The popularity of Amazon's Alexa Voice Service (AVS) is a testament that voice user interface (UI) is the next big thing, successor to the touchscreen that revolutionized smartphone and tablet designs. A variety of products—ranging from wireless speakers to smart home appliance to industrial panels—are deploying Alexa audio interface in deeply embedded systems.
And there are efforts underway to simplify the experience of creating the Alexa-enabled products, thus helping developers get AVS designs to market faster. A number of development kits built around the Alexa audio technology allow developers and OEMs to accelerate the design of voice-enabled products.
Take, for instance, Amlogic, which has unveiled an Amazon-qualified development kit for voice command products in collaboration with DSP Concepts, a developer of embedded audio processing solutions. The design solution is based on Amlogic's A113X audio processor and the Audio Weaver® Voice UI software from DSP Concepts.
Amazon-qualified kit facilitates Alexa designs in noisy environments. Image courtesy of DSP Concepts.
The soft-DSP solution eliminates the need for a dedicated DSP chip while consuming only a fraction of the A113X SoC’s processing capabilities for Alexa-centric tasks. More importantly, it tackles a critical problem that is limiting the widespread adoption of voice UI. "Voice recognition in noisy environments has been the Achilles’ heel of voice-enabled products," said Dr. Paul Beckmann, co-founder and CTO of DSP Concepts.
Alexa Design Diary
In January's CES 2018, Frontier Smart Technologies, a system integrator from the UK, demonstrated its SmartSDK solution for Alexa-based designs. The company announced that its design kit for developing voice-enabled audio devices will be available in mid-2018.
Alexa design kit is built around a single-chip solution encompassing DSP, programmable I/O and control software. Image courtesy of XMOS Ltd.
Another notable development kit for Amazon's Alexa designs comes from XMOS, a supplier of embedded voice and audio chips for consumer devices (see above figure). The Bristol, UK-based firm claims that its reference design for Alexa Voice Service is the first far-field linear mic array solution.
The VocalFusion 4-Mic Dev Kit employs a compact four-microphone linear 100 mm array to accurately capture voice signals from across the room even in noisy environments. It's built around the company's VocalFusion XVF3000 chip, a far-field voice processor that delivers voice DSP and full-duplex acoustic echo canceller (AEC) capabilities.
The two-mic solution for Alexa-enabled designs comes with proven hardware and software components. Image courtesy of Cirrus Logic.
Then, there is the two-mic solution from Cirrus Logic that comes with an easy-to-use voice capture development kit for creating Alexa-enabled products. The reference board shown above features Cirrus Logic’s CS47L24 smart codec, CS7250B digital MEMS microphones and SoundClear® algorithms for voice control, noise suppression, and echo cancelation.
Here, the smart codec comprises of an on-chip digital-to-analog converter (DAC) and a 2 W mono speaker driver to enable high-fidelity audio playback. The design solution is targeted at networked speakers and smart home appliances.
It's likely that more Alexa-focused kits will be released in the coming year. What kits have you seen that interest you? Let us know in the comments below.