AtomS3R + Atomic Voice Base Voice Assistant

This tutorial describes how to combine AtomS3R with Atomic Voice Base to integrate voice assistant functionality into Home Assistant.

1. Preparation

Environment Setup

1. Refer to the official Home Assistant documentation to install Home Assistant.
2. Refer to the official ESPHome documentation to install the ESPHome Device Builder add-on. (Note: This tutorial is based on ESPHome 2025.1.2. If compilation errors occur, switch to this version for verification.)
3. Install and configure the voice recognition add-ons:

• Home Assistant Cloud.
• Assist Pipeline.

Hardware Products

AtomS3R, Atomic Voice Base

2. Online Flashing

Click the Online Flashing button below and follow the prompts to flash the firmware and configure Wi-Fi, allowing you to quickly connect AtomS3R + Atomic Voice Base to Home Assistant.

For detailed flashing and configuration instructions, refer to Online ESPHome Firmware Flashing and Configuration.

Online Flashing

After using online flashing, you do not need to perform the "Firmware Compilation & Flashing" steps. Proceed directly to 4. Getting Started.

3. Firmware Compilation & Flashing

This section explains how to recompile the project from the yaml configuration file for secondary development, feature modifications, or new functionality.

Create Device

1. Open the ESPHome add-on page and click NEW DEVICE in the bottom right corner to create a new device.

2. Click CONTINUE in the pop-up window.

3. Select New Device Setup to create a new configuration file.

4. Fill in the configuration file name.

5. Select the device type; simply click ESP32-S3.

6. Copy the Home Assistant API Encryption Key for later use and click Skip.

Modify Configuration

1. Click EDIT on the newly generated configuration to modify it. (The Wi-Fi configuration defaults to the current HA server's Wi-Fi configuration, but you can also configure options like ssid and password according to your actual situation).

2. Add the following resource package configuration link to add the voice assistant functional components to the device.

packages:
  m5stack.atoms3r-with-echo-base: github://m5stack/esphome-yaml/common/atoms3r-with-echo-base.yaml@main

3. Click SAVE and INSTALL in the upper right corner. In the installation method pop-up, select Manual download.

Download and Flash Firmware

1. After the firmware compilation is complete, click the Download button and select the Factory Format (Previously Modern) format firmware to download locally.

2. Click Open ESPHome Web to use the web flashing tool for flashing.

3. Assemble the AtomS3R and Atomic Voice Base. Connect the AtomS3R device to your computer via a USB-C cable. Long-press the reset button until the green light turns on and then release it to enter download mode.

4. In Open ESPHome Web, click CONNECT to connect the device and select the corresponding device port.

5. Click INSTALL.

6. Select the *.bin file compiled in the previous step and click the INSTALL button again to begin flashing.

Wait for the flashing process to complete.

4. Getting Started

1. After flashing the firmware, the device will automatically connect to Wi-Fi upon startup. The Home Assistant service on the same local area network will notify you that a new device has been discovered. In Notifications, select the new device and click Check it out -> CONFIGURE. Follow the pop-up steps to add the device to the specified area to complete the configuration. If you do not receive a new device notification, click Settings -> Device & services to view the device status.

2. Next, you can continue to set up the Voice Assistant, or skip this step and set it up later.

   • Test the wake word.
     

   • Select the area.
     

   • Select the Pipeline.
     

   • Complete the setup.
     

Once the device is added and the Home Assistant Cloud and Assist pipeline add-ons are configured in the preparation steps, you can use the wake word "Okay Nabu" to wake up the voice assistant.

Demo video:

5. Video