Method and device for decomposing, recombining and playing audio data

A device for processing and playing audio data, comprising:

• an audio input unit for providing mixed input data, wherein the mixed input data are obtained from mixing a plurality of source tracks;
• an artificial intelligence (AI) system comprising a neural network trained to separate audio data of predetermined timbres from mixed audio data, wherein the AI system is configured to receive and process the mixed input data and to generate 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;
• a controlling section configured to generate a control input representing a setting of a first volume level and a second volume level;
• a recombination unit configured to:
• recombine at least the first selected track and the second selected track selected from the group of decomposed tracks to generate a first recombined track;
• and
• recombine the first recombined track at the first volume level with at least a third track selected from the group of decomposed tracks, at the second volume level, to obtain a second recombined track;
• and
• a playing unit configured to play audio data based on the second recombined track.

A method for processing and playing audio data, comprising:

• receiving mixed input data, the mixed input data being a sum signal obtained from mixing at least one first source track with at least one second source track;
• decomposing the mixed input data to obtain at least a first decomposed track resembling the at least one first source track;
• generating output data based on the first decomposed track;
• playing the output data through an audio output;
• reading a control input, the control input representing a setting of a first volume level of the first decomposed track and a second volume level of a second track, wherein the second track is an independent track obtained from second mixed input data;
• recombining at least the first decomposed track at the first volume level with the second track at the second volume level to generate recombined output data;
• and
• playing the recombined output data.

A device for processing and playing audio data, comprising:

• an audio input unit for receiving mixed input data, the mixed input data being a sum signal obtained from mixing at least a first source track with at least a second source track;
• a decomposing unit connected to the audio input unit for decomposing the mixed input data to obtain at least a first decomposed track resembling the first source track;
• a playing unit for playing output data based on the first decomposed track;
• and
• a recombination unit for recombining at least the first decomposed track with a second track to generate the output data for the playing unit, wherein the second track is an independent track obtained from second mixed input data.

The method of claim 31, further comprising recombining at least the second segment of the first decomposed track at the first volume level with the second segment of the second decomposed track at the second volume level to generate a second segment of the recombined output data,

• wherein at least one of the following steps is performed while playing the first segment of the output data:
• extracting the second segment from the input audio file,
• decomposing the second segment of the input audio file, or
• recombining the at least the second segment of the first decomposed track at the first volume level with the second segment of the second decomposed track at the second volume level, and
• wherein generation of the second segment of the output data is completed before playing the first segment of the output data is completed.