Seeed Studio ReSpeaker XVF3800 USB Microphone Array

The Seeed Studio ReSpeaker XVF3800 USB Microphone Array is a professional 4-microphone circular array based on the XMOS XVF3800 audio processor. It enables 360° far-field speech capture with a range of up to 5 meters. The array features advanced audio processing functions such as acoustic echo cancellation (AEC), multi-beam beamforming, reverberation suppression, direction of arrival (DoA) detection, DNN-based noise suppression, and automatic gain control (AGC) with a dynamic range of 60 dB. The board offers two operating modes: a USB plug-and-play mode for direct connection to PCs and an INT-Device mode (I2S) for integration into embedded systems. Twelve individually controllable WS2812-RGB LEDs provide visual feedback on device status and speech activity. A mute button allows immediate deactivation of microphone recording, with a red LED indicating the mute status. Audio outputs include a 3.5-mm headphone jack for headphones or active speakers and a JST connector for an amplified speaker with up to 5W. The integrated audio codec TLV320AIC3104 handles audio conversion, while an LM48511 Class-D amplifier drives the speaker output. The device is compatible with Windows, macOS, Raspberry Pi OS, and microcontroller hosts such as the XIAO series, ESP32, and Arduino via I2S and I2C. A unique device serial number enables identification and management in multi-device setups. Firmware can be updated via USB DFU, with various firmware options available for USB, I2S, and Home Assistant applications.

Recommended Accessories:

• 5W 4Ohm Speaker in Plastic Housing
• 2W 8Ohm Speaker in Housing (B)

Features:

• Four PDM MEMS microphones arranged in a circular pattern enable 360° far-field speech capture up to 5 meters.
• The XMOS XVF3800 processor provides AEC, beamforming, reverberation suppression, DoA detection, and DNN-based noise suppression.
• Two operating modes allow both USB plug-and-play and I2S integration with embedded systems.
• Twelve individually controllable WS2812-RGB LEDs visually indicate device status and speech activity.
• A built-in mute button deactivates microphone recording and signals the status via a red LED.
• The unique device serial number supports multi-device deployments and advanced device management.

Use Cases:

• Voice control and speech recognition in smart home systems with Home Assistant
• Conference and hands-free solution for video conferences on PC or Raspberry Pi
• Development of voice assistants with direction detection and far-field audio capture
• Embedded speech processing in robotics and IoT projects via I2S interface
• Acoustic monitoring and speech activity detection in research applications

Documentation:

• Manufacturer Website
• GitHub Link
• here
• Python Control Directory
• https://github.com/respeaker/respeaker-console/releases
• Instruction Manual

Technical Specifications:

Manufacturer: Seeed Studio

Manufacturer SKU: 101991441

Audio Processor: XMOS XVF3800

Audio Codec: TLV320AIC3104

Amplifier: LM48511 (Class D)

Microphones: 4× PDM MEMS in circular arrangement

Speech Capture Range: up to 5 m (360°)

Audio Processing: AEC, Beamforming, Reverberation Suppression, DoA, DNN Noise Suppression, AGC (60 dB)

RGB LEDs: 12× WS2812 (individually controllable)

USB Standard: USB Audio Class 2.0

Connection: USB Type-C (power and data)

Audio Output: 3.5-mm headphone jack, JST speaker connector (up to 5W)

Interfaces: I2C, I2S

Operating Modes: USB Mode, INT-Device Mode (I2S)

Compatibility: Windows, macOS, Raspberry Pi OS, XIAO Series, ESP32, Arduino

Package Dimensions: 145 × 132 × 21 mm

Gross Weight: 70 g

Contents:

• 1× Seeed Studio ReSpeaker XVF3800 USB Microphone Array