Self-abnormality has been generally accepted as the basic characteristics of depression in psychology. Many studies have shown that self-face recognition (SFR) can help us understand self-abnormality of depressed individuals. In contrast, other distinct levels of self-recognition/perception, such as self-speech recognition (SSR), are relatively understudied. Herein, 12 high depression-inclined individuals (HDII) and 15 healthy subjects were recruited to participate in this study. We measured the degree of self-processing abnormality in HDII during different speech discrimination tasks relative to controls. We observed that a significant main effect of group (F (1, 25) = 4.24, P =0.05, η2=0.027) by results of analysis of variance (ANOVA). Subsequent independent t-tests further revealed self-processing bias (t=2.633, P=0.012) and self-recognition bias (t=2.019, P=0.035) observed in selffamiliar speech recognition task and self-stranger speech recognition task respectively. It shows that the result of HDII was significantly higher than the controls. Based on selfspeech recognition, a new methodology for detecting depression was developed by exploring whether the self was impaired in HDII. Experimental results are encouraging for detecting depression, as well as provide a new perspective for future research.