what you are editing in the first place.

Questions

What people ask about this

Free to start

Now go make one

Reading about a style description only gets you so far. The studio is free to use — write one sentence and hear what comes back.

Create a free account

1,000 credits every month. No card required.

All transcription lessons

Transcription

Editing and exporting a transcript

Fix the transcript before you export it, then pick the right file: subtitles for video, a document for people, JSON for anything a machine has to read.

Lesson 3 · core · 6 min read

Open Transcription

Fix it before you export it

Every segment opens for editing, and each one carries the same 4 fields. Edits are saved against the transcript itself, so a name you correct once is correct in every format you download afterwards.

Field: What it is for
Speaker label: Who this segment belongs to. Retype it to reassign a crossed segment, or to replace the generated label with a person's actual name.
Start: When this segment begins, in seconds. Worth nudging when a caption comes in slightly ahead of the line being spoken.
End: When it ends. The other half of a caption that sits a beat off the audio.
Text: The words themselves. Names, jargon and numbers are where a transcript is wrong, and they are usually wrong the same way every time.

What an open segment lets you change.

Start with the repeated errors
Transcripts are wrong systematically, not randomly. A surname, a product name, a piece of jargon — each is usually wrong the same way every time it appears, so a handful of fixes clears most of the errors in the file.
Then the speakers
Reassign the segments that crossed, and put real names where the generated labels are. This is per segment, so it is worth doing on a transcript you are going to publish and not worth doing on one you are going to skim.
Timings last, and only if they matter
Nudge a start or an end when a caption sits off the speech. In a document the timings are invisible, so this step is skippable for everything except subtitles.

Careful

Which format goes where

The download menu offers 7 formats, and the choice is entirely about the destination rather than about quality — each one is generated from the same saved transcript.

Format: Where it belongs
Text (.txt): Just the words. The right choice when the transcript is going into something else — a draft, an email, a prompt — and nothing about timing or layout matters.
JSON (.json): The structured version: segments, times, speakers, and per-word timings when word timestamps were on. This is the one to take if code is going to read it.
SubRip (.srt): Subtitles. The format almost every video editor and video platform accepts, and the default answer for captions.
WebVTT (.vtt): Subtitles for the web — what an HTML video player expects. Take this one when the captions are going onto a page you control rather than into an editor.
Word (.docx): A document someone is going to read and mark up. The right export for an interview a human has to work through, not a machine.
PDF (.pdf): A document nobody is going to edit — a record to file, send or attach. Fixed layout, so fix the transcript before you export it.
HTML (.html): A web page you can open in a browser or paste into a CMS with its formatting intact.

Every export format, and where it belongs.

Going into a video: SubRip (.srt) and WebVTT (.vtt). SubRip is the safe default; WebVTT is for a web player.
Going to a person: Word (.docx), PDF (.pdf) and HTML (.html). Word if they will edit it, PDF if they will not.
Going into software: JSON (.json) — the only export that carries the structure, and the only one worth parsing.
Going into a draft: Text (.txt). Nothing but the words, which is the point.

Tip

Exporting is not a commitment. Download SubRip for the edit and Word for the write-up from the same transcript — the file is generated fresh each time from what is saved.

What this does not do

Renaming a speaker is per segment. Declaring the speaker count before the job is the cheaper path — see the previous lesson.
A caption file is a file, not a video. Your editor or player applies it; nothing here writes it onto the picture.
The Link tab is YouTube-only, so anything hosted elsewhere has to be uploaded.

Questions

What people ask about this

Free to start

Now go make one

Reading about a style description only gets you so far. The studio is free to use — write one sentence and hear what comes back.

Create a free account

1,000 credits every month. No card required.

All transcription lessons

Editing and exporting a transcript

Fix it before you export it

Start with the repeated errors

Then the speakers

Timings last, and only if they matter

Which format goes where

What this does not do

What people ask about this

Can I edit a transcript after it is finished?

Do I have to export again after editing?

Which format should I use for video subtitles?

What is in the JSON export?

Can I rename a speaker everywhere at once?

Can I get the transcript as a Word document?

Now go make one

Editing and exporting a transcript

Fix it before you export it

Start with the repeated errors

Then the speakers

Timings last, and only if they matter

Which format goes where

What this does not do

What people ask about this

Can I edit a transcript after it is finished?

Do I have to export again after editing?

Which format should I use for video subtitles?

What is in the JSON export?

Can I rename a speaker everywhere at once?

Can I get the transcript as a Word document?

Now go make one