In recent years, there has been a rapid advancement in the fields of artificial intelligence (AI) and robotics. This has led to the development of personal voice assistants, such as Amazon's Alexa and Apple's Siri, which use natural language processing algorithms to understand user requests and provide appropriate responses. However, these voice assistants are limited to providing audio-based responses and do not have a physical presence. We have developed a Voice Assistant robot with a vision system using machine learning. Our system is integrated with an ESP32 CAM Module for object detection, enabling it to not only respond to user commands through audio but also to use visual cues to provide a more natural and intuitive experience. For our project, we used Amazon Echo Dot 3rd generation as the voice assistant system. We customized the Eco-Dot firmware to enable the system to control various sensors, servos, and other robotic components.
The robotic head is equipped with sensors, microphones, and speakers that enable it to capture and respond to user commands in real time. A 3D printer was used to prototype the robotic head. For robotic head control, an Arduino UNO microcontroller is used to control jaw actions and robotic neck turning controls. The KA2284 LED Sound Meter Module is used for converting voice assistant audio output signals into servo motor actuation for robotic jaw motion with the help of ultrasonic sensors and a metal gear servo motor. Neck Control is implemented with sensors and a voice assistant unit. Digital eyes with realistic human eye animations are fixed in the robot head. For that, we used the Seeed XIANO-21 microcontroller module in conjunction with two pairs of 1.44-inch color LCD screen modules.
In addition to using the Amazon Eco Dot 3rd gen as the voice assistant for our robotic head, we have made several modifications to improve the system's functionality. Firstly, we have customized the wake word to "Computer" allowing users to activate the system with a unique keyword. Additionally, we have modified the voice feedback to use a male voice instead of the default female voice, providing a more personalized experience for users. The Amazon Eco Dot 3rd gen is an ideal choice for our robotic head as it provides reliable voice recognition and response capabilities, without the need for additional hardware such as a Raspberry Pi or other device. With these modifications, our Voice assistant robot with Vision system using Machine Learning offers an even more advanced and personalized user experience. Some potential uses for a Voice assistant robot with Vision system using Machine Learning include home automation, personal assistance, entertainment, and education.
ABOUT THE ENTRANT
- Name:Kamalesh K
- Type of entry:teamTeam members:
- Dhanasekar J
- Software used for this entry:Thinkercad, Ultimaker Cura, Arduino Programming, Python Programming, Amazon Alexa Service
- Patent status:pending