Skip to main content
Fig. 6 | EURASIP Journal on Audio, Speech, and Music Processing

Fig. 6

From: A depthwise separable convolutional neural network for keyword spotting on an embedded system

Fig. 6

Selection of hyper-parameter grid search. The first line in each box indicates the average accuracy in the range of 0−20 dB SNR for the matched and mismatched (in brackets) test. The number in the second line shows the number of operations and the required memory (in brackets). The third line indicates the execution time. The color indicates the accuracy (lighter color lower accuracy)

Back to article page