System and method for leveling loudness variation in an audio signal

Patent metadata

Field	Value
Publication number	US-9960742-B2
Application number	US-201715718913-A
Country	US
Kind code	B2
Filing date	Sep 28, 2017
Priority date	Apr 12, 2012
Publication date	May 1, 2018
Grant date	May 1, 2018

Abstract

Systems and methods for leveling loudness variation in an audio signal are described. Embodiments use both a perceptual leveling algorithm and a standards-based loudness measure together to minimize audio process artifacts and ensure that the measured loudness of the processed audio is close to a required measure, according to a particular standard measurement of loudness. These systems and methods can be used either offline or in real-time.

First claim

What is claimed is:

1. A system for leveling loudness variation in an audio signal, the system comprising:
a first long term loudness leveling processor configured to measure a first actual loudness of the audio signal using a first long term integration time;
a first loudness to gain processor configured to calculate a first gain value using a difference between a desired loudness of the audio signal and the first actual loudness of the audio signal;
a first audio modification processor configured to modify the audio signal using the first gain value to generate a modified audio signal;
a short term loudness leveling processor configured to receive the modified audio signal and the desired loudness, and configured to adjust a short term loudness of the modified audio signal using a short term integration time, to yield an adjusted audio signal, wherein adjusting the short term loudness comprises using spectral analysis and models of human hearing to adjust the modified audio signal;
a second long term loudness leveling processor configured to measure a second actual loudness of the adjusted audio signal using a second long term integration time, wherein the second long term integration time is longer than the short term integration time;
a second loudness to gain processor configured to calculate a second gain value using a difference between the desired loudness and the actual loudness of the adjusted audio signal; and
a second audio modification processor configured to modify the adjusted audio signal using the second gain value.

2. The system of claim 1 , wherein the short term integration time is shorter than the first long term integration time.

3. The system of claim 1 , wherein the first long term loudness leveling processor operates at a first rate, and wherein the short term loudness leveling processor operates at a second rate.

4. The system of claim 3 , wherein the first rate is faster than the second rate.

5. The system of claim 3 , wherein the first rate is slower than the second rate.

6. The system of claim 3 , wherein the first rate and the second rate are a same rate.

7. The system of claim 1 , wherein the first long term loudness leveling processor and the second long term loudness leveling processor use a same loudness measurement method.

8. The system of claim 1 , further comprising:
a memory, wherein the second long term loudness leveling processor is configured to measure the actual loudness of the adjusted audio signal over the entire adjusted audio signal.

9. The system of claim 1 , wherein the first loudness to gain processor uses rate limiting when calculating the first gain value.

10. A method of leveling loudness variation in an audio signal, the method comprising:
receiving the audio signal and a desired loudness of the audio signal;
measuring a first actual loudness of the audio signal using a first long term integration time;
calculating a first gain value using a difference between a desired loudness of the audio signal and the first actual loudness of the audio signal;
modifying the audio signal using the first gain value to generate a modified audio signal;
adjusting a short term loudness of the modified audio signal using a short term integration time, to yield an adjusted audio signal, wherein adjusting the short term loudness comprises using spectral analysis and models of human hearing to adjust the modified audio signal;
measuring a second actual loudness of the adjusted audio signal using a second long term integration time, wherein the second long term integration time is longer than the short term integration time;
calculating a second gain value using a difference between the desired loudness and the actual loudness of the adjusted audio signal; and
modifying the adjusted audio signal using the second gain value.

11. The method of claim 10 , wherein the short term integration time is shorter than the first long term integration time.

12. The method of claim 10 , wherein measuring the first actual loudness occurs at a first rate, and wherein adjusting the short term loudness occurs at a second rate.

13. The method of claim 12 , wherein the first rate is faster than the second rate.

14. The method of claim 12 , wherein the first rate is slower than the second rate.

15. The method of claim 12 , wherein the first rate and the second rate are a same rate.

16. The method of claim 10 , wherein measuring the first actual loudness and measuring the second actual loudness use a same loudness measurement method.

17. The method of claim 10 , wherein measuring the second actual loudness occurs over the entire adjusted audio signal.

18. The method of claim 10 , wherein calculating the first gain value includes using rate limiting when calculating the first gain value.

19. A non-transitory computer readable medium having computer executable instructions embedded thereon for performing, when executed by a processor, a method of leveling loudness variation in an audio signal, the method comprising:
receiving the audio signal and a desired loudness of the audio signal;
measuring a first actual loudness of the audio signal using a first long term integration time;
calculating a first gain value using a difference between a desired loudness of the audio signal and the first actual loudness of the audio signal;
modifying the audio signal using the first gain value to generate a modified audio signal;
adjusting a short term loudness of the modified audio signal using a short term integration time, to yield an adjusted audio signal, wherein adjusting the short term loudness comprises using spectral analysis and models of human hearing to adjust the modified audio signal;
measuring a second actual loudness of the adjusted audio signal using a second long term integration time, wherein the second long term integration time is longer than the short term integration time;
calculating a second gain value using a difference between the desired loudness and the actual loudness of the adjusted audio signal; and
modifying the adjusted audio signal using the second gain value.

20. The non-transitory computer readable medium of claim 19 , wherein the short term integration time is shorter than the first long term integration time.

21. The non-transitory computer readable medium of claim 19 , wherein measuring the first actual loudness occurs at a first rate, and wherein adjusting the short term loudness occurs at a second rate.

22. The non-transitory computer readable medium of claim 21 , wherein the first rate is faster than the second rate.

23. The non-transitory computer readable medium of claim 21 , wherein the first rate is slower than the second rate.

24. The non-transitory computer readable medium of claim 21 , wherein the first rate and the second rate are a same rate.

25. The non-transitory computer readable medium of claim 19 , wherein measuring the first actual loudness and measuring the second actual loudness use a same loudness measurement method.

26. The non-transitory computer readable medium of claim 19 , wherein measuring the second actual loudness occurs over the entire adjusted audio signal.

27. The non-transitory computer readable medium of claim 19 , wherein calculating the first gain value includes using rate limiting when calculating the first gain value.

Assignees

• Dolby Laboratories Licensing Corp

Inventors

• Reilly Andrew P
• Smithers Michael
• Riedmiller Jeffrey

Classifications

• H03G7/007

  of digital or coded signals · CPC title

• H03G7/002 Primary

  in untuned or low-frequency amplifiers, e.g. audio amplifiers (H03G7/007, H03G7/001, H03G7/008, H03G7/02, H03G7/06 take precedence) · CPC title

• H03G3/20 Primary

  Automatic control ({H03G3/005 takes precedence;} combined with volume compression or expansion H03G7/00) · CPC title