Why voice user interfaces with beamforming microphones work better with xCORE-VOICE
We love the Amazon Echo at XMOS. It’s a new-to-the-world category of product, brimming with possibilities as a digital assistant, a hub for home automation as well as a point of presence to allow us to access all of Amazon’s goods and services.
At its heart is a piece of technology we would call a smart microphone. This enables the Echo to capture voice samples with a high degree of accuracy before transmitting them to the Amazon Voice Services in the cloud where the query is processed before the answer is returned to the device in the form of Alexa’s soothing tones contained in an MP3 file.
The smart microphone is actually composed of a number of individual microphones linked together by a technology known as beamforming, where complex trigonometric functions enable the individual microphones to be combined to create a highly directional beam, which can focus on the current speaker. Not only that, the beam can be steered to track a speaker moving in the room so that voice samples are never lost. This capability, coupled with advanced echo cancellation to prevent whatever is being played through the Echo’s speaker at the time being captured by the microphones is combined with noise suppression to remove background sounds such as traffic noise or air conditioning from the signal.
It’s an amazing piece of kit that makes the dream of “talking” to a computer a reality. As technologists, we’re impressed by its functionality; as marketers, we’re tantalized by the view of the future that the Echo offers us.
That’s why I’m excited about introducing XMOS’ latest product, the xCORE-VOICE XVSM-2000, which is a smart microphone chip designed to bring voice interface capability to the consumer, domestic and computer markets.
The performance and flexibility of XVSM-2000 opens up new possibilities for designers and marketers looking to integrate a voice user interface into existing products, or use the capability to create new product categories, much like the Amazon Echo. We’ve integrated sophisticated microphone and voice DSP functions, delivering voice activity detection, beamforming and beam steering alongside noise suppression, echo cancellation and de-reverb functions. This sits on top of our PDM microphone aggregation capability that we announced last month.
XVSM-2000 still retains all the features that our customers love about our xCORE technology: multicore performance - up to 1800MIPS available for your application - flexible I/O to create the mix of interfaces unique to your product, our industry-standard USB or TDM backhaul solutions, and of course, all programmed via the ‘C’ language.
We’ve created the xCORE-VOICE Smart Microphone development board to enable customers to quickly evaluate our technology and develop their own applications. The board has an XVSM-2000 device, 7 PDM microphones, USB-B connector (TDM outputs via a header) and audio out via a headphone jack, and is supported by voice and microphone libraries available from XMOS.
The xCORE-VOICE Smart Microphone development board and XVSM-2000 are available today – contact your local XMOS sales representative for more details or visit www.xmos.com/products/voice.
Comment on this post via social media
« The xCORE 64-bit accumulator and memory busVoice recognition is everywhere, but accurate capture is the key »