An Analysis of Audio Features to Develop a Human Activity Recognition Model Using Genetic Algorithms, Random Forests, and Neural Networks

Table 3

Audio clips metadata per activity.

Activity	Sample rate	Encoding format	Channels
Brew coffee	8,000 Hz–44,100 Hz	m4a, amr	Stereo, Mono
Cooking	44,100 Hz	m4a	Stereo
Use microwave oven	44,100 Hz	m4a	Stereo
Take a shower	44,100 Hz	m4a, mp3	Stereo
Dish washing	44,100 Hz	m4a	Stereo
Hand washing	8,000 Hz–44,100 Hz	m4a, amr	Stereo, Mono
Brushing teeth	44,100 Hz	m4a	Stereo
No activity	8,000 Hz–44,100 Hz	m4a, amr	Stereo, Mono