Cough Detection System Based on ASR-HMM
Fatima Barkani, Hassan Satori, Mohamed Hamidi · 2020
In this paper, we propose a new approach to design a cough detection system based on speech-recognition algorithms. Our system is implemented with Kaldi opensource platform by Gaussian Mixture Model-based Hidden Markov Model (GMM-HMM) hybrid system through a simple Monophone training model. Also, a comparison between the Perceptual Linear Prediction (PLP) and Mel Frequency Cepstral Coefficient (MFCC) feature extraction methods is presented. Our proposed system can be used as a collection platform to collect naturally and spontaneous cough data from conversation or continuous speech. The system achieved the best performance when trained using the MFCC feature.