Advertisement

Multi-mic dev kit supports far-field voice pickup for Amazon Alexa

Audio front-end solution offers one-, two-, and three-mic configurations

By Heather Hamilton, contributing writer

Offering designers the flexibility to build devices with multiple mic array configurations, Microchip Technology Inc. via its Microsemi Corp. subsidiary released the AcuEdge ZLK38AVS Development Kit for Amazon Alexa Voice Service (AVS), which offers one-mic hands-free and two- or three-mic far-field configurations qualified by Amazon. The dev kit also introduced far-field voice pickup.

Microchip said that the dev kit addresses the developers’ challenges of scaling to meet different bill of materials (BOM) and form-factor requirements as they build products in various sizes and designs to meet customer demand.

The audio front-end development kit consists of the ZL38063 audio processor with a 300-MHz DSP and hardware accelerators dedicated to voice processing. It also features stereo acoustic echo cancellation (AEC), enabling barge-in recognition for customers in harsh audio playback environments. 

Microchip-ZLK38AVS-Development-Kit

Original equipment manufacturers (OEMs) and third-party developers will be able to use the kit for a variety of applications and form factors. Configurations available include:

  • One-mic configuration is designed for hands-free voice coverage in products with form-factor or cost constraints that prohibit the use of a multi-microphone array.
  • Two- and three-mic linear (180° far-field coverage) configurations can be used for products such as sound bars, smart plugs, and set-top boxes that are typically positioned against a wall.
  • Three-mic triangular (360° far-field coverage) configurations target applications such as speakers, wide-angle IP cameras, and voice-enabled Wi-Fi routers that may need interaction from any direction.

The kit allows for far-field speech recognition in noisy environments, which includes external noise sources and device playback. The field-programmable and field-upgradeable solution provides better local trigger-detection performance and cloud speech-recognition accuracy thanks to its signal-processing algorithms, said Microchip. Multi-microphone configurations include direction of arrival (DOA) estimation to indicate the location of the primary voice sound source.

Included with the development kit is a ZL38063 audio processor that connects to a Raspberry Pi 3B with plastics and mounting hardware to simulate a typical end-application mic-speaker setup. All microphone options are supported, and it includes an LED ring to report the detected sound location. Microchip audio labs and tuning facilities support customers from design to Amazon certification.

The ZLK38AVS Development Kit for Amazon AVS is priced at $299.

Advertisement



Learn more about Electronic Products Magazine

Leave a Reply