Auto Caption Video

Automatically add beautiful, animated TikTok/Reels-style subtitles to your videos using AI-powered transcription.

Features

• Accurate Transcription: AI-powered speech-to-text with word-level timestamps
• Animated Subtitles: Modern TikTok/Reels-style animations with word-by-word highlighting
• Professional Styling: Bold fonts, color highlighting, smooth pop-up animations
• Multiple Sources: Supports direct video URLs and Google Drive links
• Customizable: Adjust font size, colors, position, words per line, and more
• Fast Processing: Optimized ffmpeg settings for quality and speed
• Multiple Formats: Supports MP4, MOV, AVI, MKV, and WEBM videos
• 50+ Languages: Auto-detection or specify your language

How It Works

1. Download: Fetches your video from a URL or Google Drive
2. Extract: Extracts audio optimized for transcription
3. Transcribe: AI generates text with precise word timings
4. Generate: Creates animated ASS subtitles with professional styling
5. Process: Burns subtitles into the video permanently
6. Deliver: Returns a downloadable video with embedded subtitles

Pricing

Simple, Fixed Pricing: $0.17 per minute of video

Video Length	Cost
1 minute	$0.17
2 minutes	$0.34
3 minutes	$0.51
5 minutes	$0.85

All-inclusive pricing. No hidden fees or API costs.

Memory Requirements

Use at least 4GB RAM for optimal performance. Running with 1GB RAM will result in significantly slower processing times. The actor is configured to use 4GB by default.

Input Parameters

Required

• videoUrl (string): Direct URL to your video file or Google Drive share link
  • Example: https://example.com/video.mp4
  • Example: https://drive.google.com/file/d/{FILE_ID}/view

Optional Styling

• language (string): Language code for transcription (e.g., 'en', 'es', 'fr')

  • Default: Auto-detect
  • Improves accuracy if specified

• fontSize (number): Base font size in pixels

  • Default: 52
  • Range: 20-100

• fontColor (string): Primary font color in hex format

  • Default: 'FFFFFF' (white)
  • Example: 'FF0000' for red

• highlightColor (string): Color for highlighted words in hex format

  • Default: 'FFFF00' (yellow)
  • Creates the TikTok-style word emphasis

• wordsPerLine (number): Maximum words per subtitle line

  • Default: 4
  • Range: 1-10
  • Lower values = shorter, punchier subtitles

• position (string): Subtitle position on screen

  • Options: 'top', 'center', 'bottom'
  • Default: 'bottom'

Example Input

{
  "videoUrl": "https://example.com/my-video.mp4",
  "language": "en",
  "fontSize": 52,
  "fontColor": "FFFFFF",
  "highlightColor": "FFFF00",
  "wordsPerLine": 4,
  "position": "bottom"
}

Output

The actor returns:

• outputUrl: Direct download link to your processed video
• videoInfo: Metadata about the original video
• transcriptionInfo: Details about the transcription (language detected, word count)
• status: Status updates throughout the process

Processing Time

Video Duration	Estimated Time
1 minute	~2-3 minutes
3 minutes	~4-5 minutes
5 minutes	~6-8 minutes

Processing time depends on video resolution, complexity, and server load

Limitations

• Maximum video size: 500MB
• Maximum duration: 5 minutes
• Supported formats: MP4, MOV, AVI, MKV, WEBM
• Audio quality: Must have clear speech for accurate transcription

Google Drive Setup

To use Google Drive links:

1. Right-click your video in Google Drive
2. Click "Get link"
3. Set permissions to "Anyone with the link"
4. Copy the link (should look like https://drive.google.com/file/d/...)
5. Paste into videoUrl input

The actor will automatically convert it to a direct download link.

Tips for Best Results

Video Quality

• Use videos with clear audio and minimal background noise
• Higher resolution videos (1080p) look better but take longer to process
• Ensure good lighting and framing for professional results

Subtitle Styling

• White text with yellow highlights is most readable (default)
• Use 3-5 words per line for TikTok/Reels style
• Larger font sizes (50-60px) work better for mobile viewing

Language

• Specify the language code if you know it for better accuracy
• Supports 50+ languages
• Auto-detect works well but may be slightly less accurate

Troubleshooting

"Video file too large"

• Compress your video before uploading
• Maximum size is 500MB

"Video duration exceeds maximum"

• Maximum duration is 5 minutes
• Trim your video before processing

"Transcription failed"

• Verify the video has audible speech
• Check audio quality is clear

"Download failed"

• Check the video URL is accessible
• For Google Drive, ensure link sharing is enabled
• Try downloading the video directly to test

"Invalid video format"

• Convert your video to MP4, MOV, AVI, MKV, or WEBM
• Use a tool like HandBrake or ffmpeg to convert

Technical Details

Transcription

• Accuracy: Word-level timestamps for precise synchronization
• Languages: 50+ supported with auto-detection

Subtitle Format

• Format: ASS (Advanced SubStation Alpha)
• Features: Supports animations, colors, and effects
• Style: TikTok/Reels-inspired with word-by-word highlighting

Video Processing

• Codec: H.264 (libx264) for wide compatibility
• Quality: CRF 23 (high quality)
• Audio: AAC codec at 192kbps
• Subtitles: Permanently burned into video (hard subtitles)

Support

For issues, questions, or feature requests:

• Check the troubleshooting section above
• Review your input parameters

License

MIT

Version

1.0.0