A versatile LangChain agent that automates tasks like calculations, video/audio processing, transcription, and translation.

• Mathematical calculations
• Video to audio conversion
• Audio transcription (Whisper with GPU support)
• Text translation (Ollama or OpenRouter)
• File operations
• CUDA/GPU diagnostics

1. Clone and setup:

git clone <your-repo-url>
cd simurgh
python -m venv .venv
.venv\Scripts\activate  # Windows
# or source .venv/bin/activate  # macOS/Linux
pip install -r requirements.txt

2. Optional - GPU support (for faster transcription):

pip uninstall torch torchvision torchaudio
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
python check_cuda.py

Copy .env.example to .env and fill in your API keys:

# Windows
copy .env.example .env

# macOS/Linux
cp .env.example .env

python app.py

Ask the agent to:

• "What is 25 + 37?"
• "Convert video.mp4 to audio"
• "Transcribe audio.mp3 to text"
• "Translate 'Hello' to French"
• "Translate file.txt to Persian"
• "Check my CUDA status"

• Calculator: Math operations
• VideoToAudioConverter: Extract audio from video
• AudioToTextConverter: Whisper transcription (GPU optional)
• Translator: Multi-language translation
• TextFileSaver: Save text to files
• CUDAChecker: GPU diagnostics

• Python 3.8+
• OpenRouter or OpenAI API key
• NVIDIA GPU with CUDA (optional, for GPU acceleration)
• Ollama (optional, for local translation)

APACHE