arabic-audio-reader-worker / scripts /setup_katib_ocr.sh
Syncre's picture
Deploy Arabic Audio Reader worker
2e1a095 verified
#!/usr/bin/env bash
set -euo pipefail
ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
VENV="$ROOT/.venv-katib-ocr"
PYTHON_BIN="${PYTHON_BIN:-python3.10}"
if [ ! -d "$VENV" ]; then
"$PYTHON_BIN" -m venv "$VENV"
fi
"$VENV/bin/python" -m pip install --upgrade pip
"$VENV/bin/python" -m pip install -r "$ROOT/requirements-katib-ocr.txt"
"$VENV/bin/python" -c "import torch, torchvision, transformers; from transformers import AutoModelForImageTextToText, AutoProcessor"
echo "KATIB Arabic OCR sidecar is ready at $VENV"
echo "Default model: oddadmix/Katib-Qwen3.5-0.8B-0.1"
echo "Use OCR_ENGINE=katib-ocr for smaller Arabic-trained OCR on short samples or a worker."