Practical overview
When it fits
Use it when captions are embedded in the video image and you want separate subtitle files without routing OCR through a third-party API.
Who it is for
- People creating separate subtitle files from captions burned into video.
- Users processing multiple videos that share resolution and subtitle placement.
- Users who need local OCR and multilingual subtitle recognition.
Documented uses
- Extract searchable SRT subtitle sidecars from videos.
- Batch-process consistently formatted videos into subtitle files.
- Export recognized subtitle text as TXT.
- Replace or remove recurring recognized strings through a text-mapping configuration.
Requirements & limits
The project is open source under the Apache-2.0 license and provides a desktop GUI for Windows, macOS and Linux. Source setup requires Python 3.12 or later; the project also describes downloadable release packages.
- Fast mode favors speed but may miss some captions or introduce occasional errors.
- Automatic mode chooses a lightweight model on CPU and a more precise model on GPU, with slower extraction noted for GPU use.
- Precise mode checks frames individually and is described as very slow.
- Batch selection assumes matching video resolution and subtitle-region placement.
- The README warns against program and video paths containing spaces or Chinese characters.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Extract subtitles from one video in the GUI
Use the desktop interface to select a video, define its subtitle region and start extraction.
Before you start
- A video file.
- A downloaded and extracted application package, or a source installation with Python 3.12+ and the required runtime dependencies.
- Keep the video and program paths free of spaces and Chinese characters to avoid errors noted by the project.
Steps
Open the GUI and choose a single video with the Open control.
Adjust the subtitle region.
Choose a recognition mode and run extraction.
Expected outcome
The application extracts recognized hard-coded captions and generates an SRT or TXT subtitle file.
Guide limitations
- Fast mode may miss some captions or produce occasional recognition errors.
- Precise mode is described as very slow.
- GPU runtime setup depends on the selected hardware and supported drivers.
Prepare a batch of videos for subtitle extraction
Select multiple videos for batch subtitle extraction when they share consistent dimensions and subtitle placement.
Before you start
- Multiple video files with the same resolution and subtitle-region placement.
- The desktop application and a configured runtime.
Steps
Open the video selector.
Select multiple videos.
Ensure the videos have consistent resolution and subtitle-region placement.
Adjust the subtitle region and run extraction.
Expected outcome
The application processes the selected videos for subtitle extraction and produces subtitle files.
Guide limitations
- The project specifies consistent resolution and subtitle-region placement for batch jobs.
- Recognition modes can miss captions or trade speed for accuracy.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 14 October 2026
- Original website title
- YaoFANGUK/video-subtitle-extractor · Official homepage title · retained in its source language