L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Développement · SKILL.md

removing-subtitles

removing-subtitles

Removes hardcoded subtitles from videos using AI inpainting. Use when user wants to remove subtitles, erase text from video, clean video from captions, delete burned-in subtitles, remove video watermarks, clean hardcoded text, or strip embedded subtitles. 去字幕、去除字幕、删除字幕、清除字幕、去硬字幕、去水印、擦除字幕、移除字幕。

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: removing-subtitles description: Removes hardcoded subtitles from videos using AI inpainting. Use when user wants to remove subtitles, erase text from video, clean video from captions, delete burned-in subtitles, remove video watermarks, clean hardcoded text, or strip embedded subtitles. 去字幕、去除字幕、删除字幕、清除字幕、去硬字幕、去水印、擦除字幕、移除字幕。

Subtitle Removal

Removes hardcoded/burned-in subtitles from videos using OCR detection and AI inpainting.

When to Use

Use this skill when:

  • Removing hardcoded subtitles from videos
  • Cleaning up videos with burned-in text
  • Preparing videos for re-subtitling
  • Removing unwanted text overlays
  • Cleaning videos for repurposing

How It Works

  1. OCR Detection: AI detects subtitle regions using optical character recognition
  2. Region Masking: Identifies areas containing text
  3. AI Inpainting: Fills removed areas with reconstructed background content
  4. Output Generation: Produces clean video without subtitles

Workflow

Step 1: Get Source Video

Obtain the video URL with hardcoded subtitles.

Step 2: Submit Removal Task

Call submitAideoTask with:

  • prompt: Request subtitle removal
  • multiInputs: Array containing source video URL

Step 3: Poll for Results

  1. First check: Wait 210 seconds after submission
  2. Subsequent checks: Wait 30 seconds between polls
  3. Call getAideoTaskStatus to check progress

Step 4: Timeout Handling

Maximum wait time: 20 minutes

If task exceeds 20 minutes, report as timeout failure.

Step 5: Return Results

On completion, return the clean video URL.

Examples

Example 1: Remove Subtitles

User request: "Remove the subtitles from this video"

submitAideoTask:
  prompt: "Remove all hardcoded subtitles from this video"
  multiInputs: ["video_with_subtitles.mp4"]

Example 2: Clean Video for Re-subtitling

User request: "I want to add my own subtitles, please remove the existing ones"

submitAideoTask:
  prompt: "Erase all burned-in subtitles from this video to prepare for new subtitles"
  multiInputs: ["original_video.mp4"]

Example 3: Remove Text Overlays

User request: "Clean up this video by removing the text at the bottom"

submitAideoTask:
  prompt: "Remove the hardcoded text overlay at the bottom of this video"
  multiInputs: ["video_with_text.mp4"]

Important Notes

  • Uses OCR to detect subtitle regions automatically
  • AI inpainting fills removed areas naturally
  • Works best with subtitles on solid or simple backgrounds
  • Complex backgrounds may have some artifacts
  • Maximum wait time is 20 minutes
  • Processing time: ~10 minutes per 1-minute video
Skills enregistrées