Abstract:
A person's eating behavior, such as, chewing and swallowing can represents the health conditions of that person. From that viewpoint, we were trying to develop a system t...Show MoreMetadata
Abstract:
A person's eating behavior, such as, chewing and swallowing can represents the health conditions of that person. From that viewpoint, we were trying to develop a system that can monitor a person's chewing and swallowing, simply as well as in detail by using sound information recorded near the oropharynx. However, its accuracy was not good enough because of lack of well-labeled training data. In this paper, we propose a technique to gather a large amount of weak-labeled data easily, and leverage it to train the Long Short Term Memory-Connectionist Temporal Classification (LSTM-CTC) model for estimating the number of chewing and swallowing.
Date of Conference: 15-18 October 2019
Date Added to IEEE Xplore: 27 February 2020
ISBN Information:
Print on Demand(PoD) ISSN: 2378-8143