ESP32-S3-BOX is Espressif’s AI voice-development kit, which provides a platform for developing the control of smart devices with offline and online voice assistants. ESP32-S3-BOX is ideal for developing not only smart speakers, but also IoT devices that achieve human-computer voice interaction directly. To this end, ESP32-S3-BOX integrates a touch screen controller, various sensors, an infrared controller and a smart gateway.

In fact, ESP32-S3-BOX can function as the “control center” for users’ entire surrounding area, allowing them to easily connect multiple smart devices, either with voice commands or by using the device’s touch screen. The main functions of ESP32-S3-BOX also include Espressif’s AI image processing, Wi-Fi human-body detection, and wireless image transmission. These functions can be particularly useful in an office environment, facilitating the running of the reception area and conference rooms.

With ESP32-S3-BOX developers can quickly build a high-performance, user-friendly, AI voice-controlled solution. ESP32-S3-BOX is based on Espressif’s ESP32-S3 Wi-Fi + Bluetooth 5 (LE) SoC, with AI capabilities. In addition to ESP32-S3’s 512KB SRAM, ESP32-S3-BOX comes with 16MB of QSPI flash and 8MB of Octal PSRAM. ESP32-S3-BOX is also equipped with a variety of peripherals, such as a 2.4-inch display with a 320x240 resolution, a capacitive touch screen, a dual microphone, a speaker, and two Pmod™-compatible headers which allow for the extensibility of the hardware. ESP32-S3-BOX also uses a Type-C USB connector that provides 5 V of power input, while also supporting serial and JTAG debugging, as well as a programming interface; all through the same connector.

ESP32-S3-BOX runs Espressif’s own audio front-end (AFE) algorithm, as well as ESP-Skainet (Espressif’s offline voice-assistant SDK), and the Alexa-for-IoT SDK, thus providing users with excellent offline and online voice functions. Additionally, ESP32-S3-BOX is capable of running the LVGL-based HMI solution, and such SDKs as ESP-DL and ESP-ADF. Espressif’s complete AIoT platform, ESP RainMaker®, can also be used with ESP32-S3-BOX for configuring GPIOs and offline commands, while providing control via phone apps and/or a voice assistant. With such rich hardware and software features, as well as a product-ready form factor, Espressif envisions that ESP-S3-BOX will significantly simplify product development for its customers.

It is worth mentioning that the latest version of ESP-Skainet (Espressif’s offline voice-assistant SDK) supports two new functions which can materialize a more efficient interactive experience tailored to users’ needs:

1. 1. Support for waking up the device at any time while it is speaking or playing music.
   2. Support for talking to the device continuously once it is woken up.

In addition, ESP-Skainet reduces the rate of false wake-ups and improves the device’s speech-recognition performance while ensuring an excellent wake-up rate.

Furthermore, in order to meet the different functional and cost requirements of our diverse clientele, we have launched ESP32-S3-BOX-Lite, a simplified version of Espressif’s AI voice-development kit. It has similar offline and online voice functions to ESP32-S3-BOX but without its capacitive touch panel and mute buttons. However, ESP32-S3-BOX-Lite has three additional function buttons which can be customized by users. The table below shows a comparison between ESP32-S3-BOX and ESP32-S3-BOX-Lite.

Espressif provides a number of technical resources relating to ESP32-S3-BOX, including its hardware reference design and user guides. These will facilitate our customers to prototype their products with ESP32-S3-BOX, and then customize them quickly, reducing time-to-market and engineering costs.

If you are interested in ordering ESP32-S3-BOX, please go to Amazon in case you are a US-based customer, or to AliExpress and Adafruit from other parts of the world. If you want to know more about the product, please contact our customer support team, who will try and assist you as soon as possible.