Skip to content

CI-G16XGS02T Module Datasheet

Module Introduction

Overview

This module is a universal, portable, low-power, high-performance voice recognition module with BLE functionality, developed for cost-effective offline voice applications. Model: CI-G16XGS02T, featuring CI23161/CI23162 as the main chip, supports up to 300 offline voice commands (specific number varies by model).

Module Block Diagram

Figure 1 Module Block Diagram

The CI-G16XGS02T module features:

  • Excellent recognition performance: Supports offline neural network computation, single-microphone noise reduction enhancement, 360° omnidirectional sound pickup, and environmental noise suppression for accurate voice recognition in noisy environments. Offline voice recognition with this module is network-independent, with low latency, high performance, over 97% recognition rate, up to 10m recognition range, and response time as fast as 0.2s.
  • BLE support: The main chip supports Bluetooth BLE, allowing users to send control protocols via WeChat Mini Programs on mobile phones.
  • Compact design: Module dimensions are 22.1mm (L) × 16.2mm (W) × 3.2mm (H, including shield). Features a gold finger design for easy soldering.
  • Minimal components: The complete solution only requires an audio amplifier chip and a few passive components in addition to the voice recognition chip.
  • Simple interfaces: One microphone input, one speaker output, one 5V power supply, and one 5V-level UART communication interface.
  • Wide operating voltage: 3.6V-5V.
  • Low power consumption: Suitable for products with high energy efficiency requirements or battery-powered devices.
  • High reliability: All BOM components are industrial-grade.
  • Easy model selection: Choose between CI23161 or CI23162 chips based on the required number of voice commands, with differences as follows:

Table 1 Module Selection Guide

Module Type Local Commands ≤100 Local Commands ≤300
Gold Finger Offline Voice BLE Module CI-G161GS02T CI-G162GS02T

Main Chip Introduction

CI23161 and CI23162 are dedicated AI chips for voice processing, supporting local voice recognition in multiple global languages including Chinese, English, and Japanese. They are widely used in home appliances, lighting, toys, wearables, industrial, and automotive applications for voice interaction, control, and various intelligent voice solutions.

CI23161 and CI23162 integrate Chipintelli’s self-developed Brain Neural Processing Unit (BNPU V3.5) and CPU core, with a system frequency up to 210MHz. They feature 288KB of built-in SRAM, integrated PMU power management unit, RC oscillator, single-channel high-performance low-power Audio Codec, and multiple peripheral control interfaces including UART, I2C, PWM, and GPIO, along with high-performance low-power Bluetooth BLE. The chips require minimal external components (resistors and capacitors) to implement various intelligent voice product hardware solutions, offering excellent cost performance.

For more detailed information about CI23161 and CI23162 chips, please visit:

CI23161&CI23162 Chip Datasheet

Module Application Scenarios

This module can be used as a voice recognition frontend combined with a customer’s main control board, or as a single-chip main control module for applications such as lighting and toys. It requires external connection of a microphone, speaker, and 5V power supply. Module as Voice Recognition Frontend

Figure 2 Module as Voice Recognition Frontend

The CI-G16XGS02T series modules support 100-300 offline voice recognition commands (depending on the chip model) and can be applied to various end products such as smart fans, heating tables, clothes dryers, small household appliances, toys, and lighting systems.

Module Applications

Figure 3 Module Applications

Module Specifications

Physical Layout

Module Physical Appearance

Figure 4 Module Physical Appearance

As shown in Figure 4, the voice recognition module features single-sided SMT assembly, with main ICs including the voice recognition chip and audio power amplifier. Voice commands are input through the microphone, processed by the voice recognition IC, and the feedback announcement is sent to the audio amplifier to drive the speaker. The audio amplifier has a maximum output power of 1.5W@8Ω and 2W@4Ω.

Module Dimensions

Module Dimensions

Figure 5 Module Dimensions

Module Hardware Interface Definition

Pin functions are defined from left to right as shown in Table 2:

Table 2 Module Pin Definition Table

Pin No. Pin Name I/O Type IO Drive Capability IO Default State Function Definition
1 5V P - - 5V Power
2 GND P - - Ground
3 UART_TX0 IO, T+U 4mA IN, T+U 1. UART0_TX0
2. PB5
4 UART_RX0 IO, T+U 4mA IN, T+U 1. UART0_RX0
2. PB6
5 MICL+ - - - Microphone Positive
6 GND - - - Ground
7 SKPL- - - - Speaker Out
8 SKPL+ - - - Speaker Out

The symbols in the table above are defined as follows:

I - Input

O - Output

IO - Bidirectional

P - Power or ground

T+D - Tristate with pull-down

T+U - Tristate with pull-up

OUT - Defaults to output mode on power-up

IN - Defaults to input mode on power-up

Module Electrical Characteristics

Table 3 Electrical Characteristics

Parameter Condition Min Typ Max Unit Note
Module Supply Voltage / 3.6 5 5.5 V NOTE1
Module Current in Playback (Normal Volume) 4Ω 3W Speaker / 100 / mA NOTE2
Module Operating Current / / 55 / mA NOTE3
Chip I/O Interface Voltage / 3 3.3 3.6 V /
Module UART Interface Voltage / 4.5 5 5.5 V /

Note

NOTE1: 5V is the typical supply voltage. Voltages above 5.5V may damage the module.

NOTE2: The module can draw up to 250mA during audio playback. A power supply capable of 500mA is recommended.

NOTE3: Typical value is measured in mute state. Maximum value is measured during recognition and audio playback.

Bluetooth Electrical Characteristics

Table 4 Wireless Characteristics

Parameter Symbol Description Min Typ Max Unit
Sleep Power I_SLEEP VDDRF=3.3V - 6 - μA
Current in TX 0dBm I_TX VDDRF=3.3V - 2.5 - mA
Current in RX 1Mbps BLE I_RX VDDRF=3.3V @ -98 dBm sensitivity - 2.8 - mA
Frequency range Freq 2400 2483.5 MHz
Output power Pout -20 5 dBm

Above power consumption data is measured when VDDRF equals 3.3V

Module Temperature and Humidity Parameters

Temperature and humidity parameters for CI-G16XGS02T are shown in Table 5.

Table 5 Temperature and Humidity Parameters

Parameter Min Typ Max Unit Note
Operating Temperature -40 25 85 °C /
Storage Temperature -40 25 100 °C /
Storage Humidity 0% / 5% RH /

Module Application

Module Power Supply and Startup

The module’s voice chip and audio amplifier chip are powered by 5V. The 5V power supply must ensure a rated current of 500mA, with stable power supply and a ripple of less than 300mV.

Module startup reference print information

Figure 6 Module startup reference print information

Default Module Commands

If the module is mass-produced for users, it is generally flashed with user-specified command entry firmware before shipment. If the customer does not specify, the module will come with a default firmware that includes default commands for testing. The partial commands are shown in Figure 7:

Default module commands

Figure 7 Default module commands

Default UART Communication Protocol

The module’s default firmware supports a UART communication protocol for MCU communication. This protocol is extensible with the following features:

  • Complete transmission packet including: header/tail, length, checksum, message type, and sequence number.
  • Supports variable-length commands for easy expansion.
  • Message types (command, notification, response).
  • Command messages are configurable with ACK responses. Notification messages don’t require ACK.
  • The message format is the same as the bootloader upgrade protocol, distinguished by the header.
  • Default baud rate is 9600 bps.
  • Note: The module only reserves the UART0 interface, which defaults to print output. To use UART0 as a protocol interface, code modifications are required. Refer to the UART protocol documentation in the ☞CI13LC Series Chip SDK for implementation details.
  • Supported commands: Query protocol version, query system version, set volume (volume levels defined in user_config.h), play local audio prompts, reset command, etc. The protocol format is shown below:

UART Protocol Format

Figure 8 UART Protocol Format

Example 1:

A5 FC 07 00 A0 91 18 01 55 E0 01 00 00 1B 9B 02 FB is parsed as follows:

A5 FC: Header

07 00: 7 bytes of valid data

A0: Command word information

91: Command number 0x91 (this data contains command word data)

18: Packet sequence number (0x08th transmission), increments with each transmission

01 55 E0 01 00 00: Unique data for the current command word

1B: Command word threshold

9B 02: Checksum

FB: End marker

Note

If only the command word and threshold are of interest, focus on the 7 bytes of valid data highlighted in blue.

Example 2:

A5 FC 02 00 A3 9A 17 00 B1 05 02 FB is parsed as:

A5 FC: Header

02 00: 2 bytes of valid data

A3: Notification data

9A: Command number 0x9A (voice module content change)

17: Packet sequence number (0x07th transmission), increments with each transmission

00 B1: Valid data (indicates entering wake-up state)

05 02: Checksum

FB: End marker

Note

This is notification data. Users can choose whether to use this information based on their needs.

For more detailed protocol parsing, refer to the UART protocol section in the ☞CI13LC Series Chip SDK. Below is a reference screenshot of protocol data:

UART Protocol Data Reference

Figure 9 UART Protocol Data Reference

Software Development

The default firmware of the module is mainly used for users to experience the product. If users want to develop software, they need to register and log in to the Chipintelli AI Speech Development Platform (https://aiplatform.Chipintelli.com), and develop voice firmware quickly. At the same time, in the “Development Resources” section of the Chipintelli AI Speech Development Platform, SDK and related hardware resources can be downloaded. For new users, it is recommended to first use the Beginner’s guide to understand the specific development process, and also refer to the video tutorial in the document center for more solutions and SDK development.

  • SDK development package download
  • Model development (language model + acoustic model)
  • Voice prompt synthesis
  • Command word information table and audio file association
  • Firmware packaging

For more detailed development process, please click ☞CI13LC series chip SDK to learn.

Firmware Flashing

Flashing Preparation

Before flashing the module, the following items need to be prepared:

  • Module to be flashed
  • USB to serial debug tool
  • Firmware flashing tool (pack_update_tool.exe)
  • Firmware information (*.bin formatted file)
  • 2.0mm spacing microphone
  • 2.5mm spacing speaker
  • DuPont wire

Firmware flashing items

Figure 10 Firmware flashing items

Hardware Connection and Flashing

  • Hardware connection: Take the USB to serial debug tool shown in the figure as an example. Before flashing, connect the power, ground, and serial port receive and transmit pins of the USB to serial debug tool to the corresponding pins of the module. Note that the RXD and TXD of the USB to serial debug tool correspond to the UART0_TX and UART0_RX of the module, respectively. When flashing, the PGEN pin of the module needs to be connected to 3.3V through a 10KΩ resistor, and the connection should be disconnected after flashing. The connection method is shown in the figure below.

Module and USB to serial debug tool connection diagram

Figure 11 Module and USB to serial debug tool connection diagram
  • Firmware flashing: Open the firmware flashing tool (this tool can be found in the CI231X_SDK\tools directory of the SDK development package, PACK_UPDATE_TOOL.exe), select the corresponding model according to the chip, click the Firmware Update button, select the firmware file that has been made, and confirm the serial port port number allocated by the computer to the USB to serial debug tool. After the preparations are ready, turn on the power switch to automatically enter the programming mode (PA4 internal pull-up), and start downloading firmware. After the download is completed, the progress bar displays 100%.

Firmware flashing tool diagram

Figure 12 Firmware flashing tool diagram

Function Test

  1. Voice function test: After firmware flashing is completed, it is recommended to test the module’s functionality to verify that the flashed firmware is successful. Before testing, the module to be tested needs to first connect the microphone and speaker, turn on the power to confirm whether there is an on-power announcement, and use the wake-up word and command word to test whether it can be awakened and recognized normally. If both can work normally, then the module function is normal, and the flashing is successful; otherwise, the flashing fails, and further investigation is needed.

  2. Bluetooth function test:

  3. Search for the “Chipintelli IoT” or “AI Voice Control” mini-program in WeChat

  4. Enable Bluetooth permissions (some phones may require location permissions)

  5. Turn on the phone’s Bluetooth

  6. Add a device by clicking the “+” button or the “Add Device” button

  7. Enter the device scanning interface and click on the detected device** (Currently supported devices in this mini-program include: tea makers, air conditioners, lights, fans…)

  8. Confirm device connection

  9. Name the device

  10. Return to the home page after successful connection

  11. Click on the added device to enter the device operation interface

Mini program User Guide

Figure 13 Mini program User Guide

Troubleshooting

This section lists some problems that may occur when using the module and their corresponding solutions.

  1. Module cannot flash and update firmware. Please follow the steps below for troubleshooting:
  • First, ensure TX, RX, and GND are properly connected, pull up the PGEN pin, then select the correct COM port in the flashing tool (Figure 12) before applying 5V power.

  • Verify the UART connections: Check if TX and RX are correctly connected (not swapped), ensure the USB-to-UART driver is properly installed on your PC, and confirm the correct COM port is selected in the flashing tool.

  • If the above checks are correct but the module still cannot be programmed, use a multimeter to measure the module’s power supply voltages (5V, 3.3V, and 1.1V) at the test points shown in the figure below. If any voltage readings are incorrect, there may be a hardware fault with the module. In this case, please replace the module or repair the hardware. If all checks are normal but the issue persists, please contact our technical support for assistance.

Module test point distribution diagram

Figure 14 Module test point distribution diagram
  1. Module programmed successfully but does not broadcast. Please follow the steps below for troubleshooting:
  • Confirm that the flashed firmware matches the board;

  • Confirm that the speaker is correctly connected and powered;

  • Use a oscilloscope to measure the main chip’s voice output test point. If there is no output, check if the firmware is correct. If there is an output, check if the audio amplifier on the module is welded abnormally. If the audio amplifier is abnormal, replace it and test again. The measurement points are shown in the figure below. If all checks are normal but the issue persists, please contact our technical support for assistance.

Module voice output measurement point

Figure 15 Module voice output measurement point
  1. Module programmed successfully but does not recognize commands. Please follow the steps below for troubleshooting:
  • Check if the microphone and connector are properly connected;
  • Check if the microphone is correctly connected and powered;
  • Use a oscilloscope to measure the main chip’s MICBIAS test point. If there is no output, check if the firmware is correct. If there is an output, check if the audio amplifier on the module is welded abnormally. If the audio amplifier is abnormal, replace it and test again. The measurement points are shown in the figure below. If all checks are normal but the issue persists, please contact our technical support for assistance.

Module microphone signal test point

Figure 16 Module microphone signal test point

Other Application Notes

  • CI23161、CI23162 chips have a high ESD level and the module is designed to be convenient for users to expand, so the module only designs ESD devices on the microphone, and for products with high ESD requirements, ESD devices can be added. It is recommended to wear anti-static wristbands or anti-static gloves, fingerless gloves, etc. during inspection and production processes. Please reserve ESD protection devices at the corresponding bottom board connector positions to ensure the quality and reliability of the products.

  • The microphone, speaker, power, and serial port interfaces of the module cannot be connected incorrectly.

  • If you need to use a USB to serial debug tool to debug the module, you need to first add serial port print commands in the corresponding positions of the SDK program, compile to generate firmware and flash, and then you can debug and verify.

  • When designing the module bottom board or MCU main board, the module 5V power input port needs to be configured with a capacitance of no less than 100uF, and the microphone signal line should be short and shielded. The SPK signal line should be short and thick, and there should be no other lines across the PCB surface in the walking area and the walking area.

  • The control bottom board should have a bending degree of no more than 0.5%, to prevent module welding failure.

Production Guide, Storage and Packaging Ordering Information

Production Storage Guide

1.CI-G16XGS02T module storage conditions are as follows:

  • The module is stored in a standard vacuum moisture-proof bag, which can be stored at a temperature of -40℃-+100℃ and a humidity of 0%-85% RH.

  • The humidity indicator card inside the vacuum moisture-proof bag is as follows:

Humidity indicator card

Figure 17 Humidity indicator card

2.If the humidity indicator card has the following color change, the corresponding baking parameters need to be baked:

  • If the humidity indicator card reads that the 30% color ring is blue, the module needs to be baked for 2 hours.

  • If the humidity indicator card reads that the 30% color ring is pink, the module needs to be baked for 4 hours.

  • If the humidity indicator card reads that the 30% and 40% color rings are pink, the module needs to be baked for 6 hours.

  • If the humidity indicator card reads that the 30%, 40%, and 50% color rings are pink, the module needs to be baked for 12 hours.

3.Baking parameters are as follows:

  • Baking temperature: 125±5℃

  • Alarm temperature setting: 130℃

  • Baking times: 1 time

  • After natural cooling, the module temperature should be less than 36℃ before SMT patching can be performed.

  • If the module has not been welded after baking for more than 12 hours, it needs to be baked again.

  1. The entire SMT patching process needs to be protected against ESD, and production operations need to wear anti-static gloves.

  2. To ensure production yield, all modules should be visually inspected, AOI detected, and the temperature control of the furnace, the correct adsorption method of the module, and the correct placement method should be paid attention to.

Module mounting reflow profile

Figure 18 Module mounting reflow profile

Packaging and Ordering Information

Table 6 Module Ordering Information

Product Model Packaging Method Modules per tray Modules per tray group Modules per box
CI-G161GS02T
CI-G162GS02T
Tray+static bag+paper box 99 pcs/tray 10 trays, 9900pcs in total 5 bags, 49500pcs in total

Purchase and Technical Support

If you want to purchase our product samples, please click ☞Sample Purchase .

If you want to get technical support, please log in to ☞Chipintelli AI Speech Development Platform .