Provided are an electronic device for recognizing speech of a user, and a method, performed by the electronic device, of recognizing speech. The method includes obtaining an audio signal based on a speech input based on the audio signal being input, obtaining an output value of a first automatic speech recognition (ASR) model that outputs a character string at a first level; obtaining an output value of a second ASR model that outputs a character string at a second level corresponding to the audio signal based on the output value of the first ASR model based on the audio signal being input; and recognizing the speech from the output value of the second ASR model.