Loudness level and range processing
View Patent ↗Loudness signal processors and methods for processing an input audio signal in order to control a resulting integrated loudness level and a resulting loudness range of an output audio signal by a predetermined target loudness level and by a predetermined target loudness range, the processors and methods comprising level detection and level distribution analysis; transfer function generation based on the level distribution, the predetermined target loudness level and the predetermined target loudness range; and calculation of a gain to apply to said input audio signal, resulting in said output audio signal.
1. A loudness signal processor for processing an input audio signal in order to control a resulting integrated loudness level and a resulting loudness range of an output audio signal by a predetermined target loudness level and by a predetermined target loudness range,
the loudness signal processor comprising
an input arranged to receive an input audio signal;
a level detector block arranged to determine a time-varying level of said input audio signal;
a distribution analyzer block arranged to estimate a time-varying level distribution of said time-varying level;
a transfer function generator block arranged to determine a time-varying transfer function on the basis of said time-varying level distribution, and said predetermined target loudness level and said predetermined target loudness range;
a gain control block arranged to calculate a time-varying gain on the basis of said time-varying level of the input audio signal and said time-varying transfer function;
a multiplier block arranged to apply said time-varying gain to said input audio signal, resulting in said output audio signal,
wherein said transfer function generator block is arranged to determine a loudness level estimated on the basis of said time-varying level distribution by performing an RMS calculation in which the levels below a threshold are excluded, as specified in the calculation of integrated, gated loudness level of ITU-R BS-1770-2 and EBU R-128.
2. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to estimate said time-varying level distribution on the basis of a part of said input audio signal by considering values of said time-varying level corresponding to different times with two or more different weights.
3. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to estimate said time-varying level distribution of said time-varying level by means of a finite-length sliding window.
4. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to estimate said time-varying level distribution of said time-varying level by means of a set of IIR filters.
5. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to estimate the time-varying level distribution of said time-varying level to which said time-varying transfer function has been applied, thus effectively estimate a distribution of a level of said output audio signal.
6. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to maintain at least two instances, one weighing the older time-varying levels lower than the other instance in their respective distribution estimates.
7. The loudness signal processor according to claim 1 , wherein an estimated integrated loudness level over a finite-length window of the output audio signal substantially matches said predetermined target loudness level.
8. The loudness signal processor according to claim 1 , wherein an estimated integrated loudness level over a finite-length window of the output audio signal is substantially constrained by said predetermined target loudness level.
9. The loudness signal processor according to claim 1 , wherein an estimated loudness range over a finite-length window of the output audio signal substantially matches said predetermined target loudness range.
10. The loudness signal processor according to claim 1 , wherein an estimated loudness range over a finite-length window of the output audio signal is substantially constrained by said predetermined target loudness range.
11. The loudness signal processor according to claim 1 , wherein said predetermined target loudness level is specified as a value of integrated, gated loudness level, as standardized in ITU-R BS.1770-2 and EBU R-128.
12. The loudness signal processor according to claim 1 , wherein said predetermined target loudness range is specified as a value of LRA measure, as standardized in EBU Tech. Doc. 3342 and EBU R-128.
13. The loudness signal processor according to claim 1 , wherein said level detector block is arranged to determine said time-varying level of said input audio signal by performing RMS calculation of the input audio signal.
14. The loudness signal processor according to claim 1 , wherein the time-varying level of said input audio signal corresponds to an estimate of a time-varying loudness level of said input audio signal.
15. The loudness signal processor according to claim 1 , wherein said distribution analyzer block is arranged to enable re-initialization.
16. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution, the priming being effective immediately.
17. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution, the priming being effective at a specific relative or absolute time.
18. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution and is arranged to, over time, update the predetermined level distribution based on said time-varying level of the input audio signal.
19. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution and is arranged to, over time, update the predetermined level distribution based on said time-varying level of the input audio signal; wherein two or more predetermined level distributions are stored, each being associated with a certain audio signal type; and wherein the distribution analyzer block is arranged to perform the update only for a predetermined level distribution which is associated with an audio signal type corresponding to a type of the input audio signal.
20. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution; and wherein the loudness signal processor is arranged to let the predetermined level distribution predominantly affect the target loudness range properties of said time-varying transfer function, and the time-varying level distribution predominantly affect the target loudness level properties of said time-varying transfer function.
21. The loudness signal processor according to claim 1 , wherein said distribution analyzer block comprises a priming with a predetermined level distribution which has been generated in advance based on a collection of programs belonging to the same audio signal type that have first been loudness normalized individually, then their time-varying levels measured forming individual level distributions, and then all these distributions combined into the predetermined distribution to be used for priming.
22. The loudness signal processor according to claim 1 , wherein a time-delay block is arranged to delay the input audio signal.
23. The loudness signal processor according to claim 3 , wherein a length of the finite-length sliding window of said distribution analyzer block is determined by a genre or expected content type of the signal being processed.
24. The loudness signal processor according to claim 3 , wherein a length of the finite-length sliding window of said distribution analyzer block is determined corresponding to a duration of a particular program being processed.
25. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to maintain an estimate of a distribution of time-varying level of the output audio signal, based on past said time-varying transfer functions combined with past said time-varying level distributions; and wherein the transfer function generator block is further arranged to determine a time-varying transfer function on the basis of said estimate of the distribution of time-varying level of the output audio signal.
26. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine said time-varying transfer function such that the resulting time-varying gain, for an input audio signal with constant time-varying level, corresponds to the difference between said target loudness level and an estimated integrated loudness level of the input audio signal.
27. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine a loudness level estimated on the basis of said time-varying level distribution by performing an RMS calculation.
28. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine a loudness level estimated on the basis of said time-varying level distribution by performing an RMS calculation in which levels below a threshold level are excluded.
29. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine a loudness level estimated on the basis of said time-varying level distribution by performing an averaging calculation in which higher levels receive a higher weight than lower levels, according to a weighting function.
30. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine a degree of compression of said time-varying transfer function on the basis of a difference between said target loudness range and a distance between two percentiles estimated on the basis of said time-varying level distribution.
31. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine a degree of compression of said time-varying transfer function on the basis of a difference between said target loudness range and a distance between two percentiles estimated on the basis of said time-varying level distribution, as specified in the calculation of the LRA measure of EBU Tech. Doc. 3342 and EBU R-128.
32. The loudness signal processor according to claim 1 , wherein said transfer function generator block is arranged to determine said time-varying transfer function based on one or more heuristic rules.
33. The loudness signal processor according to claim 1 , comprising a preprocessor arranged to preprocess said input audio signal prior to said level detector block.
34. The loudness signal processor according to claim 33 , wherein said preprocessor comprises frequency weighting, preferably a filter applied to the input audio signal.
35. The loudness signal processor according to claim 33 , wherein said pre-processor comprises an A-, C-, or K-weighting filter.
36. The loudness signal processor according to claim 33 , wherein said pre-processor comprises channel weighting, preferably by being arranged to combine a multi-channel input audio signal into a single-channel preprocessed signal.
37. The loudness signal processor according to claim 36 , wherein said channel weighting comprises a per-sample power domain sum of channels of said input audio signal.
38. The loudness signal processor according to claim 36 , wherein said channel weighting comprises a gain matrix containing a gain coefficient for each individual channel of the input audio signal.
39. The loudness signal processor according to claim 1 , comprising a smoothing filter arranged to smooth said time-varying gain such that abrupt changes of said gain would not cause distortion of the output audio signal.
40. Loudness control method for controlling an integrated loudness level and a loudness range of an audio signal on the basis of a predetermined target loudness level and a predetermined target loudness range, the loudness control method comprising:
receiving said audio signal as an input audio signal;
determining a time-varying level of said input audio signal;
estimating a time-varying level distribution of said time-varying level;
determining a time-varying transfer function on the basis of said time-varying level distribution, and said predetermined target loudness level and said predetermined target loudness range;
calculating a time-varying gain on the basis of said time-varying level of the input audio signal and said time-varying transfer function; and
applying said time-varying gain to said input audio signal to establish an output audio signal,
wherein determining a time-varying transfer function on the basis of said time-varying level distribution comprises determining a loudness level estimated on the basis of said time-varying level distribution by performing an RMS calculation in which the levels below a threshold are excluded, as specified in the calculation of integrated, gated loudness level of ITU-R BS-1770-2 and EBU R-128.
41. The loudness control method according to claim 40 , wherein said estimating the time-varying level distribution is based on a part of said input audio signal by considering values of said time-varying level corresponding to different times with two or more different weights.
42. The loudness control method according to claim 40 , wherein said estimating the time-varying level distribution is performed by means of a finite-length sliding window.
43. The loudness control method according to claim 41 , wherein said estimating the time-varying level distribution is performed by means of a set of IIR filters.