Getting started with Pixi

Pixi makes cross-platform ROS 2 development easy. You can build and run both capture and playback nodes on Linux and Windows—with no system-wide ROS install.

---

Node details

For more details about the node see: ROS2 Lowwi Wakeword Detection Node page in implementation tab

📦 Install Pixi

Linux:

curl -fsSL https://pixi.sh/install.sh | bash

Windows (PowerShell):

powershell -ExecutionPolicy ByPass -c "irm -useb https://pixi.sh/install.ps1 | iex"

---

🚀 Clone & Build Project

git clone https://github.com/CLFML/lowwi.git
cd lowwi
pixi install
pixi run build

Or launch VSCode with the environment:

pixi run vscode

✅ Note (Windows): Always build in Release or RelWithDebInfo, not Debug!
(Ctrl+Shift+P → "CMake: Select Variant")

---

⚡ Using as a Pixi Dependency

Want to use custom_pkg from another Pixi-based project?

1. Init a new project

mkdir my_project && cd my_project
pixi init

2. Edit pixi.toml

Add these:

[project]
channels = [
  "https://fast.prefix.dev/conda-forge",
  "https://prefix.dev/robostack-jazzy",
  "https://clfml.github.io/conda_ros2_jazzy_channel/"
]

[dependencies]
ros-jazzy-ros-base = "*"
ros-jazzy-audio-tools = "*"
ros-jazzy-lowwi = "*"
colcon-common-extensions = "*"
rosdep = "*"

🧠 Optional: VSCode Support

Add to your pixi.toml:

[target.linux-64.dependencies]
python-devtools = "*"
pybind11 = "*"
numpy = "*"

[target.win-64.dependencies]
python-devtools = "*"

[target.linux-64.tasks]
vscode = 'env -u LD_LIBRARY_PATH code .'

[target.win-64.tasks]
vscode = "code ."

---

3. Run the node

pixi install
# Copy the models dir from Lowwi Git repo to your own project folder
pixi shell
# run this command to launch lowwi node with default models:
ros2 run lowwi lowwi_node \
  --ros-args \
  -p wakeword.phrases:="['Hey Mycroft', 'Hey Jarvis']" \
  -p wakeword.models:="['models/example_wakewords/hey_mycroft.onnx', 'models/example_wakewords/hey_jarvis.onnx']" \
  -p wakeword.min_activations:="[2, 2]"

3.5 Run the node with launch-file

For automation purposes you can use a launch-file ("lowwi_launch.py") like this:

from launch import LaunchDescription
from launch_ros.actions import Node

def generate_launch_description():
    return LaunchDescription([
        Node(
            package='lowwi',
            executable='lowwi_node',
            name='lowwi_node',
            output='screen',
            parameters=['params.yaml']  # Path to your params file
        )
    ])

with params.yaml file:

lowwi_node:
  ros__parameters:
    wakeword.phrases: ["Hey Mycroft", "Hey Jarvis"]
    wakeword.models: ["models/example_wakewords/hey_mycroft.onnx", "models/example_wakewords/hey_jarvis.onnx"]
    wakeword.min_activations: [2,2]

In pixi shell run:

ros2 launch lowwi_launch.py