Skip to main content

AI Media Workflow Studio

Transform your raw video content into polished, platform-ready assets using our intuitive, node-based media processing engine. By connecting simple functional blocks, you can automate complex AI tasks like subtitling, reframing, and highlight extraction.

Getting Started​

The Media Workflow Studio operates on a Source → Process → Sink architecture. You define where the video comes from, what happens to it, and where the final result is stored.

Step 1: Media Input​

The journey begins at the Media Input node. Currently, the studio is optimized for seamless integration with Cloudinary, with more providers coming soon.

  • Cloudinary Integration: Simply paste your Cloudinary URL.
  • Auto-Detection: The system automatically identifies the file type (e.g., .webm, .mp4) to ensure compatibility with downstream tools.
Coming Soon

Support for Amazon S3 and Public URL fetches are currently in development to give you more flexibility over your storage providers.

Step 2: Intelligent Processing Tools​

Once your media is ingested, you can pipe it into one of our AI-powered processing tools.

Our most popular tool uses advanced speech-to-text models to transcribe your audio into professional-grade subtitles.

  • Output Formats: Generates both .srt files and plain text.
  • Language Support: Features Auto-detect for input audio and multi-language output (Default: English).

2. Reframe Video​

Perfect for social media managers. Convert horizontal (16:9) video to vertical (9:16) format.

  • AI Subject Tracking: The AI doesn't just crop; it tracks the most important subject in the frame to ensure they stay center-stage.

3. Extract Highlights​

Save hours of manual editing. Our AI analyzes your footage to detect and extract the most engaging, "viral-ready" moments for short-form content.

Step 3: Media Output​

After processing, your new assets need a home. The Media Output node handles the automated upload.

  • Cloud Name: Link your specific Cloudinary instance.
  • Upload Presets: Use unsigned upload presets to define exactly how and where your processed files (like .srt subtitles) are stored in your media library.
Pro Tip

Use descriptive Upload Presets like llm_video_subtitle to keep your cloud storage organized automatically.

The Visual Workflow Canvas​

The heart of the application is the Visual Canvas. This drag-and-drop interface allows you to see the logic of your automation at a glance.

  1. Start Node: Triggers the beginning of the flow.
  2. Input Node: Fetches the source file.
  3. Tool Node: Performs the AI heavy lifting (e.g., Generate Subtitles).
  4. Output Node: Finalizes the upload to your destination.
  5. End Node: Confirms successful completion.

Key Benefits​

FeatureDescription
No-Code LogicBuild complex media pipelines without writing a single line of code.
AI-PoweredLeverage state-of-the-art models for transcription and subject tracking.
Cloud NativeBuilt to work directly with your existing cloud storage infrastructure.
ExtensibleMix and match nodes to create the perfect workflow for your specific needs.

Need Help?​

Check out our VibeAgentCopilot or reach out to the engineering team via the Support Chat node inside the application!