Espressif has launched ESP32-Korvo, an ESP32 and ESP-Skainet based AI development board for embedded IoT devices. The board is designed with multiple peripherals, earphones, speakers, an SD card slot, LED lights, and function buttons, all of which can serve a variety of AI-powered voice applications. ESP32-Korvo features a multi-microphone array that is suitable for far-field speech recognition applications with low-power consumption.
ESP32-Korvo’s motherboard contains the ESP32-WROVER-B module with 16MB flash and 8MB PSRAM, a power port, a micro-SD card slot, along with a speaker and earphone connectors. Its daughterboard contains a microphone array, function buttons, and LEDs. The ESP32-Korvo is equipped with a single ESP32 chip and two boards connected with an FPC cable, and it supports low power, far-field speech recognition applications.
The new device operates with front end acoustic algorithms such as Microphone-Array Speech Enhancement (MASE) and Acoustic Echo Cancelation (AEC) for reducing the interference of any noise in far-field speech commands along with the improvement in the performance of voice wake-up and speech recognition.
The users can combine ESP32-Korvo with Espressif’s Smart Voice Assistance, ESP-Skainet to achieve optimal application development, and customize wake-up words with its voice wake-up engine, Wakenet. The users can also add their voice commands, and realize local voice control over smart devices with ESP-Skainet’s command recognition network, Multinet.