Open-source voice recognition platform ‘for all’

Open-source voice recognition platform ‘for all’
New Products |
The Matrix Voice is a Raspberry Pi-based or stand-alone dev board that allows makers and advanced industrial hardware developers alike to create voice-control apps within minutes, at attractive unit economics.
By Rich Pell


Measuring 3.14-inches in diameter, the Matrix Voice dev board features voice recognition integration with Google Voice and Amazon Alexa. It offers a radial array of 7 MEMS microphones connected to a Xilinx Spartan6 FPGA + 64 Mbit SDRAM, allowing developers to integrate custom voice and other hardware-accelerated machine learning algorithms right onto the board.

Voice also includes 64 GPIO pins (40 pins for Raspberry Pi, 16 GPIOs, 2 i2c, and power pins) for device-to-device connection. Other features include far-field voice capture, beamforming, acoustic source localization, noise suppression, de-reverberation and acoustic echo cancellation, and more.

A stand-alone 2.4-GHz Wi-Fi and Bluetooth-enabled 32-bit microcontroller version of Matrix Voice will also be available, satisfying the demand of mass commercialization of voice solutions for the industrial IoT.

Matrix Voice has been launched on Indiegogo for an early-bird price of just $45.00; and is expected to ship in May 2017.

Visit MATRIX Labs at
Check out the Matrix Voice Indiegogo campaign

Related articles:
Home audio market: ‘It’s all about the voice’
Far-field voice capture demonstrated using radar, audio beamforming
Mic array solution targets voice user interfaces in smart homes
Optical human-to-machine communication sensors promise ‘near-perfect’ speech recognition

Linked Articles