A Next.js application for uploading and processing various types of content files to enable AI-powered learning experiences.

• Simple interface with two main actions: Upload Content and Learn
• Modern, responsive design

• Drag & drop or click to browse file upload
• Automatic file processing based on file type
• Organized storage with each file getting its own folder
• Support for multiple file types:
  • Text files (.txt, .md) - stored as-is
  • Documents (.pdf, .docx, .doc) - text extracted
  • Videos (.mp4, .mov, .avi, .webm) - audio transcribed + screenshots every 5 minutes
  • Images (.jpg, .png) - stored as-is (OCR planned)
  • Data files (.json, .csv) - stored as-is

When you upload a file, the system:

1. Creates a unique folder with timestamp and filename
2. Stores the original file
3. Processes based on file type:
   • PDFs/DOCX: Extracts text content to extracted_text.txt
   • Videos:
     • Extracts audio to extracted_audio.wav
     • Takes screenshots every 5 minutes
     • Creates placeholder for transcript (ready for Whisper API integration)
   • Text/Images: Keeps original format
4. Creates a processing summary with details

• View all uploaded files in organized folders
• See processing status and contents
• File size limit: 500MB per file
• Files stored in /content directory (gitignored)

• Interactive Q&A with uploaded content
• Content analysis and summaries
• Personalized study plans
• Auto-generated quizzes

• Frontend: Next.js 15, React, TypeScript, Tailwind CSS
• File Processing:
  • PDF parsing with pdf-parse
  • DOCX processing with mammoth
  • Video processing with ffmpeg
• Storage: Local file system (expandable to cloud storage)

1. Clone the repository

   git clone <repository-url>
   cd uc-berkeley-ai-hackathon-2025

2. Install dependencies

   npm install

3. Verify ffmpeg installation (required for video processing)

   ffmpeg -version

   If not installed on Mac: brew install ffmpeg

4. Run the development server

   npm run dev

5. Open your browser Navigate to http://localhost:3000

/content/                          # Upload directory (gitignored)
  └── 2024-01-15T10-30-00_document/ # Individual file folders
      ├── document.pdf              # Original file
      ├── extracted_text.txt        # Processed content
      ├── processing_summary.txt    # Processing details
      └── ...                       # Other processed files

/app/
  ├── api/
  │   ├── upload/                   # File upload endpoint
  │   └── files/                    # File listing endpoint
  ├── lib/
  │   └── file-processor.ts         # Core processing logic
  ├── upload/                       # Upload page
  ├── learn/                        # Learning page (coming soon)
  └── page.tsx                      # Homepage

Upload and process files

• Input: FormData with file
• Output: Processing results and folder info

List all uploaded files

• Output: Array of files/folders with metadata

• Extracts all text content
• Preserves formatting where possible
• Saves to extracted_text.txt

• Extracts raw text content
• Handles various document structures
• Saves to extracted_text.txt

• Audio Extraction: Converts to 16kHz mono WAV for transcription
• Screenshots: Captures frames every 5 minutes at high quality
• Transcript Placeholder: Ready for speech recognition integration
• Metadata: Duration and processing info

1. Audio Transcription: OpenAI Whisper API integration
2. OCR for Images: Text extraction from images
3. Cloud Storage: S3/GCS integration
4. Real-time Processing: WebSocket updates
5. Content Search: Full-text search across processed files
6. AI Integration: GPT-4 for content analysis and Q&A

• File processing is asynchronous and handles errors gracefully
• Original files are always preserved
• Processing failures don't prevent file storage
• All file operations use proper error handling
• TypeScript for type safety throughout

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test file processing with various file types
5. Submit a pull request

[Add your license here]