google_example.launch

<launch>

  <!-- you can choose chatbot_engine "Chaplus" or "A3RT" or "Mebo" currently-->
  <arg name="chatbot_engine" default="Mebo" />

  <arg name="use_sample" default="false" />
  <arg name="use_respeaker" default="true" doc="set false if you do not use respeaker"/>
  <arg name="chaplus_apikey_file" default="$(find chaplus_ros)/apikey.json" />

  <group if="$(arg use_respeaker)">
    <!-- node to convert /speech_to_text (speech_recognition_msgs/SpeechRecognitionCandidates) to /request (std_msgs/String) -->
    <node pkg="topic_tools" type="relay_field" name="sound_request_to_request"
          args="--wait-for-start /speech_to_text /request std_msgs/String
		'data: m.transcript[0]'" />
  </group>

  <group unless="$(arg use_respeaker)">
    <!-- start talker
	 subscribe /robotsound_jp sound_play/SoundRequest -->
    <include file="$(find aques_talk)/launch/aques_talk.launch" />

    <!-- start listener
	 publishes /Tablet/voice speech_recognition_msgs/SpeechRecognitionCandidates -->
    <include file="$(find ros_speech_recognition)/launch/speech_recognition.launch" >
      <arg name="launch_sound_play" value="false" />
      <arg name="continuous" value="true" />
      <arg name="engine" default="GoogleCloud" />
      <arg name="language" value="ja" />
      <arg name="n_channel" value="2" />
      <arg name="depth" value="16" />
      <arg name="sample_rate" value="44100" />
      <arg name="device" value="hw:0,0" />
    </include>
    <param name="/speech_recognition/google_cloud_credentials_json"
           value="/home/a-fujii/Downloads/eternal-byte-236613-4bc6962824d1.json" />
    <!--
	<param name="/speech_recognition/diarizationConfig"
        type="yaml"
        value="{'enableSpeakerDiarization': True, 'maxSpeakerCount': 3}" />
    -->

    <!-- node to convert /Tablet/voice (speech_recognition_msgs/SpeechRecognitionCandidates) to /request (std_msgs/String)
	 c.f.: https://github.com/ros/ros_comm/pull/639#issuecomment-618750038 -->
    <node pkg="topic_tools" type="relay_field" name="sound_request_to_request"
          args="--wait-for-start /Tablet/voice /request std_msgs/String
		'data: m.transcript[0]'" />
  </group>

  <!-- node to convert /response (std_msgs/String) to /robotsound_jp (sound_play/SoundRequest) -->
  <node pkg="topic_tools" type="relay_field" name="string_to_sound_request"
        args="--wait-for-start /response /robotsound_jp sound_play/SoundRequest
              '{sound: -3, command: 1, volume: 1.0, arg: m.data}'" />

  <!-- start chaplus
       subscribe /request and publish /response -->
  <node pkg="chaplus_ros" type="chaplus_ros.py" name="chaplus_ros"
        output="screen" >
    <rosparam subst_value="true">
      chatbot_engine: $(arg chatbot_engine)
      use_sample: $(arg use_sample)
      chaplus_apikey_file: $(arg chaplus_apikey_file)
    </rosparam>
  </node>

</launch>