Question/Issue:
Is it possible to apply more than one processing block to my data and use the resulting features of the final block to feed the learning block?
For example, my goal is to train a model using my sensor data, which is in the time domain. First, I want to use the spectral analysis processing block to apply a digital filter and normalize each sample. Then, I was wondering if it is possible to use these filtered and normalized data from the spectral analysis block as an input for a MFCC processing block, and finally used the features extracted form the MFCC block to feed the learning block.
Is this possible to achieve within the edge impulse UI, or should I build a custom block to merge both spectral analysis and MFCC blocks?