CI-D1XGS10T Module Datasheet¶

Module Introduction¶

Overview¶

This module is a general-purpose, portable, low-power, high-performance voice recognition module developed for low-cost offline voice applications. The model is CI-D1XGS10T, with CI1311 and CI1312 as the main chips, supporting recognition of up to 300 offline command words (the supported number varies by model).

Figure 1 Module Block Diagram

This module has the following features:

Compact size of 20.5 × 13.8 mm; operating voltage 3.6 V–5 V. The board integrates a power amplifier, with one microphone interface, one speaker interface, and one 5 V power and UART interface. Typically powered by the product main board’s 5 V supply with UART communication; simply add microphone and speaker connectors on the main board. The module uses a gold-finger process with vertical mounting, greatly saving main-board PCB space.

The main chip supports offline neural network computation, single-microphone noise reduction enhancement, single-microphone echo cancellation, and 360° omnidirectional pickup, suppressing environmental noise to ensure recognition accuracy in noisy environments. Offline recognition does not rely on a network, with low latency and high performance, achieving >97% recognition rate, up to 10 m recognition distance, and response time as fast as 0.2 s.
Applicable to products with energy-efficiency requirements and battery-powered products.
High reliability with industrial-grade components.

Table 1 Module Options

Module	Up to 100 local commands	Up to 300 local commands
Gold-finger offline voice module	CI-D11GS10T	CI-D12GS10T

Main Chip Introduction¶

CI1311 and CI1312 are AI chips specifically designed for voice processing. They support local voice recognition and multiple global languages including Chinese, English, and Japanese. They can be widely used in home appliances, lighting, toys, wearables, industrial, and automotive products to enable voice interaction, control, and various intelligent voice solutions.

CI1311/CI1312 integrate chipintelli’s self-developed BNPU V3 neural network processor and a CPU core; system frequency up to 220 MHz; up to 640 KByte on-chip SRAM; integrated PMU and RC oscillator; dual-channel high-performance, low-power Audio Codec; and multiple peripherals including UART, I2C, I2S, PWM, GPIO, and PDM. Only a few external components are required to realize various intelligent voice hardware solutions with excellent cost performance.

For more detailed information about CI1311 & CI1312, click the link below:

☞CI1311 & CI1312 Chip Datasheet

Module Application Scenarios¶

This module can be used as a standalone voice recognition module combined with the customer’s main control board, or as a stand-alone main control module for lighting, toys, and other solutions. Connect an external microphone and speaker, and supply power via an external 5 V source.

Figure 2 Module as Voice Recognition Front-end

The CI-D1XGS10T module supports up to 300 offline command words and can be applied to smart air conditioners, smart fans, heating tables, clothes dryers, small appliances, toys, lighting, and more.

Figure 3 Applicable Products

Module Specifications¶

Module Photos¶

Figure 4 Module Photo and Chip Location

As shown in Figure 4, the voice recognition module is vertically mounted. The main ICs include the voice recognition chip (CI1311 or CI1312) and a power amplifier. Voice commands are input from the microphone, recognized and processed by the voice recognition IC, and feedback audio is sent to the power amplifier to drive the speaker. The amplifier’s maximum output power is 1.5 W @ 8 Ω and 2 W @ 4 Ω.

Module Dimensions¶

Figure 5 Module Dimensions

As shown in Figure 5, users can design their mechanical structure based on these dimensions.

Hardware Interface Definition¶

Figure 6 Module Pin Diagram

This module provides the following functional interfaces:

Two-wire single-microphone interface. Design a microphone socket or pads on the baseboard and add ESD devices on the microphone traces. To ensure good recognition performance, it is recommended to use a microphone with sensitivity −32 ± 3 dB and SNR ≥ 65 dB. Click ☞Reference Microphone Devices for more information.
Two-wire single-speaker interface. Design a speaker socket or pads on the baseboard. For better playback, it is recommended to use a speaker with an enclosure. Click ☞Reference Speaker Devices for more information.
The UART0 interface can be used for module firmware updates. Please design corresponding header pins on the baseboard to facilitate subsequent upgrades. The UART pins can also be configured as GPIOs in addition to serial communication.

Descriptions of all external pins are shown in Table 2:

Table 2 Module Pin Mapping

Pin Name	Type	IO 5V Tolerance	IO Default State at Power-on	Function Definition
5V	P	-	-	5V power
GND	P	-	-	Ground
TX1	IO	√	IN, T+D	1. GPIO PA2 2.UART1_TX 3.I2C_SDA 4.PWM Channel 0
RX1	IO	√	IN, T+D	1. GPIO PA3 2.UART1_RX 3.I2C_SCL 4.PWM Channel 1
MIC+	-	-	-	Microphone positive
MIC-	-	-	-	Microphone negative
SKP-	-	-	-	Speaker output
SKP+	-	-	-	Speaker output

Descriptions of the symbols in the table above are as follows:

I input

O output

IO bidirectional

P power or ground

T+D tristate plus pull-down

T+U tristate plus pull-up

OUT power-on defaults to output mode

Electrical Characteristics¶

Table 3 Electrical Characteristics

Parameter	Condition	Min	Typ	Max	Unit	Remark
Module supply voltage	/	3.6	5	5.5	V	NOTE1
Module current during playback (normal volume)	8Ω 2W speaker	/	70	/	mA	NOTE2
Module operating current	/	/	40	/	mA	NOTE3
Listening current in quiet environment	5V supply	/	35	/	mA	/
Module IO interface voltage	/	3	3.3	5	V	/

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

NOTE2: The maximum current during playback can reach 250 mA. Following the 2× margin principle, a 500 mA power supply capability should be provided.

NOTE3: The typical value is measured in mute state. The maximum value is measured during active recognition and playback.

Temperature and Humidity Parameters¶

Table 4 Temperature and Humidity Parameters

Parameter	Min	Typ	Max	Unit	Remark
Operating ambient temperature	-40	25	85	°C	/
Storage ambient temperature	-40	25	100	°C	/
Storage humidity	0%	/	5%	RH	/

Module Application¶

Baseboard Design Reference¶

Using this module requires designing a debug baseboard or host mainboard. The main purpose of the debug baseboard is to carry this module, provide power, place microphone and speaker sockets, and implement communication circuits with the main controller. 1. It is recommended to place a larger-value filter capacitor at the module power input to ensure stability of the 5 V input. If the 5 V input ripple is small and noise is low, this may be omitted.

Add ESD devices near the speaker and microphone sockets to enhance ESD protection.
Reserve external header pins for UART0 on the baseboard to facilitate subsequent module upgrades.
The UART interface is used for communication with the main controller and can also be multiplexed as GPIOs.

The reference design is shown below.

Figure 7 Baseboard Design Reference

Power-Up and Startup¶

When using this module, mount it on the baseboard or host mainboard, connect the speaker and microphone, and power the module with 5 V; it will start automatically. If startup is normal, the speaker will play a boot prompt tone. Meanwhile, UART prints will be output. You can connect the UART to a PC using a USB-to-UART tool and view the prints in a serial terminal; output as shown in Figure 8 indicates normal startup. Note that the module’s UART interface is a 3.3 V logic high-speed serial port and also supports direct communication with 5 V logic without external level shifting.

The on-board power amplifier is powered by 5 V. The 5 V supply should provide a rated current of 500 mA with stable output and ripple within 300 mV.

Figure 8 Example UART Print After Startup

Default Command Set¶

For mass production, user-specified command-word firmware is usually programmed before shipment. If not specified, the module ships with default firmware containing a default command set for testing, as shown below:

Figure 9 Default Command Set

Default UART Communication Protocol¶

Modules programmed with the general-purpose firmware support UART communication for interfacing with a host or target system. The UART protocol is extensible and has the following features:

Complete transfer packet including header/footer, length, checksum, message type, and message sequence number.
Supports variable-length commands, enabling easy extension.
Message types include command, notification, and response.
Command messages are configurable and require ACK; notification messages have no ACK.
Message format is identical to the bootloader upgrade, distinguished by the header.
Default baud rate: 9600.
Note: Only UART0 is reserved on the module, and UART0 is the default print output interface. To use UART0 for the above protocol, code changes are required. Refer to the UART protocol section of ☞CI13XX Series Chip SDK.
Supported commands include: query protocol version, query system version, set volume (levels defined in user_config.h), play local broadcast audio, reset, etc. The specific protocol format is shown below:

Figure 10 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: valid data is 7 bytes

A0: command-word information

91: command number 0x91 (command-word data)

18: packet sequence number; this is the 0x08th data sent by this UART, incrementing continuously

01 55 E0 01 00 00: semantic ID, UNIQUE

1B: command-word threshold

9B 02: checksum

FB: end byte

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

Example 2:

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

A5 FC: header

02 00: valid data is 2 bytes

A3: notification data

9A: command number 0x9A (voice module content changed)

17: this is the 0x07th data sent by this UART, incrementing continuously

00 B1: valid data (indicates entering wake state)

05 02: checksum

FB: end data

Note: This is notification data; use it as needed.

For more details on message parsing, refer to the UART protocol section of ☞CI13XX Series Chip SDK. A reference screenshot is shown below:

Figure 11 Protocol Data Reference Screenshot

Software Development¶

The default firmware on the module is mainly for initial evaluation. For software development, please register and log in to the AI platform (https://aiplatform.chipintelli.com) for rapid voice firmware development. Meanwhile, you can download the SDK and related hardware materials in the “Development Materials” section of the AI platform. For first-time users of the AI platform, it is recommended to read the Getting Started Guide to understand the development process, and refer to the documentation center’s Video Tutorials for more solutions and SDK development introductions.

The software development process mainly includes the following steps:

SDK package download
Model generation(language model + acoustic model)
Text-to-speech (TTS)
Associate command-word table with audio files
Firmware packaging

For detailed development procedures, click ☞CI13XX Series Chip SDK.

Firmware Programming¶

Preparation Before Programming¶

Before programming the module, prepare the following items:

The module to be programmed
USB-to-UART tool
Firmware programming tool (pack_update_tool.exe)
Firmware file (*.bin)
Microphone and speaker matching the baseboard/host interfaces
Several Dupont wires

Figure 12 Firmware programming Required Items

Hardware Connection and Programming¶

Using the USB-to-UART tool shown above as an example, before programming, connect the tool’s power, ground, and TX/RX pins to the corresponding module pins (note that the tool’s RXD and TXD correspond to the module’s TX0 and RX0). The connection method is shown below.

Figure 13 Module and USB-to-UART Tool Connection

Open the firmware programming tool (found in CI130X_SDK\tools as PACK_UPDATE_TOOL.exe). Select the corresponding chip model, click the firmware update button, select the prepared firmware file, and confirm the COM port assigned to the USB-to-UART tool. After powering the module, it will enter firmware update mode and begin downloading the firmware. If the computer cannot recognize the USB-to-UART tool, please install the appropriate driver first.

Figure 14 Firmware programming Tool Screenshot

Functional Testing After Programming¶

After firmware programming is completed, it is recommended to perform functional testing to verify success. Before testing, connect a microphone and speaker to the module under test, power it on to confirm boot audio playback, and test wake word and command words for wakeup and recognition. If all are normal, the module is functioning properly and programming succeeded; otherwise, programming failed and further investigation is needed.

Possible Issues and Solutions During Use¶

This section lists some issues that may be encountered during use and corresponding solutions.

The module cannot be programmed/updated. If this occurs, check the following:
1. Verify UART pin connections; ensure TX and RX are not swapped; confirm the PC-side USB-to-UART driver is installed correctly; and confirm the correct COM port is selected in the programming tool.
2. Check whether the power supply is shorted, causing chip damage.
3. If the above checks are correct but programming still fails, use a multimeter to measure whether the 5 V, 3.3 V, and 1.1 V rails are correct. Refer to the test points in the figure below. If voltage issues are found, consider a hardware fault; replace the module or repair the hardware. If all checks are normal, contact our technical support team.

Figure 15 Module Testing Points Distribution Diagram

After programming, there is no playback on power-up. If this occurs, check the following:
1. Confirm the programmed firmware matches the board.
2. Confirm the speaker is properly connected and power is normal. Use an oscilloscope to measure the voice output test point of the main chip. If there is no output, check whether the firmware is correct. If there is output, check whether the power amplifier on the module has soldering anomalies; replace it if necessary and test again. The measurement point is shown below. If all checks are normal, contact our technical support team.

Figure 16 Module Audio Output Testing Points

After programming, power-on playback exists but command words are not recognized:
1. Check whether the microphone connections are intact.
2. Verify the microphone polarity matches the markings on the module and is not reversed.
3. Use a multimeter to measure whether the MICBIAS pin of the main chip is around 2.8 V, and use an oscilloscope to check whether there is an input voice waveform at the microphone input pin (set 100 mV per division). If the signal is normal, consider firmware issues; if abnormal, inspect the board hardware for physical damage. The measurement point is shown below. If all checks are normal, contact our technical support team.

Figure 17 Microphone Signal Test Point

Other Application Notes¶

Because the CI1311/CI1312 chips have relatively high ESD levels and the module is designed for easy expansion, ESD devices are not included on the module. For products with high ESD requirements, add ESD protection on the baseboard at the microphone, speaker, and power connector locations. It is recommended to wear anti-static wrist straps and gloves/finger cots during inspection and soldering to ensure product quality and reliability.
During use, ensure that microphone, speaker, and power/UART connections are correct.
You can use a USB-to-UART tool to debug your software. Add UART print statements at appropriate locations in the SDK, compile to generate the firmware, program it, and then perform debugging and verification.
All IOs on this module are typically 3.3 V level and are also 5 V tolerant.
When designing the baseboard or host mainboard, place a capacitor of at least 100 µF at the module’s 5 V input. Keep microphone traces as short as possible and shielded; keep SPK traces short and wide; avoid other traces crossing the SPK routing area.
Control baseboard warpage to less than 0.5% to prevent poor module soldering.

Production Guide, Storage, and Packaging/Ordering Information¶

Production and Storage Guide¶

Storage conditions for chipintelli gold-finger packaged modules are as follows:

Vacuum moisture-proof bags must be stored in a constant temperature and humidity room at 25 ± 5℃ and 65% ± 10% RH.
A humidity indicator card is included in the vacuum moisture-proof bag, as shown below:

Figure 18 Humidity Indicator Card

If the humidity indicator card shows the following color changes, bake the modules with the corresponding parameters:

Upon unsealing, if the 30%, 40%, and 50% rings are all blue, bake continuously for 2 hours
If the 30% ring turns pink, bake continuously for 4 hours
If the 30% and 40% rings turn pink, bake continuously for 6 hours
If the 30%, 40%, and 50% rings all turn pink, bake continuously for 12 hours

Baking parameters are as follows:

Baking temperature: 125 ± 5℃
Alarm temperature setting: 130℃
After natural cooling to < 36℃, proceed with SMT placement
Drying cycles: 1 time
If more than 12 hours elapse after baking without soldering, bake again

Before placement, provide ESD (electrostatic discharge) protection for the modules. Wear anti-static gloves and wrist straps during operation.
To ensure yield, perform visual inspection and AOI on all products to ensure correct oven temperature control, component pick-up, and placement.

Recommended Reflow Profile¶

Figure 19 Reflow Profile

Packaging and Ordering Information¶

Table 5 Bulk Ordering and Packaging Information

Product Model	Packaging Method	Modules per Tray	Modules per Pack	Modules per Carton
CI-D11GS10T CI-D12GS10T	Tray + ESD bag + Carton	– pcs	– trays, total – pcs	– bags, total – pcs

Purchasing and Technical Support¶

If you want to purchase product samples, click ☞Sample Purchase. You can also click ☞Samples and Bulk Purchase for more information.

If you need technical support, please log in to the ☞AI platform.