A wearable device capable of recognizing dozing-off stage including a processor and an electrocardiogram sensor is provided. The processor trains a neural network module. The processor is coupled to the electrocardiogram sensor. The electrocardiogram sensor is configured to generates an electrocardiogram signal. The processor executes a heart rate variability analysis operation and R-wave amplitude analysis operation to analyze R-R interval variation of the electrocardiogram signal to generate a plurality of characteristic values. The processor utilizes the trained neural network modules to perform a dozing-off stage recognition operations according to the characteristic values to obtain a dozing-off stage identification result. In addition, a recognition method is also provided.本發明提出一種可辨識打瞌睡階段的穿戴式裝置包括處理器以及心電感測器。處理器用以訓練類神經網路模組。處理器耦接心電感測器。心電感測器用以產生心電圖信號。處理器執行心律變異分析操作以及R波振幅分析操作來分析心電圖信號的心跳間期變化,以產生多個特徵值。處理器利用經訓練後的類神經網絡模組來依據這些特徵值執行打瞌睡階段辨識操作,以取得打瞌睡階段辨識結果。另外,一種辨識方法亦被提出。