Text Recognition (OCR)
Extract existing burned-in subtitles from video frames
What Is Text Recognition (OCR)
The Text Recognition (OCR) feature extracts existing burned-in subtitles from video frames and generates an editable SRT subtitle file. It is ideal for videos that already have visible subtitles but no separate subtitle file, such as downloaded videos with hardcoded subtitles, TV show recordings, etc.
Unlike speech recognition, OCR analyzes the video image rather than the audio track, so it can recognize any text that has been "burned in" to the video frames.
How to Use
- Import videos into the media libraryDrag video files into the GeekLink media library, or click the "Add Videos" button to select files.
- Switch to the "Text Recognition" tabSelect the "Text Recognition" tab at the top of the main interface.
- Choose the source languageSelect the OCR recognition language for the video subtitles.
- Click "Run Text Recognition"After confirming your settings, click the button to start processing.
- Color sampling (optional)If "Extract All Text" is not checked, you will enter the color picker step -- paint over the subtitle text on a video frame to sample its color (see details below).
- Confirm colorAfter painting in the color picker, click "Confirm Color" to continue.
- Wait for OCR to finishThe system automatically scans video frames to extract text content.
- View resultsDone! Open the subtitle editor to view and edit the recognition results.
Color Picker
Video frames often contain a lot of text besides subtitles -- watermarks, logos, on-screen labels, etc. The purpose of color sampling is to help the OCR engine focus on the target subtitles you want to extract by distinguishing them through their text color.
How It Works
- Use the mouse to paint over the subtitle text area on the video frame; the system automatically collects the color at the painted positions
- Use the +/- buttons to zoom in/out for precise painting on small subtitles
- Made a mistake? Click "Clear Paint" to start over
- Once the minimum sample size is reached, the "Confirm Color" button becomes clickable
- Use "Font Height Filter" to set minimum and maximum pixel values, filtering out non-subtitle text like watermarks (leave blank for no limit)
OCR Settings Explained
| Setting | Description | Recommendation |
|---|---|---|
| Source Language | The language for OCR recognition | Select the language of the video subtitles |
| Subtitle Region | Limit the scan area: Bottom / Bottom 20% / Top Half / Full Screen | Subtitles are usually at the bottom; choosing "Bottom" or "Bottom 20%" reduces false positives and speeds up processing |
| Detection Interval | Frame sampling rate: 0.25s / 0.3s / 0.5s / 1.0s | Default 0.5s is sufficient; lower to 0.25s for videos with rapidly changing subtitles |
| Filter Text | Exclude characters of specific scripts | Use when the frame contains unwanted Japanese/Korean/Thai text, etc. |
| Extract All Text | Skip color sampling and extract all text on screen | Use when subtitle color is inconsistent or multi-colored |
| Use Previous Style Preset | Reuse the color sampling from the last run | Saves repeated sampling when batch-processing videos in the same series |
| AI Enhancement PRO | Punctuation correction + visual re-check to improve OCR quality | Slows down processing; enable as needed |
FAQ
Why is the output full of garbled text?
The color sampling may not be precise enough, causing the OCR engine to treat background textures as text. Try re-sampling and paint only on the subtitle text strokes. You can also switch to a frame with clearer subtitles and try again.
What languages are supported?
OCR currently supports the following languages: Simplified Chinese, Traditional Chinese, Chinese-English bilingual, English, Japanese, Korean, and Vietnamese.
Why are some subtitles not detected?
The detection interval may be too large, causing subtitles that flash by quickly to be missed. Try lowering the detection interval to 0.25s to capture faster-changing subtitles. The trade-off is a longer processing time.