Table 13.

Evaluation of RPi using TFLite and post-training quantization with 8-bit on the Urban Sound dataset.

Models	Inference Time (ms)	${\mathit{F}}_1$ Micro	${\mathit{F}}_1$ Macro	${\mathit{F}}_1$ Weighted
ANN	0.089 ± 0.008	0.954	0.964	0.954
CNN-1D	21.442 ± 0.365	0.952	0.961	0.952