The embodiment of the invention provides a biological recognition method and device, and relates to the field of artificial intelligence. According to the embodiment of the invention, the identification success rate when the user identity is identified by utilizing various biological characteristics can be improved. The method comprises the following steps: acquiring characteristic parameters of nbiological characteristics of a to-be-identified user, wherein n is greater than or equal to 2; Utilizing an LSTM (Long Short Term Memory) model to fuse the characteristic parameters of the n biological characteristics respectively to generate high-order characteristic parameters corresponding to the n biological characteristics respectively; calculating comprehensive characteristic parameters ofthe n biological characteristics by utilizing an attention mechanism according to the high-order characteristic parameters corresponding to the n biological characteristics; and according to the comprehensive characteristic parameters of the n biological characteristics, identifying the identity of the to-be-identified user. The method is applied to identity recognition.