How to Extract Hardcoded Subtitles from Video (Free OCR to SRT)

BygImageReader Team
⚡ Quick Answer

To extract hardcoded (burned-in) subtitles from a video for free: use Subtitle Edit (free, open-source) — import the video or subtitle track, run its built-in OCR, review the uncertain words, and export a clean .srt file. Soft subtitles (separate tracks) need no OCR at all — just demux them.

Your video has subtitles burned into the picture — no subtitle track, no toggle, no way to copy the text. Re-typing an hour of dialogue is miserable. The fix is OCR: software that watches the video, reads the subtitle text frame by frame, and writes a timed .srt file. Here’s how to do it free.

First: What Kind of Subtitles Do You Have?

  • Soft subtitles (separate track in MKV/MP4): no OCR needed — extract the track directly with MKVToolNix. Done in seconds.
  • Hardcoded / burned-in subtitles (part of the picture): OCR is the only way. This guide covers exactly that.
  • DVD/Blu-ray image subtitles (VobSub .idx/.sub, PGS .sup — see CCExtractor’s VOBSUB documentation): also pictures — same OCR workflow below.

Free Method: Subtitle Edit + OCR (5 Steps)

1

Install Subtitle Edit

Download the free, open-source Subtitle Edit. It includes a dedicated OCR workspace built for exactly this job.

2

Import the video or subtitle images

Open your video file (for hardcoded subs) or the .idx/.sub / .sup files (for DVD/Blu-ray image subtitles).

3

Run the OCR pass

Select the subtitle language and OCR engine (Tesseract-based). Subtitle Edit scans the subtitle area frame by frame and proposes text for each cue.

4

Review uncertain words

This is the key step: the OCR workspace flags every uncertain word for manual confirmation. Names, italics, and stylized fonts are where errors hide — a 10-minute review pass makes the difference between a usable and a garbage .srt.

5

Export as .srt

Save as SubRip (.srt) — plays everywhere, editable in any text editor, translatable with any tool.

Reality check: Subtitle OCR is CPU-heavy and imperfect. Expect to correct 5–15% of cues, more with stylized fonts or low-resolution video. Anyone promising “one-click perfect” subtitle extraction is selling something.
Same engine, still images: The OCR core here is Tesseract — the same engine inside gImageReader. For extracting text from video stills (a paused frame, a thumbnail), screenshot it and OCR it with gImageReader directly.
Related guides: Translate Text in Images

Frequently Asked Questions

How can I extract hardcoded subtitles from a video for free?
Use the free, open-source Subtitle Edit: import the video, run its built-in Tesseract-based OCR on the subtitle area, review the uncertain words it flags, and export a clean .srt file. Soft subtitle tracks need no OCR — extract them directly with MKVToolNix.
Can OCR really turn burned-in subtitles into an SRT file?
Yes — that’s exactly what subtitle OCR does: it reads the subtitle text from each video frame and writes timed cues. Expect to manually review 5-15% of cues for errors, especially names and stylized fonts.
What’s the difference between soft and hardcoded subtitles?
Soft subtitles are separate text tracks you can toggle and extract losslessly. Hardcoded (burned-in) subtitles are part of the video picture itself — the only way to get them as text is OCR, frame by frame.

Same OCR engine for your documents

gImageReader brings Tesseract OCR to your scans and screenshots — free.

Download gImageReader Free

Similar Posts