USC EE434 Capstone Project.
The purpose of this project is to devise a solution to navigating around an indoor space, and to take visitors or workers to the correct destination, with minimal error and difficulties.
ROS Packages:
mpu_6050_driver robot_setup_tf jetson_nano_bot rplidar_ros hector_slam
Important Files:
RemoteController.ino ~ robot controlls
~/catkin_ws/src/mpu_6050_driver/scripts/imu_node.py
~/catkin_ws/src/mpu_6050_driver/scripts/tf_broadcaster_imu.py
~/catkin_ws/src/mpu_6050_driver/scripts/registers.py
~/catkin_ws/src/jetson_nano_bot/launch/jetson_nano_bot.launch
Run the Project:
roslaunch ~/catkin_ws/src/jetson_nano_bot/launch/jetson_nano_bot.launch Please install the software in requirements.txt before running the file.
Run Audio Processing Seperatly
Then use python3 ASR.py to run the program
If there are any programs not installed please use pip3 install #package# to install the software
When the training is over, it will display "Ready!" and then click any button of the mouse to start recording
After 3 seconds, the recording ends and the prediction starts.
folder: 'hmm-speech-recognition-0.1/audio' is all the training file stored
References: https://github.com/atta007/Speech-Recognizer/blob/main/speech_recognizer.ipynb