# Hearthwave — NVIDIA CUDA deployment (build-from-source).  *** STUB ***
#
# This profile is scaffolded but NOT yet validated on an NVIDIA host. The image
# builds onnxruntime-gpu (see backend/Dockerfile COMPUTE_BACKEND=cuda), and the
# GPU is granted via the NVIDIA Container Toolkit `deploy.resources` block below.
# Before using:
#   1. Install the NVIDIA Container Toolkit on the host.
#   2. Confirm `docker run --rm --gpus all nvidia/cuda:12.4.0-base nvidia-smi` works.
#   3. Wire the STT final pass to CUDA (the GPU transcriber currently targets
#      torch "cuda" which works on NVIDIA too, but this path is untested here).
#   4. Remove this banner once validated.
#
#   bash setup-cuda.sh
#   docker compose -f docker-compose.cuda.yml up --build
name: hearthwave

services:
  backend:
    build:
      context: .
      dockerfile: backend/Dockerfile
      args:
        COMPUTE_BACKEND: cuda
    volumes:
      - ./data:/data
      - ./Voices:/app/Voices:ro
      - ./Models:/app/backend/Models:ro
      - /run/user/${PULSE_UID:-1000}/pulse:/run/pulse:ro
    environment:
      - PULSE_SERVER=unix:/run/pulse/native
      - NVIDIA_VISIBLE_DEVICES=all
      - NVIDIA_DRIVER_CAPABILITIES=compute,utility
    devices:
      - /dev/snd:/dev/snd
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
    restart: unless-stopped
    networks:
      - radio
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8765/health')"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 20s

  frontend:
    build:
      context: ./frontend
      dockerfile: Dockerfile
    ports:
      - "80:8080"
    volumes:
      - ./data:/data:ro
    depends_on:
      backend:
        condition: service_started
    restart: unless-stopped
    networks:
      - radio

networks:
  radio:
    driver: bridge