Positioning method for specific sound source

The invention provides a positioning method of a specific sound source. Sound wave signals are collected at a plurality of positions on a preset path through a sensor. A pre-treatment is performed on the acoustic signal to obtain a plurality of signal characteristics. The signal characteristics are taken as input of a deep learning model, and signal identification is performed by using the deep learning model to obtain a plurality of specific audio signals of each position. The specific audio signals obtained at the same position are subjected to autocorrelation function operation to obtain a plurality of autocorrelation coefficients. A representative value is selected from the autocorrelation coefficients as a representative coefficient corresponding to each position. And a specific sound source position is found out according to the representative coefficient of each position.