Edge Device Dev Board Uses Machine Learning for Voice and Gesture Recognition
SparkFun Electronics® released the SparkFun Edge, a new development board powered by Apollo3 from Ambiq, which is now shipping. SparkFun created SparkFun Edge in collaboration with Google’s TensorFlow Lite team to produce new tools for developers to bring voice and gesture recognition to edge devices.
The Apollo3 from Ambiq uses a Cortex M4 processor with 384KB of RAM and 1MB of Flash storage, requiring extremely low levels of power. The low power consumption allows the SparkFun Edge to run for several days on a coin cell battery.
SparkFun Founder, Nathan Seidle, and Google Engineer, Pete Warden, will host a live stream at 11 a.m. Mountain Time on Friday, April 5, 2019 at SparkFun’s YouTube channel for developers to ask questions and learn the story behind the SparkFun Edge.
“Academia has been talking about artificial intelligence for decades, and while the field has made amazing strides, there haven’t been many tangible effects for general users. With exciting new tools like TensorFlow for machine learning, and increasing processing power, we are starting to see improvements that we can begin to reap benefits from,” said Seidle.
“We wanted to show examples of TensorFlow running on embedded systems at a super-low power—we needed a tool to do that. We talked with SparkFun and Ambiq and now we are able to share this device with the world,” said Pete Warden.
Ambiq Micro’s latest Apollo3 Blue microcontroller, an ultra-efficient ARM Cortex-M4F 48MHz (with 96MHz burst mode) processor, runs TensorFlow Lite using merely 6µA/MHz.
Apollo3 Blue features six configurable I2C/SPI masters, two UARTs, one I2C/SPI slave, a 15-channel 14-bit ADC, and a dedicated Bluetooth processor that supports BLE5. On top of that, the Apollo3 Blue has 1MB of flash and 384KB of SRAM memory – plenty for the vast majority of applications.
On the SparkFun Edge developers will have built-in access to sensors, I2C expansion, Bluetooth, and GPIO inputs/outputs. To support edge computing cases such as voice recognition the Edge board features two MEMS microphones, an ST LIS2DH12 3-axis accelerometer on its own I2C bus, and a connector to interface to an OV7670 camera (sold separately).
An onboard Bluetooth antenna gives the Edge out-of-the-box connectivity.
Also available on the board is a Qwiic® connector that adds I2C sensors/devices, four LEDs, and four GPIO pins.
The board is outfitted with battery operation from the CR2032 coin cell holder to boast its low-power capabilities.
An external USB-serial adapter like the Serial Basic Breakout via a serial bootloader enables programming of the board, but for more advanced applications the board also includes JTAG programming and a debugger port.
As an entirely open-source project, developers can take a look at the code and hardware files on GitHub and at the SparkFun website.
- 32-bit ARM Cortex-M4F processor with Direct Memory Access
- 48MHz CPU clock, 96MHz with TurboSPOT™
- Extremely low-power usage: 6uA/MHz
- 1MB Flash
- 384KB SRAM
- Dedicated Bluetooth processor with BLE 5
- ST LIS2DH12 3-axis accelerometer
- 2x MEMS microphones with operational amplifier
- OV7670 camera connector
- Qwiic connector
- 4 x GPIO connections
- 4 x user LEDs
- 1 x user button
- FTDI-style serial header for programming
- Bluetooth antenna
- CR2032 coin cell holder for battery operation
What It Does
- High processing to current consumption ratio enables machine learning applications on the ‘Edge’ of networks, without the need for a central computer or web connection.
- Voice, gesture, or image recognition possible with TensorFlow Lite. (Note: Voice examples are provided. Gesture and image examples hope to be released by TensorFlow soon)
- 1.8V to 3.6V supply voltage range
- Small 1.6in x 1.6in x 0.35in (40.6mm x 40.6mm x 8.9mm) form factor