XVF3620
The XVF3620 voice processor adds AI-based noise reduction to enable reliable audio capture with one or two microphones. The combination of a complete voice pipeline with AI and a fast-adapting acoustic echo canceller (AEC) enables voice control and communication with humans or robots in demanding acoustic environments—whether they are noisy, reverberant, or involve a moving application.

SYSTEMS

ENVIRONMENTS

APPLIANCES

ROBOTS

ASSISTANTS
Key features
AI Denoising
- Separates speech from background noise
- Adjustable target SNR
- Trained on a wide range of noise types, from industrial machinery and lawnmowers to street noise and pets demanding attention
- Preserves speech quality
- Also trained to removes reverb
- Low latency, suitable for intercom and control applications
Fast Echo Cancellation
- Tuned for speed: adapts in less than a second
- Advanced residual canceller and post-filter
- Can be combined with one or two microphone channels
- Double-Talk Detection
- Enables barge-in support for moving applications such as service robotics and kitchen appliances
Flexible Interfaces and Integration
- Audio interface options via USB or I²S (master or slave)
- Control interfaces via I²C, UART or USB HID
- Firmware updates (DFU) via I²C, USB or SPI
- Boot from QSPI flash
- Optional boot from a local host processor
Versatile Audio Options
- Supports a range of microphone configurations
- Single microphone for omnidirectional capture
- Two microphones for spatial enhancement using a generalised sidelobe canceller (GSC) in noisy environments
- Direct connection of PDM microphones to XCORE
- Integrated far-end DSP for EQ, limiting and multiband dynamics to achieve the best performance from any integrated loudspeaker
Discover more
Hardware
| PART NUMBER | DATASHEET | BUY |
|---|---|---|
| XVF3620-QF60B-C | DATASHEET | CONTACT US |
Start your development with the


2-MIC VOICE
DEVELOPMENT KIT
The 2-Mic voice development kit is available to help you evaluate the XVF3620 and prototype your voice enabled products. It supports USB plugin accessory implementations (with a USB Control Interface) and also built-in voice implementations (with an I2C Control Interface).
Our kit features:
- XU316-1024-QF60A-C24 XCORE.AI processor
- Raspberry Pi HAT connector
- 2 x Infineon IM69D130 MEMS mics
- 71mm inter-mic spacing
- Microphone mute switch
- Speaker output (Line level)
- USB / I2S host interface support
