Voice decoding method and system, storage medium and terminal

The invention provides a voice decoding method and system, a storage medium and a terminal, and the method comprises the steps: inputting to-be-decoded audio data into an acoustic model, obtaining a plurality of groups of initial decoding results through the acoustic model, and enabling the initial decoding results to comprise an initial decoding sequence and an initial likelihood score; inputting the audio data to be decoded into an unvoiced and voiced sound classification model to obtain unvoiced and voiced sound sequence information, the unvoiced and voiced sound sequence information comprising first syllable information and first moment information; and screening each group of initial decoding results according to the first syllable information and the first moment information to obtain a target decoding result, and completing decoding. According to the invention, the accuracy of voice decoding in a complex voice environment can be improved. 本发明提供了一种语音解码方法、系统、存储介质及终端，所述方法包括：将待解码的音频数据输入至声学模型，通过所述声学模型获取若干组初