🔊 3D Printed Multi-Room Smart Speaker

A complete guide to building a custom-designed, 3D-printed smart speaker. This project leverages a Raspberry Pi Zero 2 W for the client hardware and uses Music Assistant with Snapcast for sub-millisecond multi-room audio synchronization.

✨ Features

Acoustically Optimized: Custom 3D printed enclosures designed to eliminate standing waves.
Perfectly Synced: Snapcast integration ensures multi-room audio with zero echo.
Centralized Control: Managed via Music Assistant and Home Assistant.
Low Power: Client nodes run on highly efficient Raspberry Pi Zero 2 W boards.

🛠️ Hardware Prerequisites

The Client (Speaker Node)

SBC: Raspberry Pi Zero 2 W
Audio Output: HiFiBerry AMP2 (or similar I2S DAC/Amp HAT)
Drivers: Raw speaker drivers (T/S parameters required for volume calculation)
Power: High-quality 5V power supply
Hardware: M3 heat-set brass inserts, M3 machine screws, Poly-fil (pillow stuffing), and adhesive weather-stripping (or TPU for a printed gasket).

🖨️ 3D Printing the Enclosure

To achieve true high-fidelity sound rather than a hollow plastic echo, the enclosure must be rigid, airtight, and acoustically tuned.

Acoustic Design

Calculate Volume: Start with a sealed enclosure. Use the Thiele/Small (T/S) parameters of your chosen speaker driver to calculate the exact internal volume required.
Ditch the Cube: Model the enclosure with swept curves, spheres, or teardrops. Curves prevent internal standing waves and add massive structural rigidity.
Thick Baffle: Design the front face (where the speaker mounts) to be 15mm - 20mm thick.

Slicer Settings

Material: PETG, ABS, or ASA. (Avoid PLA as it creeps under screw pressure and warps in heat).
Walls over Infill: Aim for 3-4 perimeters to achieve a 2mm-3mm solid shell. Total wall thickness should be roughly 8mm-10mm.
Infill: 15% - 40% Gyroid or Cubic infill for multi-directional strength.

Assembly

Melt heat-set inserts into the mounting holes to prevent screws from backing out under vibration.
Apply a gasket between the driver and the printed baffle to ensure an airtight seal.
Loosely pack the interior with Poly-fil to slow internal sound waves and improve bass response.

💻 Software Deployment

1. The Server (Control Plane)

Music Assistant handles both the media library and the Snapcast server. Deploying via Docker Compose is the cleanest method for a dedicated server environment like TrueNAS.

Crucial Note: You must use network_mode: "host" so Music Assistant can dynamically allocate TCP ports for incoming audio streams without Docker bridge bottlenecks.

services:
  music-assistant:
    image: ghcr.io/music-assistant/server
    container_name: music-assistant
    restart: unless-stopped
    network_mode: "host"
    volumes:
      - /path/to/your/appdata/music-assistant:/data
      # Mount your local media directories (e.g., from your ZFS pool)
      - /path/to/your/media:/media:ro

2. The Client (Pi Zero 2 W)

Flash the Pi Zero 2 W with a lightweight OS like Raspberry Pi OS Lite.

Install the Snapcast client daemon:

sudo apt update
sudo apt install snapclient

Configure snapclient to point to your Music Assistant server IP, connecting on port 1704.
Configure your /boot/config.txt to enable the I2S overlay for your specific DAC/Amp HAT.

🌐 Network Requirements

Server: Wired Ethernet is highly recommended.
Clients: A strong, stable 2.4GHz Wi-Fi connection.
Ports: TCP 1704 (Audio Stream) and TCP 1705 (Control/JSON-RPC) must be open.