ipXchange has previously talked about Espressif’s ESP32-S3-BOX with our good friend Shaun O’Brien from Ineltek, but such an interview could only scrape the surface of what this inexpensive product has to offer, so let’s take a closer look at Espressif’s feature-packed AIoT development kit.
At its heart is an ESP32-S3 SoC, which contains an Xtensa LX7 dual-core CPU capable of running voice and image AI processing. In addition to this, Bluetooth 5 LE and Wi-Fi (IEEE 802.11b/g/n) functionality is present in this compact device; the ESP32-S3-BOX is even capable of using the latter for human body detection. Such high performance supports the ESP32-S3’s use as part of a smart user interface in connected smart home devices, white goods, industrial control panels, and healthcare monitors, to name just a few of the possible applications.
The ESP32-S3-BOX can be used to quickly create a proof-of-concept design for end products thanks to a wealth of human-machine interfaces (HMIs). A 2.4-inch capacitive touch display (320 x 240 px) is included for developing a visual user interface, and a lite version of the kit swaps this capacitive touch for 3 user buttons for a more cost-effective solution, or for use in applications where basic capacitive touch is simply not viable, such as where the user is likely to be wearing gloves.
Along with this visual side to the kit, dual in-built microphones and a speaker support the included offline voice assistant that can be used to integrate over 200 custom voice commands for your end product, with the device also supporting third-party online voice assistants, like Alexa, and wake functions. To support these AI voice functions, Espressif’s audio front-end algorithm provides acoustic echo cancellation, blind source separation, and noise suppression to ensure good far-field performance in noisy environments.
These fantastic HMIs will certainly get you started with your design, but the ESP32-S3-BOX is also expandable via two Pmod-compatible headers to add your own peripherals, and the ESP32-S3 itself features interfaces for 8- to 16-bit DVP camera input, an LED PWM controller with up to 8 channels, 2 × 12-bit SAR ADCs with up to 20 channels, and more.
Even in this longer-form article, we’ve not covered the ultra-low power consumption, the in-built security features, or the benefits of the connectivity options of the ESP32-S3, so follow the link to our board form to learn more, where you can also apply for a sample to evaluate the ESP32-S3 SoC for use in a commercial application.
Keep designing!