Description
See the subsystem description page for overview:
Fall 2021
Fine adjustments were made to improve the user experience with the voice subsystem. These improvements include changing the keyword from “Coborg” to “Hey Coborg” and adding a command timeout feature for when the keyword is recognized but no speech follows.
Spring 2021
- Speaker installed on top cross beam of CoBorg frame. Microphone installed on right shoulder strap facing upwards toward the user’s face. Setup was shown and demonstrated during SVD and SVD encore.
- Microphone and speaker components bought and set for SVD. False positive refinement and further validation testing performed on voice subsystem.
- Voice recognition and signal output node created on ROS Melodic framework. Demonstrations performed to verify desired output is being transmitted based on appropriate voice input.
- Microphones implemented onto ROS Melodic framework. Pocket Sphinx voice recognition package onto ROS Melodic framework.
Fall 2020
- Conceptual design made and presented. Subsystem drafted to incorporate Pocket Sphinx voice recognition package onto ROS to detect desired voice inputs.
- Schedule and technical roadmap draft built for subsystem. These include functional and cyberphysical architectures.
- Preliminary requirements built for vision subsystem.
- As-Is CoBorg was examined. Fully-developed voice subsystem not fully developed onto robot.