How do you transcribe audio in Premiere Pro?

Premiere Pro creates transcripts with its built-in Speech to Text feature ¹. The fastest route is to turn on automatic transcription, so each clip is transcribed as you import it. Then open the transcript, correct it, and create captions from the corrected text. The menu names and settings below come from a 2024 walkthrough by a working editor. Adobe's own help pages blocked our research tools, so check the names against your version of Premiere Pro.

Turn on automatic transcription before you import

The fastest setup is to tell Premiere to transcribe clips for you. Go to Premiere Pro > Settings > Transcription and enable Automatically Transcribe Clips ². With the option on, transcription starts in the background as soon as you import a clip. The author notes that you may want to turn it off when you import clips without dialogue, or a large number of clips.

On a Windows computer, the Settings menu may sit in a different place. We did not verify the Windows path.

Check the speakers, because Premiere does not label them reliably

The first transcript in the walkthrough lumped every speaker together. The author fixed it by opening the 3-dots menu in the top right corner of the Transcript panel, choosing Re-transcribe clip, and enabling the option to label separate speakers. The result separated the speakers. He then planned to add the names by hand, because Premiere does not reliably identify speakers ³.

Plan for that manual step. A transcript for an interview with two people needs speaker names added, and a panel with several speakers needs more checking.

Proofread the transcript against the audio

Adobe's automatic transcript is a draft. The same 2024 editor called the engine powerful but not very accurate, and estimated that about 80 percent of its output on three of his videos was text he did not have to transcribe himself ⁴. That figure is one person's estimate from three videos, so do not read it as the accuracy of the feature. It does show the scale of the job: a draft saves typing, but the names, numbers and quotations still need checking.

Play the clip at each point where the draft looks wrong. Pay the most attention to names, numbers, dates and any words that the speaker reads from a screen.

Create captions from the corrected transcript

Once the text is right, you can turn it into captions. The walkthrough used the 3-dots menu in the Transcript panel to create captions from the corrected transcript ⁵. The author then chose between two outputs: the video with the text burned in, or a separate SRT file for uploading to YouTube.

A caption file needs timing, and the transcript on its own does not have it. For what each format holds, see the glossary entry for SRT files and the guide to captions and subtitles.

Check the language before you transcribe

Adobe's Speech to Text page says it is powered for English, Spanish, Japanese, Korean, French, German, Chinese and Hindi, and for more languages ¹. The page does not give the full list, and it does not give a version number. Confirm that your clip's language is supported in your version before you rely on the transcript. We did not verify how Premiere behaves when a language is not supported.

It depends on how the footage was recorded

• Clean dialogue. One person speaking close to the microphone is the simplest case, with the least correction.
• Interviews and panels. Turn on separate speaker labels, then check every speaker change by hand.
• Music under speech. Expect more errors where music plays under the voice, and check those passages first.
• Work that needs a verified record. A legal or research transcript needs a person to check every line against the audio, whatever the software produces.

If a transcript does not appear, our sources do not give a fix. Check the automatic transcription setting, the language, and the speaker labels first, then consult Adobe's help for your version.

What to do

1. Open Premiere Pro > Settings > Transcription and enable Automatically Transcribe Clips on a Mac, or find the matching setting on Windows.
2. For a clip that is already in the project, open the Transcript panel for it. Use the 3-dots menu to re-transcribe the clip with separate speaker labels if more than one person speaks.
3. Correct the names, numbers and paragraph breaks against the audio.
4. Use the 3-dots menu to create captions, then export a burned-in video or an SRT file.

If you do not have Premiere Pro, or the feature is missing from your version, see the other routes in the guide to transcribing video. For a free transcript of a recording you already have, the free audio to text tool works in the browser. For DaVinci Resolve, read whether DaVinci Resolve can transcribe audio for free. For a video on YouTube, read how to get a YouTube transcript. For the wider subject, start with transcription.