· Desktop Robot Development Practical Course·
Exploring the Application Potential of Edge AI and Large Language Models
01
Course Background
With the acceleration of digital transformation across various industries, edge AI and cloud-based large model technologies are leading a new wave of innovation in smart devices. This year, the Ministry of Industry and Information Technology published a notice on organizing the collection of typical cases for the development of the Internet of Things in 2024, focusing on the integrated application of new-generation information technologies such as the Internet of Things, big data, cloud computing, and artificial intelligence to promote the development of related industries.
Huazhong University of Science and Technology’s Maker Space and the School of Telecommunications, in collaboration with ESP32-S3 chip, have specially launched a Desktop Robot Development Practical Course to help students keep up with industry trends, experience the charm of these two cutting-edge technologies, and accumulate professional skills.
02
Workshop Content
03
Specific Arrangements
# |
Project |
Main Content |
Practical Task |
Time |
1 |
Project Introduction |
1) About Espressif Technology 2) Introduction to the development platform based on the ESP32-S3 chip 3) Introduction to the development trends of the Internet of Things 4) Large model interaction Demo display |
/ |
20 minutes |
2 |
GUI Development |
1) Introduction to SquareLine Studio 2) Generate UI folder 3) Initialize screen 4) Load UI |
Use SquareLine Studio to design a simple UI interface, export the UI folder, and successfully light up the screen. |
20 minutes lecture + 40 minutes practice |
3 |
Introduction to Edge Computing |
1) Basic principles of audio encoding and decoding Codec 2) Key features of Espressif’s acoustic front-end algorithm AFE a) Voice Activation Detection (VAD) b) Local voice wake-up c) Local voice recognition |
Implement the voice activation detection function and use local voice recognition to create custom functions on the development platform. |
20 minutes lecture + 40 minutes practice |
4 |
Connecting Large Language Models for Interaction |
1) Basic concepts of large language models and human-computer interaction applications 2) Understand the development documentation 3) Interact with the cloud via HTTPS 4) Coordinate multitasking in FreeRTOS |
Complete HTTPS configuration, connect to the large language model, and achieve simple conversational functionality; free to expand. |
40 minutes lecture + 80 minutes practice |
04
Preparation Before Class
Please bring your personal computer to class and complete the following tasks in advance (please copy the link into your browser to open):
1) Set up ESP-IDF v5.2.3 development environment
– One-click installation tool download link
https://dl.espressif.com/dl/esp-idf/?idf=5.2.3
– Video tutorial
https://www.bilibili.com/video/BV1to4y177ko/
2) Download UI production tool SquareLine Studio
https://squareline.io/downloads#lastRelease
3) Watch tutorial video on the ESP-IDF project structure
https://www.bilibili.com/video/BV18keGeaEhE/?vd_source=a93a2546836b9e63ec9e2f2458ac7b4e
4) (Optional) Download HTTP debugging tool POSTMAN
https://www.postman.com/downloads/
5) (Optional) Watch tutorial video on ESP-IDF VSCode development
https://www.bilibili.com/video/BV1X34y1M7L8/?vd_source=a93a2546836b9e63ec9e2f2458ac7b4e
6) (Optional) Watch tutorial video on ESP32 voice recognition tutorial【ESP-SR】
https://www.bilibili.com/video/BV1vf421z7r8/
05
Time and Place
06
Registration Method
07
About Espressif
Espressif Technology (688018.SH) is one of the first batch of listed companies on the Science and Technology Innovation Board in China, dedicated to the integrated innovation of Internet of Things (IoT) and artificial intelligence (AI) technologies, focusing on the research and development of high-performance, low-power wireless communication chips, with a global R&D network and a rich accumulation of core self-developed technologies, including Wi-Fi/Bluetooth LE network protocol stack, RF technology, RISC-V CPU, audio and video codec, AI vector instructions and AI algorithms, operating systems, toolchains, compilers, AIOT software development frameworks, cloud services, etc., achieving a closed loop of software and hardware R&D.

Editor: Chen Xuyang
Editor-in-chief: Zheng Yuyang
Review: Guo Jing, Pang Xiumei
