Adafruit EdgeBadge - TensorFlow Lite for Microcontrollers

Machine learning has come to the 'edge' with the Adafruit EdgeBadge, a compact, credit card-sized development board tailored for TensorFlow Lite on microcontrollers. Powered by the ATSAMD51J19 processor running at 120MHz with 3.3V logic and power, it features 512KB of flash, 192KB of RAM, and an additional 2MB of SPI flash for storage of images, sounds, animations, and TensorFlow Lite files.

This board includes a 1.8" 160x128 color TFT display on its own SPI port, 8 silicone button tops for game and control inputs, 5 NeoPixels for badge dazzle or score-keeping, a triple-axis accelerometer, a light sensor, a built-in mini buzzer speaker, a mono Class-D speaker driver supporting 4-8 ohm speakers up to 2 Watts, and a PDM microphone for micro speech recognition. It also features a LiPoly battery port with built-in recharging, USB port for charging and programming, two female header strips compatible with FeatherWings, JST ports for NeoPixels, sensor input, and I2C (compatible with Grove connectors), along with an on-off switch and reset button.

You can program the EdgeBadge using CircuitPython, MakeCode Arcade, TensorFlow Lite for Microcontrollers / Arduino, and more. The included Arduino library comes with demos for voice recognition, detecting word pairs such as "yes/no", "up/down", and "cat/dog".

• ATSAMD51J19 @ 120MHz with 512KB FLASH + 192KB RAM
• 2 MB SPI Flash for extended storage
• 1.8" 160x128 Color TFT Display
• 8 Silicone tactile buttons
• 5 NeoPixels
• Triple-axis accelerometer
• Light sensor mounted on front
• Built-in buzzer speaker and mono Class-D speaker driver (up to 2 Watts)
• LiPoly battery port with charging capability
• USB port for charging, programming, debugging
• Two female Feather-compatible header strips
• JST ports for NeoPixels, sensor input, and I2C (Grove compatible)
• Reset button and On-Off switch

Note: TensorFlow Lite for Microcontrollers is a cutting-edge technology area; expect ongoing development and updates.