The kit consists of Conexant's AudioSmart CX20921 hands-free voice input processor, and Alexa wake word technology. The AudioSmart 2-mic development kit for Amazon AVS will allow product developers to build voice-enabled products with Alexa.
It has been designed to be integrated into any AVS system prototype based on the Raspberry Pi. This is intended to reduce the time needed for prototyping and engineering new digital assistant devices from over six months to a couple of weeks.
There are substantial acoustic challenges with speech recognition devices. Problems related to echo cancellation, background noise, position of microphones, and speakers need to be overcome to ensure that speech recognition and voice control from several feet away will work.
The development kit is designed to recognize the Alexa wake word, and deliver speech requests for processing from anywhere in the room, even if the room is noisy. The kit will also enable "voice barge-in." This means that users can interrupt their Alexa device when it is playing sound.
The system captures the user's voice from anywhere within the room. It can separate voice commands from music and voice prompts being played out of the device, even with background noise present.
The AVS hears only the user's command, and nothing else, to ensure that it is consistently accurate.
Steve Rabuchin, vice president of Amazon Alexa said "We created our AVS for Raspberry Pi project to help expedite the development of Alexa smart home products. Utilizing Conexant's AVS approved solutions will help third-party manufacturers quickly innovate with Alexa."