sulat 0.1.1

An opinionated way of doing AI - Task-oriented ai nodes for general-computing

sulat

an Opinionated Way of doing AI - Task-oriented AI nodes

Installation

From PyPI (Recommended)

pip install sulat

From Source

uv venv
source .venv/bin/activate
uv pip install -e .

Usage

import sulat

# Basic transcription (uses Whisper model)
text = sulat.transcribe("audio.wav")

# High performance transcription (uses Canary model)
text = sulat.transcribe("audio.wav", high_performance=True, source_lang="es")

# Control source and target languages
text = sulat.transcribe("audio.wav", source_lang="es", target_lang="en")

# Additional options
text = sulat.transcribe(
    "audio.wav",
    high_performance=False,  # Use Whisper (default) or Canary model
    source_lang="auto",      # Auto-detect or specify language
    target_lang="en",        # Target language for translation
    response_format="json",  # Response format
    temperature=0.0          # Temperature for generation
)

Environment Setup

Create a .env file in your project root:

# .env
SURUS_API_KEY=your_api_key_here

Or set the environment variable directly:

export SURUS_API_KEY="your_api_key"

Development

# Setup environment
uv venv
source .venv/bin/activate
uv pip install -e ".[dev]"

# Run tests
pytest tests/