Method, system and artificial neural network

A system for an interactive entertainment, the system comprising:

• artificial neural network circuitry implementing an artificial neural network, the artificial neural network circuitry being configured:
• obtain a target spectrum from an audio signal produced by a target instrument;
• obtain a plurality of non-target spectra;
• generate a mixture spec by multiplying each of the target spectrum and the plurality of non-target spectra by different weights and aggregating the target spectrum and the plurality of non-target spectra, after multiplication by the different weights, by the artificial neural network circuitry;
• train the artificial neural network circuitry with the mixture spectrum as input and the target spectrum as desired output of the artificial neural network circuitry;
• receive an input spectrum representing a polyphonic audio including multiple tracks;
• and
• process the input spectrum to obtain an output spectrum using the trained artificial neural network circuitry by suppressing the target spectrum of the target instrument in one of the multiple tracks in the polyphonic audio, and preserving the rest of the multiple tracks as unaltered, wherein
• the target spectrum is suppressed, using the artificial neural network of the trained artificial neural network circuitry, incompletely to be only reduced to a defined threshold, and
• wherein the system further determines a morphing spectrum based on the audio signal produced by the target instrument and a replacement spectrum, and uses the morphing spectrum as the target spectrum when training the artificial neural network circuitry.

A non-transitory computer readable medium including executable instructions, which when executed by a computer cause the computer to execute a method for an interactive entertainment performed by artificial neural network circuitry implementing an artificial neural network, the method comprising:

• obtaining a target spectrum from an audio signal produced by a target instrument;
• obtaining a plurality of non-target spectra;
• generating a mixture spectrum by multiplying each of the target spectrum and the plurality of non-target spectra by different weights and aggregating the target spectrum and the one or more non-target spectra, after multiplication by the different weights, by the artificial neural network circuitry;
• training the artificial neural network circuitry with the mixture spectrum as input and the target spectrum as desired output of the artificial neural network circuitry;
• receiving an input spectrum representing a polyphonic audio including multiple tracks;
• and
• processing the input spectrum to obtain an output spectrum using the trained artificial neural network circuitry by suppressing the target spectrum of the target instrument in one of the multiple tracks in the polyphonic audio, and preserving the rest of the multiple tracks as unaltered, wherein
• the target spectrum is suppressed, using the artificial neural network of the trained artificial neural network circuitry, incompletely to be only reduced to a defined threshold, and
• the method further comprises:
• determining a morphing spectrum based on the audio signal produced by the target instrument and a replacement spectrum:
• and
• using the morphing spectrum as the target spectrum when training the artificial neural network circuitry.