Zhiling Voice Module

Unisound Zhiling Voice Module provides high‑performance array pickup. Optimized for noise, it captures clear, accurate audio for rail transit, transport hubs and locomotive driver monitoring & training assessment.

Powerful Voice Processing

Leveraging mic‑array and deep‑learning algorithms, it realizes high‑accuracy ASR under heavy noise for smooth voice interaction.

Easy Integration & Cost Optimization

Compact for embedding in diverse smart devices. Full SDK & toolchains support rapid customization and deployment.

Flexible Custom Hardware

Optional 4/6/8‑mic arrays, cameras; built‑in/external mount, customizable high‑performance low‑power hardware.

Versatile AI Functions

Supports face, lip‑motion detection, multi‑language, voice wake‑up; integrates 3D avatars and custom TTS for virtual‑assistant interaction.

Core functional features

Excellent speech recognition and processing capabilities

Yunzhisheng's standard microphone array products, combined with deep learning algorithms and advanced speech recognition, ensure users receive a smooth and natural voice interaction experience. The microphone array technology can achieve high accuracy in ASR under various noisy environments

High integration and convenience

The product design is compact and easy to integrate, allowing for seamless embedding into various smart devices across different scenarios. Additionally, the provided development interfaces and tools are abundant, supporting rapid customization and deployment, thereby significantly reducing the integration difficulty and development cost for developers

Flexible array configuration and customizable hardware design

The product supports flexible microphone array configurations (4/6/8mic) and camera configurations according to the needs of application scenarios. It supports both pre-installed and externally attached device installations, and its customizable hardware design ensures low power consumption while maintaining high-performance operation

Driven by leading AI algorithms

Supporting face recognition and lip movement detection, Chinese and multiple languages including Minnan dialect, combined with interactive large-screen 3D virtual image and TTS customization services, it supports automatic voice to activate devices, creating a more vivid and lifelike virtual assistant image

Zhiling noise-cancelling microphone array parameter information

8-microphone noise reduction module

Number of microphone arrays: 8

Microphone type: PDM digital microphone

Acoustic Overload Point (AOP): 10% THD@1kHz, S=Typ, 120db SPL

Microphone sensitivity (S): 94 dB SPL @ 1 kHz, -26dBFS

Total harmonic distortion (THD): 94 dB SPL @ 1 kHz, S = Typ, less than 0.1%

Microphone directionality: omnidirectional

Microphone Signal-to-Noise Ratio (SNR): 20 kHz bandwidth, A-weighted fCLOCK=2.4 MHz, >64 dB

Microphone array layout: Linear arrangement

Microphone array interface type: USB

Microphone array power supply mode: USB bus power supply

Industrial PC

CPU：RK3588 Cortex‑A76 and Cortex‑A55， Frequency 2.4GHz

RAM：LPDDR4X，8G、16G、32G、64G

NPU：6.0 Tops

OS: Android 12 (Ubuntu/Linux optional)

Built in memory: support eMMC5.1, 64G (16GB/32GB/64G/128G optional)

Display output: HDMI 2.1 interface * 1, maximum output 8K 60Hz, HDCP2.3; DP interface * 1, 4Lane, maximum output 8K 30Hz, HDCP2.3

HDMI: up to 4K 60HZ

Ethernet: 1000M network interface

WIFI: wireless network access module, supporting WIFI6, 5G networks

Timing clock: support RTC

USB: USB 3.0 HOST Type A * 1; USB3.1 OTG Type‑C * 1； USB2.0 HOST Type A * 4

PCIE：PCIE3.0 M.2 M‑Key

Audio: module array: number of parallel array elements: 8, array element size: 35 * 8mm, wake-up distance:<20m, recognition distance:<5m, SNR>63dB, operating temperature: ‑ 20 ° C~80 ° C, sound source positioning: horizontal 360 °, positioning accuracy: ± 12 °, keyword wake-up, sound source positioning, noise suppression VAD

Industrial interface: RS232 interface * 2; RS485 interface * 1

Input voltage range: working voltage DC12V

Working current: 3A

Working environment: ‑ 10~75 °, relative humidity ≤ 80%

Speaker: connection 8 Ω 5W * 2, speaker output 4pin 2.0mm

Customer Stories

Benchmark customer trust choice, witness product value

Comprehensive transportation

AFC self-service intelligent voice ticket vending machine is equipped with smart listening noise reduction microphone array, which can deal with the noisy environment in the subway, complete voice ticket purchase, and achieve clear and reliable voice recognition.

Railway transportation

Intelligent listening noise reduction microphone array is used in the cab of high-speed railway locomotive to suppress the noise of complex equipment in the cab, clearly collect the voice of the stewards, and ensure the stability and availability of voice interaction and intercom calls.

Tod life circle

The customer service screen at the entrance and exit of TOD life circle is equipped with a smart listening noise reduction microphone array, which can effectively filter background noise such as noisy traffic at the entrance and exit, environmental broadcasting, etc., accurately pick up the voice of visitors, and ensure that the remote customer service intercom and voice consultation interaction are clear and stable.

Start your journey of intelligence

Our industry experts can provide you with targeted consultation on intelligent hardware solutions to help upgrade products intelligently

Cooperation consulting