A self-hosted, browser-based utility for file conversion, OCR and audio/video transcription. It wraps common CLI and Python converters (FFmpeg, LibreOffice, Pandoc, ImageMagick, etc.), plus faster-whisper and Tesseract OCR.

• Convert between many file formats
• OCR for PDFs and images (Tesseract / ocrmypdf)
• Audio & Video transcription using Whisper (MP4, MKV, AVI, MOV, etc.)
• Speaker diarization - automatically identify different speakers (requires pyannote.audio)
• torchcodec for enhanced audio decoding (requires FFmpeg DLLs on Windows)
• Simple, responsive dark UI with drag-and-drop
• Background job processing with real-time status updates
• /settings page for configuring tools and OAuth
• CPU-only by default; GPU acceleration available

# Clone this repository
git clone https://github.com/akron2/filewizard-win.git
cd filewizard-win

# Python 3.10-3.12 recommended (3.13+ may have compatibility issues with some packages)
python --version

# Create and activate virtual environment
python -m venv venv

# Allow script execution (required once)
Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

.\venv\Scripts\Activate.ps1

# Install dependencies
pip install --upgrade pip
pip install -r requirements_windows.txt

# Run the application
.\run.bat

Open http://localhost:8000 in your browser.

For full functionality, install these tools:

# Install Chocolatey (run PowerShell as Administrator)
Set-ExecutionPolicy Bypass -Scope Process -Force
[System.Net.ServicePointManager]::SecurityProtocol = [System.Net.ServicePointManager]::SecurityProtocol -bor 3072
iex ((New-Object System.Net.WebClient).DownloadString('https://community.chocolatey.org/install.ps1'))

# Install tools
choco install ffmpeg
choco install tesseract
choco install libreeoffice
choco install pandoc
choco install poppler
choco install pkgconfiglite  # for html5_parser

• Tesseract OCR: https://github.com/UB-Mannheim/tesseract/wiki
• FFmpeg: https://ffmpeg.org/download.html
• LibreOffice: https://www.libreoffice.org/download/
• Pandoc: https://pandoc.org/installing.html
• Poppler: https://github.com/oschwartz10612/poppler-windows/releases

Speaker diarization automatically identifies different speakers in conversations.

When you first use diarization:

1. The app will automatically open Hugging Face pages in your browser
2. Log in (or create account)
3. Click "Accept" on model pages:
   • https://huggingface.co/pyannote/speaker-diarization-3.1
   • https://huggingface.co/pyannote/segmentation-3.0
4. Return to terminal and press Enter
5. Models will download automatically (~500MB)

• Enable "Identify Speakers (Diarization)" checkbox when transcribing
• Output format:

  [SPEAKER_00]:
  Hello, how are you?
  
  [SPEAKER_01]:
  I'm fine, thank you!
  

1. Open http://localhost:8000
2. Drag & drop or select files
3. Choose action: Convert, OCR, or Transcribe
4. Track progress in History table

Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

choco install tesseract

choco install ffmpeg

Close previous instance or change port in run.bat.

1. Ensure pyannote.audio is installed: pip install pyannote.audio pyannote.pipeline
2. Accept model terms on Hugging Face (see Speaker Diarization section)

If you see errors like OSError: Could not load this library: ...torchcodec\libtorchcodec_core*.dll:

1. Ensure torchcodec is installed: pip install torchcodec
2. If the error persists, torchcodec requires FFmpeg DLLs to be in PATH
3. Add your FFmpeg bin directory to PATH: set PATH=%PATH%;C:\path\to\ffmpeg\bin
4. Restart the application

Warning: Exposing this app publicly without authentication risks arbitrary code execution. Intended for local use or behind OAuth/OIDC.

• Original Repository: https://github.com/LoredCast/filewizard
• This Repository: https://github.com/akron2/filewizard-win
• Issues: https://github.com/akron2/filewizard-win/issues