Intelligent voice chatbot implemented based on esp32-s3-wroom

Video Link:
Bilibili Video -- Function Demonstration and
Project Introduction
This project is an intelligent voice chatbot implemented based on esp32-s3-wroom. By configuring prompts and voice timbres, you can experience various characters and voices. See Xia Ge's Bilibili videos for examples:


Funny Potato with a Northeastern Accent: https://www.bilibili.com/video/BV1uR26YHEcu/?spm_id_from=333.999.0.0


Flirting with Your Taiwanese-Speaking Ex-Girlfriend: https://www.bilibili.com/video/BV1G4pweCERX/?spm_id_from=333.999.0.0&vd_source=d54d1251b3a2da5a425913a5f429be57


English Teacher: https://www.bilibili.com/video/BV18zsDe1ENN/?spm_id_from=333.999.0.0


The project's functionality
is based on Xia Ge's server, achieving a low-latency, high-quality voice chat experience. The server-side console: https://xiaozhi.me/ allows you to define the AI's self-awareness through command words. It has voiceprint recognition capabilities and supports speaker configuration. It also has memory capabilities.
The project's firmware
uses Xiaozhi's open-source code: https://github.com/78/xiaozhi-esp32
, only the pinouts are different. You need to modify the pin definitions to build the firmware: Modify the SDK configuration:
CONFIG_AUDIO_DEVICE_I2S_MIC_GPIO_WS=5
CONFIG_AUDIO_DEVICE_I2S_MIC_GPIO_BCLK=4 CONFIG_AUDIO_DEVICE_I2S_MIC_GPIO_DIN=6
CONFIG_AUDIO_DEVICE_I2S_SPK_GPIO_DOUT =15 CONFIG_AUDIO_DEVICE_I2S_SPK_GPIO_BCLK =16 CONFIG_AUDIO_DEVICE_I2S_SPK_GPIO_WS=17 Alternatively, use the SDK configuration editor (recommended): WS2612 LED pinout: # Builtin LED # CONFIG_BUILTIN_LED_GPIO=45 # end of Built-in LED or modify using the SDK configuration editor (recommended): Xiaozhi AI Chatbot Encyclopedia: https://ccnphfhqs21z.feishu.cn/wiki/F5krwD16viZoF0kKkvDcrZNYnhb Xiaozhi AI Backend Configuration Panel: https://xiaozhi.me/ ( QQ group communication : 946599635)