GYKL-5M-DOA360-I is an AI-enabled microphone array module designed for reliable far-field voice pickup in noisy environments.
Based on a multi-microphone array architecture and directional beamforming, combined with deep learning–based noise reduction algorithms, the module can separate target speech from background noise in real time and deliver clear voice signals for intelligent voice interaction.

In typical home and indoor environments, GY6228AEC supports far-field voice pickup at distances of approximately 3–5 meters, providing high recognition accuracy and fast response through a USB 2.0 standard audio interface.

The module integrates a comprehensive set of audio processing functions, including Acoustic Echo Cancellation (AEC), speech enhancement, noise suppression, sound source localization, and Automatic Gain Control (AGC). Even in high-noise scenarios, users can experience clear and smooth audio and video communication.

Beamforming & Sound Source Localization

 GYKL-5M-DOA360-I combines noise reduction and beam angle estimation algorithms to support omnidirectional sound source localization.
The module can calculate the direction-of-arrival (DoA) of human speech and output independent beam angle information via USB, enabling advanced spatial audio awareness for intelligent systems.

Audio Architecture & Development Support

• Supports up to 5-channel microphone array signal processing

• All microphone channels are sampled using a shared clock, ensuring strict inter-channel synchronization for accurate beamforming and localization

• Digital microphone audio is output in PCM format

• Noise reduction and digital microphone gain parameters can be configured

• Supports secondary development on host platforms such as Android, Windows, and Linux, enabling array signal processing algorithms including:

  • Direction estimation

  • Speech enhancement

  • Noise suppression

DSP Processing Platform

The microphone array system is built on a high-performance audio DSP processor optimized for AI audio applications:

• Ultra-low-power, high-performance 32-bit RISC core

• Maximum operating frequency up to 320 MHz

• Integrated DSP instruction set and floating-point unit (FPU) for advanced audio algorithms

Audio Performance & Interfaces

• Microphone Input:

  • Multi-channel digital microphone inputs (array architecture)

• DAC:

  • 2-channel line-level audio input/output

  • Signal-to-noise ratio (SNR) ≥ 105 dB

• Audio Resolution:

  • Maximum bit depth: 16-bit

• Supported Sampling Rates:

  • 16 kHz / 22.025 kHz / 24 kHz

  • 32 kHz / 44.1 kHz / 48 kHz

• Headphone Output:

  • Directly drives 16 Ω or 32 Ω headphones

• Digital Audio Interface:

  • One S/PDIF interface (half-duplex TX/RX), supporting HDMI audio and ARC

Power, Clock & USB Interface

• Power Supply:

  • USB DC 5 V input @ VDDIN

• Clock Sources:

  • 12 MHz RC oscillator

  • PLL clock source

• USB Interface:

  • USB 2.0 Full-Speed (OTG) controller

  • Supports multiple endpoints

  • Integrated PHY, compliant with standard USB audio protocols

• USB-C Audio Codec Solution:

  • High-quality voice pickup

  • Active noise cancellation

  • Noise suppression

  • Wind noise reduction

Typical Applications

• AI voice interaction terminals

• Smart home devices

• Voice-controlled consumer electronics

• Audio and video communication systems

• Intelligent assistants

• Human–machine interaction products