pyVideoTrans

Appearance

Local Video Hard Subtitle Extraction Tool ​

In video processing, it's now quite convenient to recognize speech and auto-generate subtitles (e.g., using Whisper models). However, if the video has subtitles "burned" into the image frames (hard subtitles), extracting them is not easy: online OCR interfaces often hit rate limits or errors, while paid services are costly.

To solve this problem, this tool integrates two excellent open-source projects to create a fully local, completely offline video hard subtitle extractor. It directly recognizes text in video frames and automatically generates .srt subtitle files. Currently supports Chinese, English, Japanese, and Korean — 4 languages.

Download: https://pan.baidu.com/s/1FrBVmdXAQBtnTelONv8Wog?pwd=1234

Must be extracted to an English-named folder — Chinese paths may cause errors.

Core Principles: ​

  • VideoSubFinder: Quickly locates frames in the video where hard subtitles appear and captures screenshots.
  • PaddleOCR-json (based on Baidu PaddleOCR): Recognizes text in these screenshots and converts it to text.

Features ​

  • Visual area selection: After importing video, use the mouse to drag-select the subtitle region in the preview window, avoiding interference from other on-screen text (logos, watermarks, etc.).
  • CUDA GPU acceleration: Improves text recognition performance.
  • Multi-language recognition: Supports Chinese, English, Japanese, and Korean hard subtitle extraction.
  • Environment: Currently supports Windows 10 and Windows 11 only.

Usage ​

Step 1: Launch ​

Download and extract the package, then double-click ocrsp.exe in the extracted folder.

Step 2: Import Video ​

On the left side of the interface, click to select the video file to process (ensure the video contains hard subtitles to extract).

Step 3: Select Subtitle Region ​

  1. In the video preview area on the right, hold the left mouse button and drag to select the subtitle region.
  2. For precise selection, use the "Step" button below the preview window to fine-tune the frame. Check different time points to ensure subtitles are fully contained within the selection.

Step 4: Start Extraction ​

Click the "Start" button. The log window at the bottom shows real-time progress. Wait for completion.

Step 5: Get Subtitles ​

After extraction, the generated subtitle file is automatically saved to the video's folder, named [original_video_name]-ocr.srt.

Hard Subtitle Removal Tools ​

Acknowledgments ​

This tool would not be possible without the following open-source projects: