In collaboration with Edge Impulse, Syntiant presents a new board for evaluating their NDP101 Neural Decision Processor™. Unlike the NDP9101B0 development platform, which is implemented as a Raspberry Pi shield, the ‘Tiny Machine Learning’, a.k.a. TinyML, board is self-contained and smaller than a smart-watch display (24 mm x 28 mm), providing an incredibly compact, low-power edge-AI inference system for audio and motion; a BMI160 6-axis motion sensor definitely gears this board more towards wearables than the purely speech-oriented NDP9101B0.
I won’t go into detail again about the NDP101 – see details of this chip in the application form or previous post – but if I could sum up its benefits, I would bullet-point it as so:
-> Deep learning algorithms for speech recognition
-> At-memory computation for 20x more throughput than stored program architectures
-> 200x less energy per inference than stored program architectures
-> 32-pin QFN SoC with dual inputs for PDM microphone or PCM-over-SPI
-> Embedded Arm Cortex-M0 with 112 kB SRAM
-> Active power consumption of <140 µW while recognising words
Bosch! But enough about that 6-axis motion sensor – Let’s continue to the TinyML!
Syntiant’s TinyML development board is the ideal platform for building low-power voice, acoustic event detection (AED), and sensor ML applications. This new developer kit provides stand-alone access to Syntiant’s technology for anyone who wants to use it, test it, and design with it, and contains a host processor, the SAMD21 Cortex-M0+ 32-bit, low-power ARM MCU with 256 kB flash memory and 32 kB host processor SRAM, instead of requiring an additional board. This infrastructure is supported by 2 MB of on-board serial flash and a 48 MHz system clock. With a 32 GB micro-SD card, not included, one can store >3 days of uncompressed audio data (Fs = 16 kHz) and >300 days of 6-axis IMU sensor data (Fs = 100 Hz), making the TinyML an idea platform for data collection.
The TinyML operates on 5 V over micro-USB or 3.7 V via LiPo battery. The 5 digital I/Os are compatible with Arduino MKR series boards, and there is a UART and I2C interface included in these pins. An onboard SPH0641LM4H microphone and BMI160 sensor enable easy configuration for any speech, AED, or 6-axis motion-and vibration-related application, and there is also a user-defined RGB LED to serve as an immediate visual HMI during operation. Trained models can be easily downloaded on the TinyML board through a micro-USB connection without the need for any specialised hardware, and the board is currently shipping with an ‘Alexa play music’ model.
A pretty cool update, right?
(Image sourced from Syntiant)