Hugging Face's logo

Spaces:
Shreevathsam
/
Video_generator_tool
Runtime error

App Files Community
Video_generator_tool / DEPLOYMENT_GUIDE.md
Shreevathsam's picture
Shreevathsam
Upload 7 files
598e1f3 verified
preview code
|
raw
history blame contribute delete
8.01 kB

A newer version of the Gradio SDK is available: 6.3.0

Upgrade

🚀 Complete Deployment Guide to Hugging Face

Before You Start

Prerequisites

  • A Hugging Face account (Sign up here)
  • A Google API key with Gemini API access (Get one here)
  • At least 5-10 video clips (MP4 format recommended)

Part 1: Create Your Hugging Face Space

Step 1: Create New Space

  1. Go to https://huggingface.co/spaces
  2. Click the "Create new Space" button (top right)
  3. Fill in the details:
    • Owner: Your username
    • Space name: Choose a name (e.g., ai-video-generator)
    • License: Apache 2.0 (recommended)
    • Select the Space SDK: Choose Gradio
    • Space hardware: Start with CPU basic (free) - you can upgrade later
    • Space visibility: Public or Private (your choice)
  4. Click "Create Space"

Part 2: Upload Your Code Files

Step 2: Upload Main Files via Web Interface

  1. You should now be in your Space's main page
  2. Click on "Files and versions" tab
  3. Click "Add file" → "Upload files"
  4. Upload these files (download from the artifacts above):
    • app.py
    • requirements.txt
    • packages.txt
    • README.md
    • .gitignore
    • .gitattributes
  5. In the commit message box, type: Initial commit - upload main files
  6. Click "Commit changes to main"

Alternative: Using Git (Advanced)

If you're comfortable with Git:

# Clone your space
git clone https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE_NAME
cd YOUR_SPACE_NAME

# Copy your files
cp /path/to/downloaded/app.py .
cp /path/to/downloaded/requirements.txt .
cp /path/to/downloaded/packages.txt .
cp /path/to/downloaded/README.md .
cp /path/to/downloaded/.gitignore .
cp /path/to/downloaded/.gitattributes .

# Commit and push
git add .
git commit -m "Initial commit - upload main files"
git push

Part 3: Configure API Key

Step 3: Add Google API Key

🚨 CRITICAL: Never put your API key directly in the code!

  1. In your Space, click "Settings" (gear icon in top right)
  2. Scroll down to "Repository secrets"
  3. Click "New secret"
  4. Enter:
    • Name: GOOGLE_API_KEY
    • Value: Paste your actual Google API key
  5. Click "Add"

Part 4: Upload Video Clips

Step 4: Create video_clips Folder and Upload Videos

Option A: Using Web Interface

  1. Go back to "Files and versions" tab
  2. Click "Add file" → "Upload files"
  3. Create folder structure by typing: video_clips/
  4. Upload your video files (.mp4, .avi, .mkv, or .mov)
  5. If videos are larger than 10MB, you'll need to use Git LFS (see below)
  6. Commit with message: Add video clips

Option B: Using Git with LFS for Large Files

# Make sure you're in your cloned space directory
cd YOUR_SPACE_NAME

# Install Git LFS if not already installed
git lfs install

# Create video_clips folder
mkdir video_clips

# Copy your video files
cp /path/to/your/videos/*.mp4 video_clips/

# Add and commit
git add video_clips/
git commit -m "Add video clips"
git push

Video Requirements:

  • At least 5-10 video clips
  • Formats: .mp4, .avi, .mkv, or .mov
  • Resolution: 720p or 1080p recommended
  • Duration: 5-30 seconds per clip works best

Part 5: Optional - Add Background Music

Step 5: Upload Background Music (Optional)

  1. Go to "Files and versions"
  2. Click "Add file" → "Upload files"
  3. Create folder: background_music/
  4. Upload your music file (.mp3 or .wav)
  5. Commit with message: Add background music

Part 6: Wait for Build and Test

Step 6: Monitor Build Process

  1. Go back to your Space's main page (click the Space name at top)
  2. Click on "Logs" tab
  3. Watch the build process:
    • Installing system packages (imagemagick, ffmpeg)
    • Installing Python packages
    • Starting the app

Build Status Indicators:

  • 🟡 Yellow "Building": In progress
  • 🟢 Green "Running": Successfully built!
  • 🔴 Red "Failed": Check logs for errors

Step 7: Test Your App

Once the status shows "Running":

  1. Go to the "App" tab
  2. You should see your video generator interface!
  3. Test it:
    • Enter some text (e.g., "Hello world, this is a test video")
    • Select a voice
    • Click "Generate Video"
    • Wait for the video to generate

Part 7: Troubleshooting Common Issues

Issue 1: "Video clips folder not found"

Solution: Upload videos to the video_clips folder (Step 4)

Issue 2: "TTS failed: Error"

Solution:

  • Check that API key is correctly set in Settings → Repository secrets
  • Verify the key name is exactly GOOGLE_API_KEY
  • Test your API key at https://aistudio.google.com/

Issue 3: Build Fails

Solution:

  • Check the Logs tab for specific error messages
  • Common fixes:
    • Ensure requirements.txt is properly uploaded
    • Ensure packages.txt is properly uploaded
    • Try restarting the Space (Settings → Factory reboot)

Issue 4: Out of Memory

Solution:

  • Upgrade to better hardware (Settings → Change hardware)
  • Or use fewer/smaller video clips

Issue 5: Videos Upload Fails

Solution:

  • Use Git LFS for files over 10MB
  • Or use smaller video files
  • Or compress videos before uploading

Part 8: Optimize and Share

Step 8: Upgrade Hardware (Optional)

For better performance:

  1. Go to Settings
  2. Scroll to "Hardware"
  3. Choose a better option:
    • CPU Upgrade ($0.60/hour)
    • T4 GPU ($0.60/hour) - Recommended for faster processing
    • A10G GPU ($3/hour) - For very fast processing
  4. Click "Update"

Step 9: Share Your Space

Your Space is now live! Share it:

  • Direct URL: https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE_NAME
  • Click "Share" button to get embed codes
  • Add to your portfolio or website

Quick Checklist

Use this to ensure everything is set up:

  • ✅ Created Hugging Face Space
  • ✅ Uploaded app.py
  • ✅ Uploaded requirements.txt
  • ✅ Uploaded packages.txt
  • ✅ Uploaded README.md
  • ✅ Added GOOGLE_API_KEY to Repository secrets
  • ✅ Uploaded video clips to video_clips/ folder
  • ✅ (Optional) Uploaded background music
  • ✅ Space shows "Running" status
  • ✅ Tested video generation successfully

Need Help?

  1. Check Logs: Always check the Logs tab for error messages
  2. Community: Ask in Hugging Face Discord
  3. Documentation: Gradio Spaces Docs

Summary of File Locations

After setup, your Space structure should look like:

your-space-name/
├── app.py                    ✅ Main application file
├── requirements.txt          ✅ Python dependencies
├── packages.txt             ✅ System dependencies  
├── README.md                ✅ Documentation
├── .gitignore               ✅ Git ignore rules
├── .gitattributes           ✅ LFS configuration
├── video_clips/             ✅ Your video files (REQUIRED!)
│   ├── clip1.mp4
│   ├── clip2.mp4
│   └── clip3.mp4
├── background_music/        ⭕ Optional music files
│   └── bgmusic.mp3
├── voice_over/              ⚙️ Auto-created by app
└── exports/                 ⚙️ Auto-created by app

Legend:

  • ✅ = Required, you upload this
  • ⭕ = Optional
  • ⚙️ = Auto-created by the app

🎉 Congratulations!

You've successfully deployed your AI Video Generator to Hugging Face Spaces!

Now you can create amazing videos with AI voiceovers and animated subtitles right from your browser.