http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Supervised learning에 기반한 돼지 발성음 분류에 관한 연구
민경진(KyoungJin Min),이혁재(HyeokJae Lee),황현진(HyunJin Hwang),이상엽(SangYeob Lee),이강휘(KangHwi Lee),문상호(SangHo Moon),이정은(JungEun Lee),이정환(Jeong Whan Lee) 대한전기학회 2021 전기학회논문지 Vol.70 No.5
This study categorizes the current situation of the pig as supervised learning through the analysis of the pig’s sound. Audio data were obtained from video data obtained by recording at a barn. Speech data was preprocessed to extract features in the time domain and frequency domain, and formants and MFCC were extracted in the frequency domain. Decision Tree, K-Nearest Neighbors, and Support Vector Machine were used for classification, and linear and RBF kernels were used for SVM. The experiment was conducted two times: classifying using features used in Praat and MDVP, which are speech analysis programs, and classifying using MFCC used in speech recognition. After classification, k-fold verification was performed. As a result of the experiment, it can be seen that there is a difference in classification according to the characteristics of using the same voice, and in the case of a situation in which the sound is unified, such as ‘cough,’ the judgment of any classifier is clear. However, in other situations, it is considered necessary to consider the characteristics of pigs by further observing their socialization behavior.