This repository contains drivers and ROS interfaces for the Picovoice libraries. Input files can be managed via the Picovoice console.
Prerequisites: Working ROS Noetic environment
sudo apt-get install ros-noetic-picovoice-driver
roscd && cd ../src # Navigate to the 'src' dir of your catkin workspace
git clone https://github.com/reinzor/picovoice_ros.git
cd picovoice_ros
rosdep install --from-path -y -i .
catkin_make # Compile workspace (or catkin build)
Start a roscore
:
roscore
Start the porcupine
recognizer from the picovoice_driver
package
Access key: https://console.picovoice.ai/
rosrun picovoice_driver picovoice_driver_porcupine _access_key:=[YOUR_ACCESS_KEY_HERE]
Start the axclient
in order to evaluate the action interface of the driver
rosrun actionlib_tools axclient.py /get_keyword # ros-noetic-actionlib-tools (for earlier distro's it should be included in the 'actionlib` package)
Set the following as Goal
keywords: [{
name: "porcupine",
url: "porcupine_linux"
}]
press SEND GOAL
, and say "Porcupine"
Start the rhino
recognizer from the picovoice_driver
package
Access key: https://console.picovoice.ai/
rosrun picovoice_driver picovoice_driver_rhino _access_key:=[YOUR_ACCESS_KEY_HERE]
Start the axclient
in order to evaluate the action interface of the driver
rosrun actionlib_tools axclient.py /get_intent # ros-noetic-actionlib-tools (for earlier distro's it should be included in the 'actionlib` package)
Set the following as Goal
context_url: 'coffee_maker_linux'
and press SEND GOAL
, and say "Small cappuccino"