Method and unit for performing dynamic range control
A dynamic range control unit ( 210 ) configured to apply dynamic range control, referred to as DRC, to an audio signal ( 211 ). The DRC unit ( 210 ) is configured to downsample a subband signal ( 212 ) derived from the audio signal ( 211 ), to provide a downsampled subband signal ( 321 ), to determine a DRC gain ( 329 ) based on the downsampled subband signal ( 321 ), and to apply the DRC gain ( 329 ) to the subband signal ( 212 ), to provide a compressed subband signal ( 213 ) of a compressed audio signal ( 214 ).
1. A dynamic range control, referred to as DRC, unit configured to apply DRC to an audio signal; wherein the DRC unit is configured to
downsample a subband signal derived from the audio signal, to provide a downsampled subband signal;
determine a DRC gain based on the downsampled subband signal; and
apply the DRC gain to the subband signal, to provide a compressed subband signal of a compressed audio signal.
2. The DRC unit of claim 1 , wherein
a frame of the subband signal comprises N time slots and a corresponding frame of the downsampled subband signal comprises N/M time slots; and
M>1 is a downsampling factor.
3. The DRC unit of claim 2 , wherein the DRC unit is configured to
determine N/M DRC gains for the corresponding N/M time slots of the downsampled subband signal, based on the downsampled subband signal; and
upsample the N/M DRC gains to N DRC gains for the corresponding N time slots of the subband signal.
4. The DRC unit of claim 3 , wherein the DRC unit is configured to multiply N samples for the corresponding N time slots of the subband signal with the corresponding N DRC gains to determine the compressed subband signal, and/or wherein upsampling comprises
repeatedly using the same DRC gain for M subsequent time slots; or
interpolating between two successive DRC gains from the N/M DRC gains to determine M−1 DRC gains for M−1 subsequent time slots between the two successive DRC gains from the N/M DRC gains.
5. The DRC unit of claim 1 , wherein the DRC unit is configured to
downsample the subband signal without making use of a lowpass filter; and/or
downsampled subband signal by selecting every M th sample of the subband signal as a sample of the downsampled subband signal.
6. The DRC unit of claim 1 , wherein
the audio signal is represented using K subband signals for K different frequency bins, with K>1; and
the DRC unit is configured to
group at least some of the K different frequency bins into Q DRC bands, with Q<K; and
determine a DRC gain for at least one of the Q DRC bands, based on the subband signals assigned to the at least one of the Q DRC bands, respectively.
7. The DRC unit of claim 6 , wherein the DRC unit is configured to ignore one or more of the K frequency bins when determining the DRC gain for the Q DRC bands.
8. The DRC unit of claim 1 , wherein the DRC unit is configured to
determine a loudness estimate for a sample of the downsampled subband signal; and
determine the gain value based on the loudness estimate using a compression curve.
9. The DRC unit of claim 8 , wherein the compression curve is defined by one or more parameters comprised within profile data, and/or wherein the DRC unit is configured to
determine a magnitude of the sample of the downsampled subband signal; and
apply Steven's power law and a conversion from a linear-domain to a logarithmic-domain to the magnitude of the sample of the downsampled subband signal to determine the loudness estimate for the sample of the downsampled subband signal.
10. The DRC unit of claim 8 , wherein
the compression curve is configured to provide a mapped gain value in dependence of a deviation of the loudness estimate from an operating point; and
the operating point is dependent on a dialnorm parameter or on an average dialog level of dialog content within the audio signal or on an average power level of the audio signal.
11. The DRC unit of claim 10 , wherein the DRC unit is configured to
determine an average block power value based on power values of a block of samples of the downsampled subband signal; and
determine the operating point for the block of samples of the downsampled subband signal also based on the average block power value.
12. The DRC unit of claim 11 , wherein the DRC unit is configured to
determine a long-term average block power value for the block of samples of the downsampled subband signal based on the average block power value for the block of samples of the downsampled subband signal and a sequence of previous blocks of samples of the downsampled subband signal;
determine a power deviation of the long-term average block power value from the dialnorm parameter; and
determine the operating point for a sample from the block of samples of the downsampled subband signal based on the power deviation.
13. The DRC unit of claim 1 , wherein the DRC unit is configured to
determine a mapped gain value for a sample of the downsampled subband signal for a current time slot based on a loudness estimate of the loudness of the sample of the downsampled subband signal for the current time slot;
perform gain smoothing of the mapped gain values for samples of the downsampled subband signal for the current time slot and for one or more previous time slots, to determine a smoothed gain value 326 ; for the sample of the downsampled subband signal for the current time slot; and
determine the DRC gain for the sample of the downsampled subband signal for the current time slot based on the smoothed gain value for the sample of the downsampled subband signal for the current time slot.
14. The DRC unit of claim 1 , wherein the DRC unit is configured to
determine a DRC gain for a sample of the downsampled subband signal; and
apply the DRC gain for the sample of the downsampled subband signal to a plurality of successive samples of the subband signal to determine the compressed subband signal.
15. The DRC unit of claim 1 , wherein
the audio signal comprises a plurality of different channels; and
the DRC unit is configured to determine a joint DRC gain for the plurality of different channels or for one or more subgroups of the plurality of different channels.
16. The DRC unit of claim 15 , wherein the DRC unit is configured to
select one of the channels from the plurality of different channels; and
determine a DRC gain for the selected channel as the joint DRC gain for the plurality of different channels or for the one or more subgroups of the plurality of different channels.
17. The DRC unit of claim 1 , wherein the DRC unit is configured to
determine and apply a plurality of DRC gains for a corresponding plurality of subband signals to provide a corresponding plurality of compressed subband signals; and
determine the compressed audio signal by applying an inverse transform or a synthesis filter bank to the plurality of compressed subband signals; wherein the inverse transform or synthesis filter bank are complementary to a forward transform or analysis filter bank used for deriving the subband signal from the audio signal.
18. A decoding unit configured to derive a dynamic-range compressed audio signal from a bitstream indicative of an audio signal;
wherein the decoding unit comprises the DRC unit according to claim 1 , configured to apply dynamic range control to the audio signal.
19. A method for applying dynamic range control, referred to as DRC, to an audio signal; wherein the method comprises
downsampling a subband signal derived from the audio signal, to provide a downsampled subband signal;
determining a DRC gain based on the downsampled subband signal; and
applying the DRC gain to the subband signal, to provide a compressed subband signal of a compressed audio signal.
20. A non-transitory computer-readable storage medium comprising instructions which, when executed by a processor cause the processor to perform the method of claim 19 .