Voice recognition interfacing is the new frontier of IoT and smart home technology, and while innovative applications like Amazon’s Alexa make voice activated interaction look easy, the technology behind it is quite advanced.
Engineers experimenting with voice recognition using the Raspberry Pi and a simple microphone have run into difficulties with accuracy even from just a short distance within a room. The difficulties of voice recognition at a distance—called Far Field Detection—stem from the fact that there is a lot of conflicting noise in the world around us. Parsing what is an actual command from what is just secondary conversation or background noise involves complex filtering and mathematics.
Developing solutions that integrate accurate voice recognition has been difficult and cost-prohibitive for many developers—until now.
The Alexa Voice Service (AVS) approved AudioSmart™ 2-mic Development Kit from Conexant gives engineers an affordable alternative to multi-mic array designs and expedites the development of 3rd Party Alexa smart home products.
The AudioSmart™ 2-mic Development Kit features Conexant’s powerful CX20921 high-performance hands-free Voice Input Processor System-on-Chip (SoC). Its dual microphone voice processing capability uses Alexa “wake word” technology to respond to voice commands from anywhere in the room, even in noisy conditions. A wake word is a word that triggers a system to wake from lower power mode to fulfill a command, allowing the device to save power but still have the capability to respond quickly as needed.
The proprietary Smart Source Pickup™ (SSP) noise-suppression algorithm in the CX20921 detects voice and cancels surrounding noise. It can separate the voice commands from music and voice prompts being played out of the device, even with background noise present. It even allows for “barge in”, which means users can interrupt the device when it’s playing sound—like during a song, for instance.
Not only is Arrow Electronics the distributor of the AVS approved AudioSmart™ 2-mic Development Kit, but we also source the components and are able to provide technical design support for kit applications.
This partnership between Arrow and Conexant was developed with the engineer in mind. Arrow is uniquely positioned to fulfill the demand for the Amazon AVS-approved AudioSmart™ dev kit, which will allow engineers, makers, and innovators of all types to create Alexa enabled prototypes faster and more economically than ever before. The system is designed to be easily integrated into any AVS prototype based on the Raspberry Pi—one of the most accessible and innovation-friendly SBCs on the market.
With the AudioSmart™ 2-mic Development Kit from Conexant, the power of Alexa wake word technology is now in the hands of engineers and makers everywhere. All that remains is to see what innovative voice command solutions we’ll see next.