LLM Module UIFlow2.0 Quick Start

Overview

The LLM Module can be used with various M5 controllers. This tutorial will use the M5Core series as an example to demonstrate how to program and control the LLM Module in UIFlow 2.0.

Preparation

• Refer to the UIFlow Web IDE Tutorial to understand the basic workflow of using UIFlow 2.0, and follow the M5Core Firmware Flashing and Program Uploading Tutorial to complete the device's UIFlow firmware flashing and connection to UIFlow 2.

Example

Here are some basic example programs. You can download and import them into UIFlow 2 according to your needs.

• examples:
  • get_model_list : Gets a list of models for the Module LLM
  • kws_asr : Wake word detection through KWS -> Trigger ASR for speech-to-text conversion. (KWS + ASR)
  • text_assistant : Input content into the LLM model via text, and output the inference results as text. (LLM)
  • tts : Convert text to speech playback using the TTS module. (TTS)
  • voice_assistant : Wake word detection through KWS -> Trigger ASR for speech-to-text conversion -> Use the converted content as input for LLM inference -> Finally, output the inference results as speech through TTS. (KWS + ASR + LLM + TTS)
  • yolo11n with UVC :Read the video stream through the USB camera, send the video stream to yolo11, deduce the output recognition result.

Related Links

• LLM Module UIFlow2.0 API

Video