AudioNN + ObjDet about ambpro2_arduino HOT 3 CLOSED

Hi @rkuo2000,

To be able to run Audio NN and Object Detection together, you have to combine both of the examples. I have combined the code (please refer to the attached zip folder), but I have not tried it. However, as the YAMNet model is big, it is impossible to run these 2 models together on the board.

You will get this error " Sketch too big; see https://support.arduino.cc/hc/en-us/articles/360013825179 for tips on reducing it.
text section exceeds available space in board"

To note: modelSelect function in the code is used to select NN task and the model used. Therefore, you need to call this function to be able to upload the neccessary model into the board.

Hence, you have to retrain the YAMNet model to reduce the size. You may refer to https://github.com/tensorflow/models/tree/master/research/audioset/yamnet.

ObjDetAudioNN.zip

from ambpro2_arduino.

This issue is stale because it has been open for 14 days with no activity.

from ambpro2_arduino.

Loading NN model using SD card feature is up. You can give it a try using early release v4.0.7-build20240313

You may refer to this YouTube video for tutorial: https://www.youtube.com/watch?v=cVvdnXiCAa4

Below is the updated code.
ObjectDetectionAudioClassification.zip

from ambpro2_arduino.