Module ASR — Custom Firmware Generation and Flashing

Based on the actual command words and multi-language requirements used by the customer, the Module ASR offline speech recognition module can flexibly and efficiently adjust command words and response phrases via the SmartPi online platform, and regenerate matching firmware.

1. Preparation

Account registration: Visit the SmartPi platform to register and log in, then open the Product Management page.
Download the product configuration template (.json): The links below include both Chinese and English versions of the configuration template. Choose the template that matches the language you will use; you can further customize the firmware based on the template.

Download Module ASR Config Template Json File

Language support

The current version does not support mixing languages — only one language can be used at a time. If you need different languages, download the corresponding configuration template for each language or regenerate via the platform.

Hardware:

Module ASR only needs to be connected to a computer via a USB Type-C data cable for flashing.

Software tools used:

Module ASR flashing tool

2. Configuration Template

Click Import Product and import the product configuration template (.json) downloaded in the previous step. This tutorial uses the English template as an example.

Click View Details, select the template and click Inherit. This will create a new version based on the factory firmware template for editing.

Click the Edit button to enter the firmware configuration page. Scroll to the Command Words Customization option and click + Add to create a new command.

Fill in the trigger phrase and the reply phrase according to the language you will use. Do not mix languages; configure according to the language you enable. When finished, click the arrow button (>>) and then click + Add Control to configure detailed parameters for the current command.

Wake-word settings

By default the template uses several phrases from the normal command list as no-wake-word triggers (e.g. "Hi M Five", "Hi ASR") for wake interactions. This option conflicts with the custom wake word section; if both are enabled, only the custom wake word will be triggered. To avoid conflicts, simply select commands from the command list as no-wake-word triggers and do not set a separate custom wake word.

Note

If you want other trigger types (for example, UART input, GPIO input, etc.), change the trigger type in the Command Words section shown above.

Configure the UART response payload that the Module ASR will send when a voice command is triggered. To be compatible with the M5Unit-ASR Arduino library, follow the existing format. The format rule is AA 55 ID 55 AA, where the command ID must be unique and must not conflict with other commands.

After completing the configuration above you have added the new command Add. Next, click the Configuration Check option in the menu to validate for errors. If everything is correct, click Publish Version to package the firmware.

The version publishing process takes a few minutes. Once the firmware is generated, click Download Firmware. The firmware archive is named jx_firm.tar.gz.

3. Flashing the Firmware

Open the Module ASR flashing tool (.exe) downloaded in previous steps. Ensure the product model is set to CI1302, then click Update to start the tool.

Import the extracted firmware file (.bin).

Flashing procedure

The Module ASR flashing port is a USB Type-C connector — connect the module to your computer using this port.

Once the flashing tool detects the correct COM port, check the port to activate download mode, then press the Debug Rst button to start downloading the program.

Program download complete.

Note

After flashing completes, please uncheck the port option in the flashing tool; otherwise the software will continuously attempt to re-flash the firmware.

4. Command Test

Power the Module ASR module and use the wake phrase Hi M Five. If it responds with I'm here, the firmware flashing was successful.

Next Overview

Linux PC

CM4Stack

CoreMP135

AI Accelerator Card

LLM-8850 Card

Quick Start

Vision Models

Large Language Models

Multimodal Models

Audio Models

Generative Models

Application List

Advanced Usage

LLM

Real-Time AI Voice Assistant

OpenAI Voice Assistant

XiaoZhi Voice Assistant

XiaoLing Voice Assistant

AtomS3R-M12 Volcengine Kit

Offline Voice Recognition

Unit ASR

Module ASR

Industrial Control

StamPLC

IoT Measuring Instruments

Air Quality

PowerHub

Module13.2 PPS

VAMeter

T-Lite

Input Device

Ezdata

Ethernet Camera

PoECAM

Wi-Fi Camera

TimerCAM

Unit CamS3/-5MP

AI Camera

UnitV2

M5StickV/UnitV

LoRa & LoRaWAN

TTN (The Things Network)

Meshtastic

Motor Control

Unit Roller485/CAN

Develop Tools