respeaker_ros

A ROS Package for Respeaker Mic Array

Supported Devices

• Respeaker Mic Array v2.0

  

Preparation

1. Install this package

   Assumed that ROS is properly installed.

   You can install this package via apt-get.

    sudo apt-get install ros-$ROS_DISTRO-respeaker-ros
    

Or you can also build from the source.

    mkdir -p ~/catkin_ws/src && ~/catkin_ws/src
    git clone https://github.com/jsk-ros-pkg/jsk_3rdparty.git
    cd ~/catkin_ws
    source /opt/ros/kinetic/setup.bash
    rosdep install --from-paths src -i -r -n -y
    catkin config --init
    catkin build respeaker_ros
    source ~/catkin_ws/devel/setup.bash
    

1. Register respeaker udev rules

   Normally, we cannot access USB device without permission from user space. Using udev, we can give the right permission on only respeaker device automatically.

   Please run the command as followings to install setting file:

    roscd respeaker_ros
    sudo cp -f $(rospack find respeaker_ros)/config/60-respeaker.rules /etc/udev/rules.d/60-respeaker.rules
    sudo systemctl restart udev
    

And then re-connect the device.

1. Update firmware

    git clone https://github.com/respeaker/usb_4_mic_array.git
    cd usb_4_mic_array
    sudo python dfu.py --download 6_channels_firmware.bin  # The 6 channels version 
    

How to use

1. Run executables

    roslaunch respeaker_ros sample_respeaker.launch
    rostopic echo /sound_direction     # Result of DoA
    rostopic echo /sound_localization  # Result of DoA as Pose
    rostopic echo /is_speeching        # Result of VAD
    rostopic echo /audio               # Raw audio
    rostopic echo /speech_audio        # Audio data while speeching
    

You can also set various parameters via `dynamic_reconfigure`.

    sudo apt install ros-kinetic-rqt-reconfigure  # Install if not
    rosrun rqt_reconfigure rqt_reconfigure
    

To set LED color, publish desired color:

    rostopic pub /status_led std_msgs/ColorRGBA "r: 0.0
    g: 0.0
    b: 1.0
    a: 0.3"
    

Use cases

Voice Recognition

• ros_speech_recognition
• julius_ros

Notes

The configuration file for dynamic_reconfigure in this package is created automatically by reading the parameters from devices. Though it will be rare case, the configuration file can be updated as followings:

File truncated at 100 lines see the full file

Changelog for package respeaker_ros

2.1.24 (2021-07-26)

2.1.23 (2021-07-21)

2.1.22 (2021-06-10)

• [respeaker_ros] Specify correct Python version in package.xml (#247)

  • Fixes typo introduced in https://github.com/jsk-ros-pkg/jsk_3rdparty/commit/39be21894a38112a1633cf8385caf079c35536ff

• Add respawn_delay to respeaker_node.py to reduce CPU load (#241)

  • Kill respeaker_node.py when USBError occur to reduce CPU load

• [respeaker_ros] fix default sound_play action name of speech_to_text.py (#239)

• Mistakes I found when I tried to use respeaker_ros on hirovision (#217)

  • add speech_recognition_msgs
  • change the name from respeaker.launch to sample_respeaker.launch

• Contributors: Koki Shinjo, Miyabi Tanemoto, Naoya Yamaguchi, Shun Hasegawa

2.1.21 (2020-08-19)

2.1.20 (2020-08-07)

2.1.19 (2020-07-21)

2.1.18 (2020-07-20)

• Fix for noetic (#200)
  • respeaker_ros: use catkin_install_python to install scripts
  • fix 2to3, with print, raise, exception
  • use package.xml format 3 for package contains python depends
• [respeaker_ros] use rospy.logerr instead of print (#206)
• Add args to ros_speech_recognition (#197)
  • [respeaker_ros] update run_depend and test file to pass travis test properly
• Contributors: Kei Okada, Naoya Yamaguchi, Shingo Kitagawa

2.1.17 (2020-04-16)

2.1.16 (2020-04-16)

• [respeaker_ros] increase timeout to pass the test https://api.travis-ci.org/v3/job/554008643/log.txt
• Contributors: Kei Okada

2.1.15 (2019-12-12)

• [respeaker_ros] add pixel-ring in run_depend of respeaker_ros (#184)
• [respeaker_ros] install config dir in respeaker_ros (#185)
• Contributors: Shingo Kitagawa

2.1.14 (2019-11-21)

2.1.13 (2019-07-10)

• [respeaker_ros] increase timeout to pass the test (#170)
  • default tts_action_names should be soundplay
  • [respeaker_ros] increase timeout to pass the test https://api.travis-ci.org/v3/job/554008643/log.txt
  • [respeaker_ros] add python-speechrocognition-pip to package.depends, because scripts/speech_to_text.py depends on it
• [respeaker_ros] Add test file for speech_to_text (#164)
  • add test file for speech_to_text
• [respeaker_ros] add tts_action_names param: do not listen when the robot is speaking either japanese or english (#168)
  • add tts_actions param: do not listen when the robot is speaking either japanese or english
• Contributors: Naoya Yamaguchi, Shingo Kitagawa

2.1.12 (2019-05-25)

• Make sample_respeaker.launch re-usable (#161)
  • [respeaker_ros] add docs for each args in sample_respeaker.launch
  • make sample_respeaker.launch re-usable
• respeaker_ros: cleanup error messages (#155)
• Contributors: Yuki Furuta, Kei Okada, Naoya Yamaguchi

File truncated at 100 lines see the full file