Generation of multichannel audio signal and data signal representing a multichannel audio signal

An apparatus comprising:

• a downmixer circuit,
• wherein the downmixer is arranged to generate a downmix audio signal by downmixing a multichannel audio signal,
• wherein the downmixer is arranged to generate upmix parametric data for upmixing of the downmix audio signal;
• an artificial neural network,
• wherein the artificial neural network is arranged to generate a plurality of control data values,
• wherein the plurality of control data values is arranged to change a processing,
• wherein the processing is arranged to generate an auxiliary audio signal for upmixing from the downmix audio signal,
• wherein the artificial neural network comprises output nodes and input nodes,
• wherein the output nodes are arranged to provide the plurality of control data values,
• wherein the input nodes are arranged to receive samples of at least one of the downmix audio signal and the multichannel audio signal;
• a generator circuit,
• wherein the generator circuitry is arranged to generate a data signal,
• wherein the data signal represents the multichannel audio signal,
• wherein the data signal comprises the downmix audio signal, the upmix parametric data, and the plurality of control data values.

A method comprising:

• receiving a data signal, the data signal comprising:
• a downmix audio signal for a multichannel audio signal;
• upmix parametric data for upmixing the downmix audio signal;
• and
• a plurality of control data values for adapting a processing so as to generate an auxiliary audio signal for upmixing from the downmix audio signal;
• receiving samples of the downmix audio signal using the input nodes of an artificial neural network;
• providing samples of the auxiliary audio signal using the output nodes of the artificial neural network,
• wherein nodes are arranged to comprise contributions from the plurality of control data values,
• wherein the artificial neural network comprises the nodes;
• and
• generating the multichannel audio signal from the downmix signal and the auxiliary audio signal based on the upmix parametric data.

A method comprising:

• generating a downmix audio signal by downmixing a multichannel audio signal;
• generating upmix parametric data for upmixing of the downmix audio signal;
• generating a plurality of control data values using an artificial neural network,
• wherein the plurality of control data values is arranged to change a processing,
• wherein the processing is arranged to generate an auxiliary audio signal for upmixing from the downmix audio signal;
• providing samples of at least one of the downmix audio signal and the multichannel audio signal to input nodes of the artificial neural network;
• and
• receiving the plurality of control data values from output nodes of the artificial neural network;
• and
• generating the data signal to comprise the downmix audio signal, the upmix parametric data, and the plurality of control data values.