AI Large-Scale Online Voice Assistant

1. Online Voice Configuration

Before setting up auto-start, we must ensure that the program itself can work independently while online. This requires modifying the configuration file.

1. Locate the Configuration File:

In your project code, find and open the configuration file:

config/yahboom.yaml

2. Modify Configuration Parameters:

Please check the following parameters in the file and ensure that their values are consistent with those shown below. If the parameters are not present, please add them.

xxxxxxxxxx
   asr:                                    # Voice Node Parameters
     ros__parameters:
       VAD_MODE: 2                        #VAD Sensitivity
       sample_rate: 16000                  #ASR Recording Audio Sampling Rate
       frame_duration_ms:  30              #VAD frame size (in milliseconds)
       use_oline_asr: True                # Whether to use online ASR (True: use online; False: use offline)
       mic_serial_port: "/dev/ttyUSB0"     #Microphone Serial Port Alias
       mic_index: 0                        #Microphone Index
       language: 'zh'                      #ASR Language
   
   model_service:                          # Model Server Node Parameters
     ros__parameters:
       language: 'zh'                      #Large Language Model API
       useolinetts: True                   # Whether to use online speech synthesis (True for online, False for offline)
   
        # Large Model Configuration
       # llm_platform: 'ollama'              # Available Platforms: 'ollama', 'tongyi', 'spark', 'qianfan', 'openrouter'
       llm_platform: 'tongyi'              # Here, we take Tongyi as an example.
   

• use_online_asr and use_online_tts must be set to tongyi.
• llm_platform must be set to tongyi.

With these settings, the entire system will operate in a purely online mode.

3. Save the file and rebuild the project to apply the changes:

xxxxxxxxxx
cd ~/yahboom_ws
colcon build
source install/setup.bash

Once this step is complete, the program will function as a purely online voice service.

2. Creating a Startup Script

Now, we will create a startup script to ensure that largemodel_control.launch.py runs automatically whenever the system boots up.

2.1 Creating the Startup Script

To ensure the system correctly loads the ROS 2 environment, the best practice is to create a simple script that opens a new terminal window to execute the launch command.

1. Create the script file: In the directory ~/yahboom_ws/src/largemodel/, create a file named start_largemodel.sh.

xxxxxxxxxx
vim ~/yahboom_ws/src/largemodel/start_largemodel.sh

2. Write the script content: Copy and paste the following content into the script file.

xxxxxxxxxx
#!/bin/bash
​
sleep 10
gnome-terminal -- bash -c "export RUN_PROGRAMS='ros2 launch largemodel largemodel_control.launch.py';exec bash"

Here, RUN_PROGRAMS specifies the command to be executed.

3. Save and exit the editor.
4. Grant execute permissions to the script:

xxxxxxxxxx
chmod +x ~/yahboom_ws/src/largemodel/start_largemodel.sh

5. Add the execution command to .bashrc

Open ~/.bashrc:

xxxxxxxxxx
vim ~/.bashrc

Copy and paste the following line to the very end of the file:

xxxxxxxxxx
eval $RUN_PROGRAMS

Save and exit. ### 2.2 Creating the Startup File

1. Open the system terminal, then create and edit the start_largemodel.desktop startup file:

xxxxxxxxxx
vim ~/.config/autostart/start_largemodel.desktop

2. Copy the following content into start_largemodel.desktop, then save and exit.

xxxxxxxxxx
[Desktop Entry]
Type=Application
Exec=sh /home/jetson/yahboom_ws/src/largemodel/start_largemodel.sh
Hidden=false
NoDisplay=false
X-GNOME-Autostart-enabled=true
Name[en_US]=start_largemodel
Name=start_largemodel
Comment[en_US]=
Comment=

In the above configuration: Exec specifies the command used to execute the script; X-GNOME-Autostart-enabled determines whether the program launches automatically upon system startup; and Name specifies the display name.

Restart the system, and the startup program will automatically launch upon boot.

xxxxxxxxxx
sudo reboot

3. Startup Management

After restarting the system, a system terminal window will pop up, automatically running the voice assistant program. If you need to temporarily exit the program, press the [Ctrl+C] keys, and then close the terminal window.

3.1 Disabling the Startup Program

If you no longer require the program to launch automatically upon system startup, edit the start_largemodel.desktop startup file. Change the X-GNOME-Autostart-enabled parameter to false, then save and exit. This ensures that the program will not launch the next time the system starts up.

xxxxxxxxxx
vim ~/.config/autostart/start_largemodel.desktop

Example Content:

xxxxxxxxxx
[Desktop Entry]
Type=Application
Exec=sh /home/jetson/yahboom_ws/src/largemodel/start_largemodel.sh
Hidden=false
NoDisplay=false
X-GNOME-Autostart-enabled=false
Name[en_US]=start_largemodel
Name=start_largemodel
Comment[en_US]=
Comment=

3.2 Enabling the Startup Program

If you require the program to launch automatically upon system startup, edit the start_largemodel.desktop startup file. Change the X-GNOME-Autostart-enabled parameter to true, then save and exit. Upon restarting the system, the program will launch automatically.

xxxxxxxxxx
vim ~/.config/autostart/start_largemodel.desktop

Example Content:

xxxxxxxxxx
 Multimodal autonomous proxy application
OpenClaw Quick Start on Jetson Platform