Method and device for decomposing, recombining and playing audio data

A method for playing recombined audio data, comprising:

• providing, to one or more processors, mixed input data, the mixed input data being obtained from mixing a plurality of source tracks;
• processing, by the one or more processors, the mixed input data by an artificial intelligence (AI) system comprising a neural network trained to separate audio data of predetermined timbres from the mixed input data, wherein the mixed input data are processed by the AI system to obtain a group of decomposed tracks comprising at least a first decomposed track representing audio signals of a first predetermined timbre, a second decomposed track representing audio signals of a second predetermined timbre different from the first predetermined timbre, and a third decomposed track representing audio signals of a third predetermined timbre different from the first predetermined timbre and the second predetermined timbre;
• reading, by the one or more processors, a control input, the control input representing a setting of a first volume level and of a second volume level;
• recombining, by the one or more processors, at least a first selected track and a second selected track selected from the group of decomposed tracks comprising the first decomposed track, the second decomposed track, and the third decomposed track, to generate a first recombined track;
• recombining, by the one or more processors, the first recombined track at the first volume level with at least a third selected track from the group of decomposed tracks at the second volume level to obtain a second recombined track;
• and
• playing, via one or more speakers coupled to the one or more processors, the recombined audio data based on the second recombined track.