Microphone dev kit delivers far-field speech accuracy for Alexa-enabled devices

By Gina Roos, editor-in-chief

DSP Group, Inc. has launched its HDClear 3-Mic Development Kit for Amazon Alexa Voice Service (AVS). At the heart of the kit is DSP Group’s DBMD5 advanced audio/voice processor (SoC) and HDClear voice enhancement processing technology. The company claims that the development kit is the lowest-power solution for developers who want to build high-accuracy Alexa-enabled devices, including smart speakers, wearable computing devices, smart home devices, and remote controls.

The three main kit components include a printed circuit board (PCB) with the DBMD5 audio processor, Raspberry Pi, and a three-microphone array. It also provides pre-process noise-reduction, beamforming, and Acoustic Echo Canceling (AEC) algorithms for enhanced voice accuracy.

Touted as being simple to operate — just connect it to a speaker and plug it into power — the kit is designed to reduce time-to-market and speed up development efforts.

DSP Group’s list of key benefits of the development kit include:

• Voice enhancement capabilities — Sophisticated voice algorithms including acoustic echo cancellation, noise reduction, beamforming, and far-field enhancement for greater voice command accuracy.
• Barge-in — Enables detection of voice commands when audio, including music, is played by the device.
• Seamless operation with Amazon AVS — Supports Amazon Alexa wake word and fully integrates with AVS SDK.
• Advanced audio SoC — Embedded programmable dual-core DSP supports digital and analog microphones and incorporates various application processor interfaces such as SPI, I² C, UART, and SLIMbus.
• Low power consumption — Power-optimized implementation for always-on IoT devices allows application processors to remain in sleep mode during voice sensing and enables battery-powered devices to access Alexa.  Power consumption well below 1 mA for Alexa wake-word detection is supported.
• Far-field enhancement — Use of three mics enables up to 360° voice coverage.
• Local detection of Alexa — Enables low-power detection and integration in IoT devices with small amounts of memory for wake-word detection.
• Fast time-to-market — Provides a complete suite of host processor/communications processor/MCU software drivers.
• Scalable — Software configuration supports several types of devices including smart speakers, digital assistants, wearable computing devices, smart home devices, remote controls, and IoT devices.

A product brief and user guide with step-by-step instructions are available at the company’s website.