Can embedded subtitles be extracted from a video?

Yes, if the embedded subtitle stream is text-based, such as SRT, ASS, SSA, or WebVTT. Image-based streams such as PGS or VobSub usually need OCR before they become editable text.

Can burned-in subtitles be extracted as SRT?

Not directly. Burned-in subtitles are part of the video image, so a subtitle extractor cannot save them as SRT text. You need OCR and manual cleanup.

How do I know if subtitles are embedded or burned in?

Try turning captions off in a player, check for subtitle tracks, or use a local extractor. If captions cannot be disabled and no subtitle stream is found, they are probably burned in.

Should I use a subtitle extractor or OCR first?

Use a subtitle extractor first when the video may contain a text track. Use OCR only when the words are permanently visible in the video picture or the extracted stream is image-based.

Embedded vs Burned-In Subtitles

Subtitle extraction depends on whether the captions are a real text track or part of the video image. Understanding this difference is crucial before attempting to extract subtitles from a video file.

Quick answer

Embedded subtitles are separate text streams stored inside the video container. They can be extracted as .srt, .vtt, or .ass files.

Burned-in subtitles are pixels permanently rendered into the video frames. They cannot be extracted as text without OCR (Optical Character Recognition).

Use Extract Subtitles from Video when the video file contains an embedded text subtitle stream.

Should you extract or use OCR?

Use this quick check before uploading a private video to another service:

What you see	Likely subtitle type	Best next step
Captions can be turned on and off in the player	Embedded or external subtitle track	Try Extract Subtitles from Video and save the text track locally.
FFmpeg or the extractor lists SRT, ASS, SSA, or WebVTT	Embedded text subtitles	Extract the track, then validate or convert it if needed.
FFmpeg lists PGS, VobSub, DVD subtitle, or another image stream	Embedded image subtitles	Extracting may find a stream, but OCR is needed for editable SRT text.
Captions are always visible and cannot be disabled	Burned-in subtitles	Use OCR; a normal extractor cannot read video pixels as subtitle text.
A website shows captions but the downloaded video has no track	Separate streaming captions	Look for the original VTT, SRT, or platform caption download instead of extracting from the video.

If you get an extractable text file, follow up with SRT Validator or Subtitle Cleaner before uploading it to YouTube, Plex, or a web player.

Key differences at a glance

Feature	Embedded Subtitles	Burned-in Subtitles
Can be turned off	✅ Yes	❌ No
Can be extracted	✅ Yes (as text)	❌ No (requires OCR)
File size impact	Minimal (~100KB per language)	None (already in video)
Quality	Perfect text	Depends on video quality
Editability	Easy to edit after extraction	Must re-encode video to change
Multiple languages	Can have many tracks	Only one language visible
Accessibility	Screen readers can access	Screen readers cannot access

Embedded subtitles (extractable)

Embedded subtitles are separate text streams stored inside a video container file. They exist as data alongside the video and audio streams.

How embedded subtitles work

Video container structure:

MKV file (example)
├── Video stream (H.264)
├── Audio stream (AAC)
├── Subtitle stream 1: English (SubRip/SRT)
├── Subtitle stream 2: Spanish (SubRip/SRT)
└── Subtitle stream 3: French (ASS)

Each subtitle stream is a separate data track that can be:

Turned on/off by the player
Extracted as a standalone file
Replaced or removed without re-encoding the video

Common embedded subtitle formats

Text-based (easily extractable):

SubRip (SRT) - Most common, simple text format
ASS/SSA - Advanced styling, used in anime
WebVTT - Web standard, used in HTML5 video
SubStation Alpha - Older format with styling

Image-based (harder to extract):

PGS (Presentation Graphic Stream) - Used in Blu-ray
VobSub - Used in DVD
DVB subtitles - Used in broadcast TV

Note: Image-based subtitles are embedded but not plain text. Extracting them gives you images, not editable text files.

Containers that support embedded subtitles

MKV (Matroska) - Best support, can hold unlimited subtitle tracks
MP4 - Limited support, mainly for streaming
AVI - Limited support, older format
MOV - Apple format, supports subtitles
WebM - Web format, supports WebVTT

Advantages of embedded subtitles

✅ User control - Viewers can turn captions on/off
✅ Multiple languages - One video file can have many subtitle tracks
✅ Small file size - Text subtitles add minimal size (~100KB per language)
✅ Easy to edit - Extract, edit, and re-embed without touching the video
✅ Accessibility - Screen readers can access the text
✅ Quality - Perfect text quality regardless of video resolution

Disadvantages of embedded subtitles

❌ Player support required - Not all players support all subtitle formats
❌ Extraction needed - Must extract to edit or convert
❌ Container limitations - Some containers (MP4) have limited subtitle support

Burned-in subtitles (hardcoded, not extractable)

Burned-in subtitles (also called “hardcoded” or “hardsubbed”) are permanently rendered into the video frames as pixels. They are part of the video image itself, not a separate data stream.

How burned-in subtitles work

Video structure:

MP4 file (example)
├── Video stream (with subtitles rendered into frames)
└── Audio stream

The subtitle text is literally painted onto each video frame during encoding. A player cannot turn them off because they are part of every frame, just like any other visual element in the video.

Visual comparison

Embedded subtitles:

Video frame: [person talking]
Subtitle layer: "Hello, how are you?"
↓
Player can show/hide the subtitle layer

Burned-in subtitles:

Video frame: [person talking + "Hello, how are you?" text]
↓
Text is part of the image, cannot be removed

When burned-in subtitles are used

Common scenarios:

Social media videos - Instagram, TikTok, Facebook videos often have hardcoded captions
Fansubs - Anime fan translations are often burned in
Broadcast TV recordings - News, sports with permanent captions
Accessibility compliance - Some platforms require always-visible captions
Pirated content - Watermarked subtitles to identify the source
Legacy formats - Old VHS or DVD rips

Advantages of burned-in subtitles

✅ Universal compatibility - Works on any player, no subtitle support needed
✅ Always visible - Guaranteed to display, no user action required
✅ No extraction needed - Subtitles are already “there”
✅ Styling preserved - Custom fonts, colors, and effects are permanent

Disadvantages of burned-in subtitles

❌ Cannot be turned off - Always visible, even if unwanted
❌ Cannot be extracted as text - Requires OCR to get editable text
❌ Single language only - Can’t have multiple language options
❌ Larger file size - Increases video bitrate and file size
❌ Quality depends on video - Low-resolution video = blurry subtitles
❌ Not accessible - Screen readers cannot read burned-in text
❌ Cannot be edited - Must re-encode the entire video to change subtitles
❌ Blocks video content - Can cover important visual elements

Why burned-in subtitles increase file size

When subtitles are burned in, the video encoder must work harder to compress frames with text. Text creates sharp edges and high-contrast areas that are harder to compress efficiently.

Example:

Video without subtitles: 500 MB
Same video with burned-in subtitles: 550-600 MB (10-20% larger)

Embedded subtitles add only ~100KB per language, regardless of video length.

How to tell the difference

Before attempting extraction, determine which type of subtitles you have.

Method 1: Try to turn captions off in the player

Steps:

Play the video in VLC, MPC-HC, or another media player
Look for a subtitle menu or CC button
Try to disable subtitles

Results:

✅ Captions disappear → Embedded subtitles
❌ Captions stay visible → Burned-in subtitles

Method 2: Check with FFmpeg (most reliable)

Command:

ffmpeg -i video.mkv

Look for subtitle streams in the output:

Embedded subtitles example:

Stream #0:0: Video: h264
Stream #0:1: Audio: aac
Stream #0:2(eng): Subtitle: subrip
Stream #0:3(spa): Subtitle: ass

Burned-in subtitles example:

Stream #0:0: Video: h264
Stream #0:1: Audio: aac
(no subtitle streams listed)

If FFmpeg lists subtitle streams, they are embedded and can be extracted.
If FFmpeg finds no subtitle streams, the captions are either burned-in or the video has no subtitles.

Method 3: Check file properties

In VLC:

Open the video
Go to Tools → Media Information
Click the “Codec Details” tab
Look for subtitle tracks

In Windows:

Right-click the video file
Properties → Details tab
Look for “Subtitle” field

In macOS:

Select the video file
Press Cmd+I (Get Info)
Look for subtitle track information

Method 4: Visual inspection

Embedded subtitles typically:

Have clean, sharp text
Use standard fonts
Appear in a consistent position
Have a semi-transparent background

Burned-in subtitles often:

Match the video quality (blurry if video is low-res)
Use custom or stylized fonts
May have creative positioning or effects
Are part of the video compression artifacts

Method 5: Check the file extension

Likely to have embedded subtitles:

.mkv (Matroska) - Best subtitle support
.mp4 (with subtitle tracks)
.mov (QuickTime)

Less likely to have embedded subtitles:

.avi (older format, limited support)
.wmv (Windows Media)
.flv (Flash video)

Note: File extension alone doesn’t guarantee embedded subtitles. Always check with FFmpeg or a player.

Common mistakes

Assuming every MKV has subtitles

Problem: MKV is a container format that supports subtitles, but not every MKV file contains subtitles.

Reality:

MKV can hold 0 to unlimited subtitle tracks
Just because a file is .mkv doesn’t mean it has subtitles
Always check with FFmpeg or a player before attempting extraction

Example:

video1.mkv → Has 3 subtitle tracks (English, Spanish, French)
video2.mkv → Has 0 subtitle tracks (video and audio only)

Expecting OCR from a subtitle extractor

Problem: Trying to use Extract Subtitles from Video on burned-in subtitles.

Why this fails: Extraction tools read subtitle data streams, not video pixels. They cannot “see” burned-in text.

Solution for burned-in subtitles: Use OCR (Optical Character Recognition) software:

Subtitle Edit (free, Windows) - Has built-in OCR
SubtitleExtractor (paid) - Automated OCR
Manual transcription - Most accurate but time-consuming

OCR limitations:

Requires manual correction (70-90% accuracy)
Slow process (can take hours for a movie)
Struggles with stylized fonts or low-quality video
Cannot handle overlapping text or complex layouts

Ignoring image-based subtitle streams

Problem: Extracting PGS or VobSub subtitles and expecting editable text.

What happens: You get image files (.sup, .idx/.sub), not text files (.srt).

Why this matters: Image-based subtitles are embedded but not plain text. They’re essentially a series of images timed to appear at specific moments.

Common image-based formats:

PGS - Blu-ray subtitles (.sup files)
VobSub - DVD subtitles (.idx + .sub files)
DVB - Broadcast TV subtitles

Solution: Use OCR to convert image-based subtitles to text:

Extract the image-based subtitle stream
Use Subtitle Edit or similar tool to OCR the images
Export as SRT or VTT

Or: Look for a text-based subtitle track in the same video file. Many Blu-rays include both PGS (for players) and SRT (for accessibility).

Confusing “closed captions” with embedded subtitles

Closed captions can be:

Embedded (e.g., CEA-608/708 in broadcast video)
Burned-in (e.g., open captions on social media)
Separate files (e.g., SRT uploaded to YouTube)

The term “closed captions” refers to the content (includes sound effects, speaker IDs), not the technical implementation.

Trying to extract from streaming video

Problem: Attempting to extract subtitles from Netflix, YouTube, or other streaming services.

Why this fails: Streaming subtitles are delivered separately from the video, not embedded in the video file you’re watching.

Solution:

YouTube: Download subtitles directly from the video page (if available)
Netflix, etc.: Use browser extensions or official download features (where legal)
Live streams: Captions are often not saved and cannot be extracted after the stream ends

When to use each type

Use embedded subtitles when:

✅ You want viewers to control caption visibility
✅ You need multiple language options
✅ You want to keep file size small
✅ You need accessibility compliance (screen readers)
✅ You might need to edit subtitles later
✅ You’re distributing via Plex, Kodi, or similar media servers

Use burned-in subtitles when:

✅ You need guaranteed caption visibility (social media)
✅ Target platform doesn’t support subtitle files
✅ You want custom styling that players might not support
✅ You’re creating content for platforms that strip subtitle tracks
✅ You want to prevent subtitle removal or editing

Frequently asked questions

Can I convert burned-in subtitles to embedded subtitles?

Not directly. You would need to:

Use OCR to extract the text from video frames
Manually correct OCR errors
Time the subtitles to match the video
Embed the resulting SRT file into the video

This is time-consuming and often imperfect.

Can I remove burned-in subtitles?

No, not without re-encoding the video and losing quality. Burned-in subtitles are permanent. You would need the original video without subtitles to start over.

How do I extract embedded subtitles?

Use the Extract Subtitles from Video tool or FFmpeg:

ffmpeg -i video.mkv -map 0:s:0 subtitles.srt

Why do some videos have both embedded and burned-in subtitles?

This happens when:

Someone burned in subtitles to a video that already had embedded subtitles
A video was re-encoded with hardcoded captions added
Fansubs added their own burned-in subtitles to official releases

Result: You see two sets of subtitles (one you can turn off, one you can’t).

Are YouTube subtitles embedded or burned-in?

YouTube subtitles are neither. They’re delivered as separate WebVTT files that the player overlays on the video. You can download them directly from YouTube without extracting from the video file.

Can I add embedded subtitles to a video?

Yes, using FFmpeg or video editing software:

ffmpeg -i video.mp4 -i subtitles.srt -c copy -c:s mov_text output.mp4

This creates a new video file with the subtitle track embedded.