Object recognition using Artificial Intelligence has never been this convenient ever before.

Sipeed MaixCube is an all-in-one development platform based on the M1n module which is powered by the Kendryte K210 core. It is equipped with a camera, TF card slot, user buttons, TFT display, lithium battery, speaker, microphone and an expansion interface. This provides a great platform for everyone to start learning about AI development.

The KendryteK210 which acts as the main control chip for the MaixCube, is very powerful. The module has a built-in 64-bit dual-core processor chip and 8M of on-chip SRAM. It has outstanding performance in AI machine vision and Auditory applications. It has a variety of hardware acceleration units such as KPU, FPU, FFT, etc. and the total computing power can be up to 1TOPS, which can conveniently realize machine vision/hearing algorithms of various application scenarios, and can also perform pre-processing work of voice scanning and voice data output.

Users can use MaixCube to easily build a face recognition access control system.

Features:

All-in-One Platform with Rich Peripherals
K210 RISC-V 64-bit Dual-Core CPU for Powerful AI Applications
Computing Power up to 1TOPS for Heavy Machine Vision Applications
Built-in FPU, KPU, FFT Hardware Acceleration Units
Built-in APU for High-Quality Audio Processing

A Wide Range of Sipeed Modules (SP-MOD) Supported:

Through the SP-MOD interface on the MaixCube which is a standard 2.54mm 2x4 Pin female header, with Vcc, GND, and 6 signals, you can connect to various SP-MOD modules and extend your AI projects further.

Handle Module
USB In-line Module
Microphone Array Adapter Board
Weather Station Module
PSRAM Module
Credit Card Module
Grove Adapter Board
SP-MOD One-to-Multi-Adapter Board
1.14 inch LCD module
Single point ToFu Module
FPC Extension Module
Bluetooth Module
Ink Screen Module
LoRa Module

Specifications:

Kendryte K210:

Core : RISC-V Dual-Core 64-bit, with FPU
Frequency : 400MHz (overclockable to 600MHz)
SRAM : Built-in 8M Byte
Image recognition : QVGA @ 60fps / VGA @ 30fps
Voice recognition : Microphone array (8mics)
Network Model : supports YOLOv3/TinyYOLOv2/face recognition, etc.
Deep learning framework : Support frameworks such as TensorFlow / Keras / Darknet / Caffe
Peripherals : FPIOA, UART, GPIO, SPI, PC, PS, TIMER
Hardware acceleration unit :
- KPU convolution operation accelerator
- FPU floating point accelerator
- APU Audio processor
- FFT Fourier transform accelerator

MaixCube Module:

Onboard Peripherals :
- Button x 3
- Camera x 1
- RGB LED x 2
- 1.3 inch TFT screen
- Electret Microphone x 1
- 128Mbit Flash x 1
- Accelerometer x 1
Onboard Interfaces :
- USB Type-C Interface
- Audio interface (supports external speakers)
- TF card slot
- Grove standard interface
- SP-MOD interface (supports SP-MOD interface module)
Power supply : USB-Type-C / Internal lithium battery (200mAh)