课程: Machine Learning Foundations: Prototyping with Edge Impulse

Running the model - Arduino教程

课程: Machine Learning Foundations: Prototyping with Edge Impulse

开始一个月试用 购买团队方案

Running the model

“

- [Instructor] Now that you've deployed an Arduino library from Edge Impulse, it's time to upload it and see if the prototype works on the development board. For this, we're back in the Arduino IDE. To install the Edge impulse library, go to Sketch, Include Library, Add Zip Library, and choose the zip file you downloaded earlier. When you see the library installed message, you can review the added examples by going to File, Examples, Voice Commander Inference. Looking under the Arduino Nano 33 BLE Sense Menu, there are several examples. We're interested in the ones using the microphone. One microphone example monitors input continuously. The other example samples in discreet windows. For our prototype, we want to be listening continuously, so choose the continuous microphone example. Make sure to select your board if it's not already selected. Now compile and upload the program. This can take some time. Once the example is completed uploading, to see the output, we need to use the serial monitor, which will print out information coming over the USB connection from the development board. You can open the serial monitor by clicking the control at the top of the Arduino IDE window. The example's monitoring sounds coming through the microphone. When you say your keyword, you should see its percentage increase. Shield up, shield up, shield up, shield up. You can see that when you say shield up, its percentage goes up and that confirms that inferencing is working on the development board. The next step is to make some modifications to the sample code to get closer to our goal of locking the screen when you say shield up.

内容