Espressif’s EchoEar ESP32-S3 voice-controlled AI chatbot runs esp-brookesia firmware

Espressif Systems’ EchoEar is a compact ESP32-S3 AI chatbot designed for voice interaction and edge AI applications, for smart toys, voice-enabled speakers, and control systems. It features a 1.85-inch circular touch display, a dual microphone array with local wake-word detection, and support for large AI models from OpenAI, Xiaozhi AI, and Gemini.

The kit is built around the ESP32-S3-WROOM-1 Wi-Fi 4 and Bluetooth 5 module, and also integrates a 3W speaker for audio interaction, and a microSD card slot for data storage. Other hardware features include a BMI270 IMU, a green LED, a USB-C port, a magnetic connector, and a battery management chip.

Espressif EchoEar Specifications

• Wireless Module – ESP32-S3-WROOM-1
  • SoC – Espressif Systems ESP32-S3R8
    • CPU – Dual-core Tensilica LX7 up to 240 MHz with vector extension for AI/ML workloads
    • RAM – 512KB SRAM, 8MB PSRAM
    • ROM – 284KB
    • Wireless – WiFi 4 and Bluetooth LE 5
  • Storage – 16MB QSPI flash
  • Antenna – PCB antenna
• Storage – MicroSD card slot
• Display
  • 1.85-inch round QSPI LCD, 360 × 360 resolution, ST77916 driver
  • Capacitive touch with ESP32-S3 native touch sensors
  • Backlight control via GPIO43
• Audio
  • Dual LMA3729T381-OY3S microphone array (supports local wake-word, sound source localization)
  • Built-in 3W mono speaker (connector)
  • NS4150B 3W Class-D amplifier
  • ES7210 audio ADC (4-ch)
  • ES8311 audio codec (ADC + DAC)
• USB – USB Type-C port for programming, power, and log printing
• Expansion
  • Pogopinand magnetic connector (serial + 5 V)
  • Internal I2C header
• Misc
  • Buttons – Power (on/off toggle), Reset, and BOOT (download mode)
  • Green status LED
  • M1.6 welded nuts for housing mounting
  • Bosch BMI270 6-axis IMU Sensor
  • Sliding touch interface (2 pads on v1.2)
• Power
  • 5V via USB-C port
  • 3.7 V Li-ion battery (connector)
  • Magnetic connector (5 V + UART passthrough)
  • BQ27220 battery management chip
  • TP4057 Li-ion charging IC (250 mA charging current)
  • TlV62569 / SY8088AAC buck converters (5 V → 3.3 V)
  • SAM8108 power control chip (power on/off management)
• Dimensions – TBD

On the software side, the EchoEar can be programmed and debugged over USB-C using the ESP-IDF framework. Espressif provides a dedicated Board Support Package (BSP) that simplifies access to key peripherals such as the LCD display, audio subsystem, buttons, and LEDs. Developers can also find ready-to-use application examples in the esp-brookesia project, covering voice interaction, multimedia, and AI agent use cases. The kit is designed for edge AI applications and integrates smoothly with large-model services such as OpenAI, Gemini, and Xiaozhi AI, enabling offline voice wake-up, multimodal interaction, and full-duplex conversational agents directly on the device. More information is available on the wiki.

Previously, we have written about other ESP32-S3-based development boards that support AI voice features, like the ESP32-S3-Touch-LCD-3.49 touch display development board, the ESP32-S3-AUDIO-Board, the LILYGO T-Circle-S3, and other ESP32-S3 devices. But the EchoEar is an official kit from Espressif, and goes further, combining a 1.85″ round display, dual-mic array, 3W speaker, battery management, and esp-brookesia support for a complete voice-AI kit with offline wake-word, sound localization, and large-model integration.

The Espressif EchoEar AI chatbot can be purchased from AliExpress for $40 and on Amazon for $55.00.

Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.