A modern web application for transcribing audio files and YouTube videos using AssemblyAI or OpenAI Whisper.

• Multiple Transcription Models: Support for AssemblyAI and OpenAI Whisper
• YouTube Integration: Direct transcription from YouTube URLs with automatic title extraction
• File Upload: Upload local audio files (MP3, WAV, M4A, OGG, FLAC, AAC)
• Smart Filenames: Downloads use original file/video names
• Modern UI: Clean, responsive card-based design
• Copy & Download: Copy to clipboard or download as TXT files
• Security: Input validation, file size limits, and security headers

1. Build the Docker image:

   docker build -t transcriber .

2. Run the container:

   docker run -p 3000:3000 transcriber

3. Access the application:

   http://localhost:3000
   

1. Install dependencies:

   pip install -r requirements.txt

2. Install yt-dlp (for YouTube support):

   pip install yt-dlp

3. Start the web server:

   python3 app.py

4. Access the application:

   http://localhost:3000
   

You'll need API keys for the transcription services:

• AssemblyAI: Sign up at assemblyai.com
• OpenAI: Sign up at openai.com

• MP3, WAV, M4A, OGG, FLAC, AAC
• Maximum file size: 100MB

• All YouTube URLs (youtube.com, youtu.be)
• Automatic audio extraction
• Video title used for download filename

• Minimal Design: Clean black and white theme
• Responsive Layout: Works on desktop and mobile
• Card-based UI: Modern card design with proper spacing
• Model Selection: Radio button interface for choosing transcription model
• Smart Filenames: Downloads use original file/video names

• Copy to Clipboard: One-click copy with visual feedback
• Download as TXT: Automatic download with original filename
• New Transcription: Start over with fresh transcription
• Form Auto-hide: Form disappears when transcript is shown

• Input Validation: API key format and length validation
• File Validation: Type and size restrictions
• URL Validation: YouTube URL format checking
• Security Headers: XSS protection, clickjacking prevention
• HTTPS Ready: Works with HTTPS proxy (as you mentioned)

docker build -t transcriber .

docker run -p 3000:3000 transcriber

docker run -p 3000:3000 -v /host/path:/app/uploads transcriber

• Python 3.11+
• yt-dlp (for YouTube support)
• Flask, requests, openai

• Port: 3000 (configurable in app.py)
• Host: 0.0.0.0 (accessible from any IP)
• Debug Mode: Enabled for development
• Temp Directory: /tmp (for file processing)

• Development Server: Uses Flask development server (not for production)
• Temporary Files: Automatically cleaned up after processing
• API Keys: Not stored - entered per session for security
• Rate Limiting: Removed for personal use

1. "yt-dlp not found": Install with pip install yt-dlp
2. "API key invalid": Ensure key is at least 20 characters
3. "File too large": Maximum file size is 100MB
4. "Invalid YouTube URL": Use standard YouTube URLs

1. Port already in use: Change port with -p 3001:3000
2. Permission denied: Run with sudo or check file permissions
3. Build fails: Ensure Dockerfile and requirements.txt are in the same directory