ESP32-S3 开发实战:从零搭建AIoT项目全流程ESP32-S3 Development in Practice: A Complete Guide to Building an AIoT Project from Scratch


AI free token/AI免费积分

                   

       

ESP32-S3 开发实战:从零搭建AIoT项目全流程

               

在物联网和边缘AI的浪潮中,ESP32-S3 凭借其集成的AI向量指令集、丰富的外设接口和成熟的开发生态,成为众多开发者的首选芯片。本文基于实际项目经验,从硬件选型、环境搭建到实战案例,带你完整走一遍ESP32-S3的AIoT开发流程。文中所有技术参数和代码示例均来自可验证的公开项目,确保每个步骤都可复现。

       

一、ESP32-S3 硬件规格详解

               

在开始开发之前,我们需要先了解ESP32-S3的核心能力。以下是基于乐鑫官方数据和实际项目验证的硬件规格:

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
规格项 参数 说明
处理器 双核 Xtensa 32位 LX7 主频高达 240 MHz,支持动态调频优化功耗
内置 SRAM 512 KB SRAM(静态随机存取存储器),TCM(紧耦合内存)架构
外置 Flash 最大 16 MB 通过 Octal SPI 接口扩展,存储固件和语音模型
外置 PSRAM 最大 8 MB 伪静态RAM,用于运行AI模型和数据处理
GPIO 45 个可编程 支持丰富的外设扩展
无线功能 Wi-Fi + Bluetooth 5 (LE) 2.4 GHz Wi-Fi (802.11 b/g/n),支持 Bluetooth Mesh
AI加速 向量指令集 专为加速神经网络计算和复杂信号处理设计
       
数据来源:ROS小课堂 - 初识ESP32-S3芯片(2026-06-20)、博客园 - ESP32-S3实战指南(2025-01-16)
       

1.1 为什么选择 ESP32-S3?

               

相比前代ESP32,S3版本在AIoT应用场景下的核心优势在于:

               

1. AI向量指令集:新增专为神经网络加速的向量指令,配合ESP-DSP和ESP-NN库,可显著提升图像识别、语音唤醒等AI应用的性能(来源:搜狐 - 乐鑫ESP32-S3芯片AI应用,2024-10-15)。

               

2. 大容量PSRAM支持:最大8MB外置PSRAM,可以运行更大的神经网络模型,而不仅限于简单的TensorFlow Lite Micro示例。

               

3. 成熟的开发生态:支持ESP-IDF、Arduino、MicroPython等多种开发框架,社区活跃,开源项目丰富。

       

二、开发环境搭建(三种方式)

       

ESP32-S3支持多种开发方式,针对不同需求的开发者,这里介绍三种主流方案。

       

2.1 方式一:ESP-IDF(官方推荐,适合商用项目)

       

ESP-IDF是乐鑫官方的物联网开发框架,包含工具链、API和完整文档,适合需要深度定制的商用项目。

       

步骤1:克隆ESP-IDF仓库

       
# 使用国内Gitee镜像加速(推荐)
git clone --recursive https://gitee.com/EspressifSystems/esp-idf.git
cd esp-idf

# 切换到稳定版本(推荐v5.0+)
git checkout v5.3.1

# 安装工具链
./install.sh

# 设置环境变量
source export.sh
       
           💡 国内加速技巧:使用Gitee镜像源可以大幅提升下载速度。同时,将Python的pip软件源切换至阿里云镜像,可加速依赖安装(来源:企鹅号 - 代码全开源!ESP32-S31-TINY快速上手指南,2026-06-11)。        
       

步骤2:配置VS Code开发环境

               

安装VS Code的ESP-IDF扩展插件,配置工具链路径和串口监控参数。对于LCD显示屏等项目,还需要添加LVGL图形库:

       
# 使用IDF组件管理器安装LVGL
idf.py add-dependency "espressif/lvgl^8.3.6"

# 或者手动克隆到项目的components目录
git clone https://github.com/lvgl/lvgl.git components/lvgl
       
代码示例来源:CSDN - ESP32-S3电子相册实战(2026-03-11)
       

2.2 方式二:Arduino IDE(适合快速原型)

       

对于熟悉Arduino生态的开发者,使用Arduino IDE是最快的上手方式。

       

步骤1:添加ESP32板卡支持

               

打开Arduino IDE,进入"文件"→"首选项",在"附加开发板管理器网址"中添加:

       
https://raw.githubusercontent.com/espressif/arduino-esp32/gh-pages/package_esp32_index.json
       

步骤2:安装ESP32开发板支持包

               

打开"工具"→"开发板"→"开发板管理器",搜索"esp32"并安装。安装完成后,选择"ESP32S3 Dev Module"作为目标开发板。

       
操作步骤来源:CSDN - ESP32-S3物联网开发实战:从点灯到上云Adafruit IO(2026-05-14)
       

2.3 方式三:MicroPython(适合Python开发者)

       

如果更熟悉Python语法,可以使用MicroPython快速验证想法。

       

固件刷写命令:

       
esptool.py --chip esp32s3 --port /dev/ttyUSB0 write_flash 0x0 firmware.bin
       

刷写完成后,即可通过串口或WebREPL进行Python交互式开发。

       
命令来源:CSDN - ESP32-S3物联网开发实战:从零搭建AI语音交互设备(2026-04-04)
       

三、实战项目1:本地语音识别与控制

       

语音交互是AIoT设备的核心能力之一。本小节基于实际开源项目,介绍如何在ESP32-S3上实现本地语音识别。

       

3.1 硬件选型

                                                                                                                                                                                                                                                                                                                                                                                                                                                               
组件 推荐型号 说明
ESP32-S3模组 ESP32-S3-WROOM-1-N16R8 16MB Flash + 8MB PSRAM
麦克风阵列 双麦克风 支持波束成形和噪声抑制
音频编解码器 ES8388 或 ES7210 低功耗,支持16位/16kHz采样
音频功放 NS4150B 单声道D类音频放大器
扬声器 4Ω/8Ω 匹配2W Class D功放
       
硬件选型来源:CSDN - ESP32-S3物联网开发实战:AI语音交互设备
       

3.2 软件框架:ESP-SR / MultiNet

       

乐鑫提供了ESP-SR(语音识别)开发框架,其中的MultiNet组件支持本地命令词识别,无需联网即可实现语音控制。

       

项目示例:语音控制Web塔防游戏

               

一个完整的实战项目流程(来源:CSDN - ESP32-S3实战教程:本地语音识别控制Web塔防游戏,2026-05-29):

       
// 1. ESP32-S3采集麦克风音频
// 2. ESP-SR / MultiNet 在本地识别命令词
// 3. 通过WebSocket把命令发送给电脑端的FastAPI服务
// 4. Web页面实时显示游戏状态
       

这个项目不是单纯介绍某一个库,而是提供了一个可以运行的完整闭环,适合学习语音识别的实际集成方式。

       

四、实战项目2:摄像头图像识别与优化

       

ESP32-S3配合PSRAM,可以实现基本的图像采集和本地推理。但在实际开发中,图像采集管线的优化是关键。

       

4.1 帧缓存(Frame Buffer)管理机制

       

在使用PSRAM的摄像头模块中,图像采集管线如下(来源:腾讯云 - ESP32-S3手势识别延迟优化实战,2026-06-24):

       
摄像头采集 → 驱动缓存队列(fb) → PSRAM存储帧 → 算法取帧识别 → 返回结果
       

默认机制下,ESP32摄像头驱动会预分配2~3个帧缓存(frame buffer):

       
camera_config_t config;
config.fb_count = 2;  // 默认2个帧缓存
       

这意味着:摄像头持续采集新帧,算法线程从队列中取出"上一帧"来处理。可以通过打印帧地址来确认:

       
camera_fb_t *fb = esp_camera_fb_get();
Serial.printf("Frame addr: %p\n", fb);
esp_camera_fb_return(fb);
       

4.2 延迟优化技巧

       

方法1:降低缓存数量(fb_count = 1)

               
config.fb_count = 1;  // 关键设置
       

这会强制摄像头只使用单帧缓存。效果:只有在esp_camera_fb_return(fb)后,摄像头才会采集下一帧。

       
           适用场景:手势识别、目标检测等单帧触发类任务;对实时性比帧率更敏感的项目(如游戏交互)。        
       

方法2:主动控制采集节奏

               

如果算法部分较耗时(例如推理模型500ms+),建议采用"同步采集"方式,确保每一帧都得到及时处理。

       

五、实战项目3:LCD显示屏与图形界面

       

对于需要本地显示的项目,ESP32-S3可以驱动多种接口的LCD屏幕,并运行LVGL图形库提供流畅的UI体验。

       

5.1 硬件连接示例(4.3寸屏 + RGB接口)

                                                                                                                                                                                                                                                                                                                                                                                                                                                                               
ESP32-S3 GPIO 功能说明
GPIO12 CLK 像素时钟
GPIO11 HSYNC 行同步信号
GPIO10 VSYNC 场同步信号
GPIO9 DE 数据使能
GPIO0-7 DATA0-7 RGB数据总线
GPIO16 SCL I2C时钟
GPIO17 SDA I2C数据
       
引脚定义来源:CSDN - ESP32-S3电子相册实战:从零搭建4.3寸屏的LVGL界面(2026-03-11)
       

5.2 LVGL图形库移植

       

LVGL(Light and Versatile Graphics Library)是一个开源的图形库,特点包括:

               

○  内存占用小(最低配置仅需16KB RAM)

○  支持多种显示控制器和输入设备

○  提供丰富的控件和动画效果

       

在ESP32-S3上移植LVGL的典型项目结构:

       
esp32_s3_photo_frame/
├── components/
│   ├── lvgl/              # LVGL图形库(v8.3或更高)
│   ├── fatfs/              # FATFS文件系统驱动
│   ├── jpeg/               # JPEG解码库
│   └── spiffs/             # SPIFFS或LittleFS
├── main/
│   ├── main.c
│   └── CMakeLists.txt
└── CMakeLists.txt
       

对于1.14英寸ST7789V RGB LCD屏幕(135×240分辨率,SPI接口),已有完整的开发包提供参考,包括BSP层驱动封装和LVGL事件绑定示例(来源:CSDN - ESP32-S3开发包:16MB闪存8MB PSRAM,LVGL 8.3.11驱动ST7789V彩屏,2026-05-30)。

       

六、网络配置:SoftAP配网实战

       

在物联网设备中,如何让用户方便地将设备接入Wi-Fi网络,是一个常见需求。ESP32-S3的SoftAP(软件接入点)模式提供了优雅的解决方案。

       

6.1 SoftAP工作原理

       

想象以下场景(来源:CSDN - ESP32-S3 SoftAP模式开发实战,2025-12-03):

       
你买了一个基于ESP32-S3的智能温湿度计。第一次使用时,它自动开启名为"SmartSensor_XXXX"的Wi-Fi热点。你用手机连上去,浏览器自动跳转到配置页面,选择你的家庭Wi-Fi并输入密码,点击连接——几秒钟后,设备重启并成功接入家中网络。整个过程完全离线、无需云端中转,安全又高效。
       

这就是SoftAP的魅力所在。

       

6.2 配网流程代码框架

       
// 1. 设备开机,检查是否已保存Wi-Fi信息
if (nvs_get_wifi_config() == SUCCESS) {
    // 2. 有保存的信息,直接尝试连接
    wifi_sta_connect();
} else {
    // 3. 无保存信息,开启SoftAP模式
    wifi_softap_start("SmartSensor_XXXX");
    // 4. 启动HTTP服务器,提供配置页面
    http_server_start();
}

// 5. 用户在网页中提交Wi-Fi信息
// 6. 设备接收到SSID/密码,切换为Station模式尝试联网
// 7. 成功后关闭热点,进入正常工作状态
       

官方提供了完整的一键安装脚本和示例项目,开发者可以直接 clone 并修改适配自己的产品。

       

七、常见问题与解决方案

       

7.1 环境配置问题

       

问题:ESP-IDF扩展配置阶段,版本兼容问题导致环境初始化失败。

               

解决:降低ESP-IDF版本至稳定版(如V1.11.1),重新按照扩展引导流程完成工具链、依赖库的下载与配置(来源:CSDN - ESP32-S3开发板LCD显示屏点亮项目实践报告,2026-03-09)。

       

7.2 编译失败问题

       

问题:ESP32-S31当前编译必须使用Master分支版本的ESP-IDF,版本不正确会导致编译失败或组件不兼容。

               

解决:拉取最新ESP-IDF源码,并切换至Master分支(来源:企鹅号 - 代码全开源!ESP32-S31-TINY快速上手指南,2026-06-11)。

       

7.3 PSRAM缓存优化

       

问题:摄像头采集和算法推理之间的帧缓存队列积压,导致延迟增加。

               

解决:将fb_count设置为1,强制单帧缓存模式;或者采用同步采集方式,确保算法处理完成后再采集下一帧。

       

八、总结与资源推荐

       

本文从硬件规格、开发环境搭建到三个完整的实战项目(语音识别、摄像头优化、LCD图形界面),系统性地介绍了ESP32-S3的AIoT开发流程。所有代码示例和硬件选型均来自可验证的开源项目和官方文档,确保读者可以按图索骥,完整复现。

       

推荐资源

       

○  官方文档:ESP-IDF编程指南 - ESP32-S3

○  GitHub仓库:espressif/esp-idf(官方框架)

○  图形库:lvgl/lvgl(LVGL官方仓库)

○  社区论坛:ESP32官方论坛

○  国内镜像:EspressifSystems - Gitee(加速下载)

       
           💡 实践建议:建议从本文"实战项目1"(本地语音识别)开始,因为该项目的硬件要求相对简单(只需麦克风+扬声器),且ESP-SR框架提供了完整的示例代码。在成功运行第一个项目后,再逐步扩展到摄像头、显示屏等更复杂的应用。        
       

致谢

       

本文的实战案例部分参考了以下开源项目和社区分享(按文中出现顺序):

               

○  腾讯云开发者社区 - ESP32-S3手势识别延迟优化实战

○  CSDN博客 - ESP32-S3电子相册实战系列

○  CSDN博客 - ESP32-S3物联网开发实战系列

○  博客园 - ESP32-S3实战指南(立创实战派S3)

○  企鹅号 - ESP32-S31-TINY快速上手指南

       

感谢所有开源贡献者的无私分享,让嵌入式AI开发的门槛不断降低。

           

希望这篇实战指南能帮助你在ESP32-S3的AIoT开发之路上少走弯路。如果有任何问题,欢迎在评论区交流讨论。

           

— 贾湖新郾

       
   

 

 

ESP32-S3 Development in Action: Building an AIoT Project from Scratch

In the wave of IoT and edge AI, the ESP32-S3 has become the chip of choice for many developers, thanks to its integrated AI vector instruction set, rich peripheral interfaces, and mature development ecosystem. Based on real project experience, this article guides you through the complete AIoT development workflow for the ESP32-S3—from hardware selection and environment setup to hands-on projects. All technical parameters and code examples in this article are sourced from verifiable public projects, ensuring every step is reproducible.

I. ESP32-S3 Hardware Specifications Explained

Before we begin development, we need to understand the core capabilities of the ESP32-S3. Below are the hardware specifications based on Espressif's official data and actual project validation:

Data sources: ROS小课堂 - Getting Started with the ESP32-S3 Chip (2026-06-20), Cnblogs - ESP32-S3 Practical Guide (2025-01-16)

1.1 Why Choose the ESP32-S3?

Compared to its predecessor, the ESP32, the S3 version offers key advantages for AIoT application scenarios:

1. AI Vector Instruction Set: New vector instructions specifically designed for neural network acceleration, working with the ESP-DSP and ESP-NN libraries to significantly improve performance in AI applications such as image recognition and voice wake-up (Source: Sohu - Espressif ESP32-S3 Chip AI Applications, 2024-10-15).

2. Large-Capacity PSRAM Support: Up to 8MB external PSRAM, enabling larger neural network models to run, beyond simple TensorFlow Lite Micro examples.

3. Mature Development Ecosystem: Supports multiple development frameworks including ESP-IDF, Arduino, and MicroPython, with an active community and abundant open-source projects.

II. Development Environment Setup (Three Approaches)

The ESP32-S3 supports multiple development approaches. For developers with different needs, here are three mainstream methods.

2.1 Method One: ESP-IDF (Officially Recommended, Suitable for Commercial Projects)

ESP-IDF is Espressif's official IoT development framework, including toolchains, APIs, and comprehensive documentation, ideal for commercial projects requiring deep customization.

Step 1: Clone the ESP-IDF Repository

💡 Domestic Acceleration Tip: Using the Gitee mirror can significantly speed up downloads. Additionally, switching Python's pip source to the Aliyun mirror can accelerate dependency installation (Source: QQ Hao - Fully Open-Source Code! ESP32-S31-TINY Quick Start Guide, 2026-06-11).

Step 2: Configure the VS Code Development Environment

Install the ESP-IDF extension plugin in VS Code, and configure the toolchain path and serial monitor settings. For projects involving LCD displays, the LVGL graphics library also needs to be added:

Code example source: CSDN - ESP32-S3 Digital Photo Frame Project (2026-03-11)

2.2 Method Two: Arduino IDE (Suitable for Rapid Prototyping)

For developers familiar with the Arduino ecosystem, using the Arduino IDE is the quickest way to get started.

Step 1: Add ESP32 Board Support

Open the Arduino IDE, go to "File" → "Preferences," and add the following URL under "Additional Board Manager URLs":

Step 2: Install the ESP32 Board Support Package

Open "Tools" → "Board" → "Boards Manager," search for "esp32," and install it. After installation, select "ESP32S3 Dev Module" as the target board.

Operation steps source: CSDN - ESP32-S3 IoT Development in Action: From Blinking an LED to Cloud Connectivity with Adafruit IO (2026-05-14)

2.3 Method Three: MicroPython (Suitable for Python Developers)

If you are more comfortable with Python syntax, you can use MicroPython to quickly validate your ideas.

Firmware flashing command:

After flashing, you can proceed with Python interactive development via serial port or WebREPL.

Command source: CSDN - ESP32-S3 IoT Development in Action: Building an AI Voice Interaction Device from Scratch (2026-04-04)

III. Hands-On Project 1: Local Voice Recognition and Control

Voice interaction is one of the core capabilities of AIoT devices. This section, based on actual open-source projects, introduces how to implement local voice recognition on the ESP32-S3.

3.1 Hardware Selection

Hardware selection source: CSDN - ESP32-S3 IoT Development in Action: AI Voice Interaction Device

3.2 Software Framework: ESP-SR / MultiNet

Espressif provides the ESP-SR (Speech Recognition) development framework, whose MultiNet component supports local command word recognition, enabling voice control without an internet connection.

Project Example: Voice-Controlled Web Tower Defense Game

A complete hands-on project walkthrough (Source: CSDN - ESP32-S3 Practical Tutorial: Local Voice Recognition Controlling a Web Tower Defense Game, 2026-05-29):

This project doesn't just introduce a single library—it provides a fully runnable end-to-end solution, making it ideal for learning practical voice recognition integration.

IV. Hands-On Project 2: Camera Image Recognition and Optimization

With PSRAM, the ESP32-S3 can perform basic image capture and local inference. However, in actual development, optimizing the image capture pipeline is the key challenge.

4.1 Frame Buffer Management Mechanism

For camera modules using PSRAM, the image capture pipeline is as follows (Source: Tencent Cloud - ESP32-S3 Gesture Recognition Latency Optimization in Practice, 2026-06-24):

By default, the ESP32 camera driver pre-allocates 2–3 frame buffers:

This means: the camera continuously captures new frames, while the algorithm thread processes the "previous frame" from the queue. You can confirm this by printing frame addresses:

4.2 Latency Optimization Techniques

Method 1: Reduce the Number of Frame Buffers (fb_count = 1)

This forces the camera to use only a single frame buffer. Effect: the camera will only capture the next frame after esp_camera_fb_return(fb) is called.

Suitable scenarios: Single-frame trigger tasks such as gesture recognition and object detection; projects where real-time responsiveness is more important than frame rate (e.g., gaming interaction).

Method 2: Actively Control the Capture Rhythm

If the algorithm portion is relatively time-consuming (e.g., model inference takes 500ms+), it is recommended to use a "synchronous capture" approach to ensure each frame is processed in time.

V. Hands-On Project 3: LCD Display and Graphics Interface

For projects requiring local display, the ESP32-S3 can drive various interface LCD screens and run the LVGL graphics library to deliver smooth UI experiences.

5.1 Hardware Connection Example (4.3-inch Screen + RGB Interface)

Pin definition source: CSDN - ESP32-S3 Digital Photo Frame Project: Building an LVGL Interface for a 4.3-inch Screen from Scratch (2026-03-11)

5.2 LVGL Graphics Library Porting

LVGL (Light and Versatile Graphics Library) is an open-source graphics library with the following features:

○ Low memory footprint (minimum configuration requires only 16KB RAM)

○ Supports multiple display controllers and input devices

○ Provides a rich set of controls and animation effects

Typical project structure for porting LVGL on the ESP32-S3:

For a 1.14-inch ST7789V RGB LCD screen (135×240 resolution, SPI interface), a complete development kit is available as a reference, including BSP layer driver encapsulation and LVGL event binding examples (Source: CSDN - ESP32-S3 Development Kit: 16MB Flash + 8MB PSRAM, LVGL 8.3.11 Driving ST7789V Color Screen, 2026-05-30).

VI. Network Configuration: SoftAP Provisioning in Practice

In IoT devices, how to let users easily connect the device to a Wi-Fi network is a common requirement. The ESP32-S3's SoftAP (Software Access Point) mode provides an elegant solution.

6.1 SoftAP Working Principle

Consider the following scenario (Source: CSDN - ESP32-S3 SoftAP Mode Development in Practice, 2025-12-03):

You've purchased a smart temperature and humidity sensor based on the ESP32-S3. On first use, it automatically creates a Wi-Fi hotspot named "SmartSensor_XXXX." You connect your phone to it, and the browser automatically redirects to a configuration page. You select your home Wi-Fi, enter the password, and click Connect—a few seconds later, the device restarts and successfully joins your home network. The entire process is completely offline with no cloud relay, secure and efficient.

This is the charm of SoftAP.

6.2 Provisioning Flow Code Framework

The official documentation provides a complete one-click setup script and example projects. Developers can directly clone and adapt them to their own products.

VII. Common Issues and Solutions

7.1 Environment Configuration Issues

Problem: During the ESP-IDF extension configuration phase, version compatibility issues cause environment initialization to fail.

Solution: Downgrade the ESP-IDF version to a stable release (e.g., V1.11.1), then follow the extension's guided flow to complete toolchain and dependency installation and configuration (Source: CSDN - ESP32-S3 Development Board LCD Display Lighting Project Practice Report, 2026-03-09).

7.2 Compilation Failure Issues

Problem: Compiling the ESP32-S31 currently requires the Master branch version of ESP-IDF; an incorrect version leads to compilation failure or component incompatibility.

Solution: Pull the latest ESP-IDF source code and switch to the Master branch (Source: QQ Hao - Fully Open-Source Code! ESP32-S31-TINY Quick Start Guide, 2026-06-11).

7.3 PSRAM Cache Optimization

Problem: Frame buffer queue backlog between camera capture and algorithm inference causes increased latency.

Solution: Set fb_count to 1 to force single-frame buffer mode; or adopt a synchronous capture approach to ensure the algorithm finishes processing before the next frame is captured.

VIII. Summary and Resource Recommendations

This article systematically introduces the ESP32-S3 AIoT development workflow—from hardware specifications and environment setup to three complete hands-on projects (voice recognition, camera optimization, and LCD graphics interface). All code examples and hardware selections come from verifiable open-source projects and official documentation, ensuring readers can follow along and reproduce every step.

Recommended Resources

○ Official Documentation: ESP-IDF Programming Guide - ESP32-S3

○ GitHub Repository: espressif/esp-idf (Official Framework)

○ Graphics Library: lvgl/lvgl (LVGL Official Repository)

○ Community Forum: ESP32 Official Forum

○ Domestic Mirror: EspressifSystems - Gitee (for accelerated downloads)

💡 Practical Advice: It is recommended to start with "Hands-On Project 1" (Local Voice Recognition), as this project has relatively simple hardware requirements (only a microphone and speaker needed), and the ESP-SR framework provides complete example code. After successfully running the first project, gradually expand to more complex applications such as cameras and displays.

Acknowledgements

The hands-on project sections of this article reference the following open-source projects and community contributions (in order of appearance):

○ Tencent Cloud Developer Community - ESP32-S3 Gesture Recognition Latency Optimization in Practice

○ CSDN Blog - ESP32-S3 Digital Photo Frame Project Series

○ CSDN Blog - ESP32-S3 IoT Development in Action Series

○ Cnblogs - ESP32-S3 Practical Guide (LiChuang Practical School S3)

○ QQ Hao - ESP32-S31-TINY Quick Start Guide

Thanks to all open-source contributors for their selfless sharing, which continues to lower the barrier to embedded AI development.

I hope this practical guide helps you avoid unnecessary detours on your ESP32-S3 AIoT development journey. If you have any questions, feel free to leave a comment and discuss.

— Jiahu Xinyuan

暂无评论,快来发表第一条评论吧!No comments yet. Be the first!

📮 需求咨询📮 Inquiry