Connect the LEDs to the RTL8722DM_MINI board as shown in the following wiring diagram.

Download the Ameba customized version of TensorFlow Lite for Microcontrollers Library at:

https://github.com/ambiot/ambd_arduino/tree/master/Arduino_zip_libraries. 

Follow the instructions at  https://www.arduino.cc/en/guide/libraries to install it.

Ensure that the patch files found at  https://github.com/ambiot/ambd_arduino/tree/master/Ameba_misc/  are also installed.

To open the example code, Go to "Files"→ "Examples" → "TensorFlowLite_Ameba" → "micro_speech"

Upload the code and press the reset button on the Ameba board once the upload has completed.

Once it is running, you should see the yellow led flashing, which indicates that it is processing the audio. Saying the word "yes" will cause the green led on the board to light up. "no" will cause the red LEd to light up. If the word is not recognised, the blue LED on the board will light up.

The result will be output to the Serial Monitor, which appears like the following picture.

Here are some tips when troubleshooting:

• Ensure that your surroundings are quiet with minimal noise.
• Experiment with varying the distance of the microphone, starting with it at an arm's length
• Experiment with different tones and volume when saying the words.
• Depending on how you pronounce the words, the characteristics of the microphone used, getting one keyword recognized may be easier than the other.