How to add automatic subtitles in DaVinci Resolve

Typing subtitles by hand is one of the slowest jobs in editing. The good news is that DaVinci Resolve can listen to your audio and write them for you. The less good news is that there is a condition worth knowing before you start.

First things first: you need Studio

The feature that transcribes audio is called Create Subtitles from Audio, and it is exclusive to DaVinci Resolve Studio, the paid version. On the free version that option simply does not appear in the menu.

This is not a problem with your install or something you forgot to activate: it runs on Blackmagic’s AI engine, and that engine ships only with Studio.

If you are on the free version you can still work with subtitles: you can create them by hand or import an .srt made elsewhere. What you cannot do is generate them automatically from the audio.

Step 1 — Prepare the timeline

Before generating anything, make sure your video is already cut on the Edit page with audible audio. Resolve transcribes what it hears, so the cleaner the audio, the fewer errors you will fix later.

If your project has loud background music, drop its volume temporarily or mute it. The transcription improves a lot.

Step 2 — Generate the subtitles

With the timeline open, go to the top menu:

Timeline → Create Subtitles from Audio

A window opens with several options. The ones that matter:

• Language: the language spoken in the video. Choose carefully — get it wrong and the transcription comes out unreadable.
• Caption preset: how many characters per line. For vertical social video you want less text per line.
• Split captions: how it breaks up sentences.

Hit Create and wait. Resolve analyses the audio and creates a new subtitle track above your video. On a one-minute clip it usually takes seconds.

Step 3 — Fix the mistakes

No automatic transcription is perfect. Proper nouns, jargon and foreign words are the usual suspects.

Click any subtitle and correct the text in the Inspector, on the caption tab.

Here is a detail that catches a lot of people out: if you plan to use word-by-word animation, correcting the text alone is not enough. The animation does not read the subtitle text — it reads the per-word timings Resolve calculated during transcription. Fix only the text and the animation may still show the old word.

The reliable way to correct without losing the timings is to do it from the original transcription:

1. In the Media Pool, right-click the clip → Audio Transcription → Transcribe
2. Fix the words there, in the transcription window
3. Go back to the timeline and regenerate the subtitles

It is more awkward, but it preserves the timing data the animations need.

Step 4 — Style them

By default Resolve subtitles are white text on a black box. Functional, but not what you see in videos that hold attention.

To change the look of the whole track at once, select the subtitle track and open the Inspector, Track Style tab. There you can change:

• Typeface, size and colour
• Outline and shadow
• Background or box behind the text
• Position on screen

Everything you change here applies to every subtitle on the track at the same time, which is exactly what you want.

Step 5 — Word-by-word highlighting

From DaVinci Resolve 20 onwards, the word being spoken can be highlighted as it is said. It is the effect CapCut popularised, and the one that keeps people reading.

Resolve ships with a few animated styles. You apply them by dragging from Effects → Titles → Subtitles onto the subtitle track header (the grey area on the left that says “Subtitle 1”), not onto an individual subtitle.

If the words are out of sync, select a subtitle and use Analyze Word Timing so Resolve re-adjusts the rhythm word by word.

Common problems

“I cannot find Create Subtitles from Audio.” You are on the free version, or you do not have a timeline with audio open.

“I changed the style and the track did not update.” When you apply a style, the track keeps its own copy. To change it, drag the new style onto the track header again.

“The animation shows a word I already corrected.” That is what step 3 covers: the animation uses transcription timings, not the subtitle text. Correct from the original transcription.

“Short words dangle at the end of lines.” Adjust characters per line in the track settings, or split the subtitle manually.

A shortcut if you are short on time

Setting up each style from scratch takes a while: choosing a typeface, adjusting the outline, calibrating the highlight, checking it reads on a phone. And you repeat it on every project.