The present disclosure provides a method, an apparatus, a wearable device and a storage medium for audio playback, and relates to the fields of machine learning, deep learning, intelligent wearable, etc. The method includes: measured vital sign data of a user within a first time range is obtained; a vital sign prediction is performed based on the measured vital sign data within the first time range, to obtain predicted vital sign data of the user within a second time range after the first time range; one or more target audio parameters are determined based on the predicted vital sign data and user attribute information of the user; and matched target audio data is obtained based on the one or more target audio parameters.