This paper describes the integration of a spoken dialogue system and a nursing care recording on an Android smartphone application intending to help nurses reduce doc-umentation time and improve the overall experience of a healthcare setting. We develop a machine learning model based on a bidirectional long-short term memory and condi-tional random fields (Bi-LSTM-CRF) to extract record details from unstructured utterances and transform them into record inputs on the application. The model achieves the highest F1 score at 98%. We also conduct a preliminary experiment and demonstrate its powerful ability to record, in which the speed increases by 75% compared to a traditional keyboard-based.