This despository is used for judging why baby wa crying? In the early statge, the cry of baby has some common reasones. Under this background, we collect and slect some database that includes discomfort, sleep, hunger, wakefulness, and hug five types. Finally ,we develop a method which has good accuarcy in baby crying reason judgement.
-
Download some cry data: Most of all the data have been proved by chris.
-
Download this code despository. put database into wavs_all_cut. Test data into data/test.
-
developing python environment for train.
-
run WriteMfcc2CSV.py for train data generate.
-
run ModelTrain.py for model train.
-
run predict.py for test prediction.
-
get the result.
note : due time reason ,some complex code is not in here. We give the DataAugument.py and some reference paper for your review.
-
aniel S Park, William Chan, Yu Zhang, Chung-Cheng Chiu, BarretZoph, Ekin D Cubuk, and Quoc V Le. Specaugment: A simple dataaugmentation method for automatic speech recognition.arXiv preprintarXiv:1904.08779, 2019.
-
Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz.mixup: Beyond empirical risk minimization.arXiv preprintarXiv:1710.09412, 2017
-
hi Hua Zhou. Ensemble learning.Encyclopedia of Biometrics, pages270–273, 2009.