Audio recognition method and device and data processing equipment

The application discloses an audio recognition method and device and data processing equipment. The method comprises the following steps: acquiring to-be-recognized audio, and splitting the to-be-recognized audio into a plurality of audio segments; for each audio segment, obtaining the feature of the audio segment; inputting the feature of the audio segment into a trained classification model forclassification recognition to obtain whether the audio segment is a classification result of a suspected specific audio segment or not; and inputting the classification result of the plurality of audio segments into a long short-term memory network LSTM network for processing to obtain whether the to-be-recognized audio is the recognition result of the specific audio or not. Compared with the modethat the whole audio is integrally recognized in the prior art, the scheme provided by the embodiment can more accurately recognize the specific audio content with small proportion in the to-be-recognized audio. 本申请一种音频识别方法、装