Granted Patent B2

US 9,509,267 · App. 14/356,678 · Granted Nov 29, 2016

Abstract

Claims (54)

1. A method for loudness leveling of a queue of audio tracks in an audio player comprising:

determining a first loudness estimate for at least one frame of an audio track in the queue, wherein at least two frequency band energy values from the at least one frame of the audio track are generated and an estimated loudness value based on the at least two frequency band energy values for the at least one frame of the audio track is determined;

determining a second loudness estimate for at least one frame of a second audio track in the queue;

comparing the second loudness estimate against the first loudness estimate;

selecting a loudness estimate based on one of the first or second loudness estimate dependent on the comparison;

determining a level value for the second audio track based on the selected loudness estimate; and

adjusting the loudness of the second audio track based on the level value so as to reduce a disparity in loudness level across the first and second audio tracks in the queue.

2. The method as claimed in claim 1 , further comprising determining a desired level, and wherein adjusting the second audio track is further based on the desired level.

3. The method as claimed in claim 2 , wherein determining a desired level comprises at least one of:

receiving a desired level from a user interface input;

determining a desired level from a selected audio signal; and

receiving a desired level from a memory.

4. The method as claimed in claim 1 , further comprising comparing the first frame loudness estimate against at least one historical loudness estimate.

5. The method as claimed in claim 4 , further comprising generating an estimate of a distribution model dependent on the at least one historical loudness estimate; and

determining whether the first frame loudness estimate fits within the estimate of the distribution model.

6. The method as claimed in claim 5 , further comprising selecting the first frame loudness estimate when the first frame loudness estimate fits within the estimate of the distribution model; and selecting the at least one historical frame loudness estimate when the first frame loudness estimate fails to fit the estimate of the distribution model.

7. The method as claimed in claim 1 , wherein determining the first loudness estimate for the at least one frame of the audio track further comprises:

applying a regression coefficient to each of the at least two frequency band energy values; and

combining the regression coefficient modified at least two frequency band audio signals to generate a first frame loudness estimate.

8. An apparatus comprising at least one processor and at least one memory storing computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least:

determine a first loudness estimate for at least one frame of an audio track in a queue, wherein at least two frequency band energy values from the at least one frame of the audio track are generated and an estimated loudness value based on the at least two frequency band energy values for the at least one frame of the audio track is determined;

determine a second loudness estimate for at least one frame of a second audio track in the queue;

compare the second loudness estimate against the first loudness estimate;

select a loudness estimate based on one of the first or second loudness estimate dependent on the comparison;

determine a level value for the second audio track based on the selected loudness estimate; and

adjust the loudness of the second audio track based on the level value so as to reduce a disparity in loudness level across the first and second audio tracks in the queue.

9. The apparatus as claimed in claim 8 , further caused to determine a desired level, and wherein adjusting the second audio track is further based on the desired level.

10. The apparatus as claimed in claim 9 , wherein determining a desired level causes the apparatus to perform at least one of:

receive a desired level from a user interface input;

determine a desired level from a selected audio signal; and

receive a desired level from a memory.

11. The apparatus as claimed in claim 8 , wherein the first frame loudness estimate is compared against at least one historical loudness estimate.

12. The apparatus as claimed in claim 11 , wherein the apparatus further estimates a distribution model dependent on the at least one historical loudness estimate and determines whether the first frame loudness estimate fits within the estimate of the distribution model.

13. The apparatus as claimed in claim 12 , wherein the apparatus further selects the first frame loudness estimate when the first frame loudness estimate fits within the estimate of the distribution model.

14. The apparatus as claimed in claim 12 , wherein the apparatus further selects the at least one historical frame loudness estimate when the first frame loudness estimate fails to fit the estimate of the distribution model.

15. The apparatus as claimed in claim 8 , wherein determining the first loudness estimate for the at least one frame of the audio track further causes the apparatus to:

apply a regression coefficient to each of the at least two frequency band energy values; and

combine the regression coefficient modified at least two frequency band audio signals to generate a first frame loudness estimate.

16. A computer program product comprising at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions comprising program code instructions configured to:

determine a first loudness estimate for at least one frame of an audio track in a queue, wherein at least two frequency band energy values from the at least one frame of the audio track are generated and an estimated loudness value based on the at least two frequency band energy values for at least one frame of the audio track is determined;

determine a second loudness estimate for at least one frame of a second audio track in the queue;

compare the second loudness estimate against the first loudness estimate;

select a loudness estimate based on one of the first or second loudness estimate dependent on the comparison;

determine a level value for the second audio track based on the selected loudness estimate; and

adjust the loudness of the second audio track based on the level value so as to reduce a disparity in loudness level across the first and second audio tracks in the queue.

17. A computer program product according to claim 16 further comprising program code instructions configured to determine a desired level, and wherein adjusting the second audio track is further based on the desired level.

18. A computer program product according to claim 17 wherein the program code instructions configured to determine the desired level comprise program code instructions configured to:

receive a desired level from a user interface input;

determine a desired level from a selected audio signal; and

receive a desired level from a memory.

19. A computer program product according to claim 16 , further comprising program code instructions configured to estimate a distribution model dependent on at least one historical loudness estimate and determine whether the first frame loudness estimate fits within the estimate of the distribution model.

20. A computer program product according to claim 16 , wherein the program code instructions configured to determine the first loudness estimate for the at least one frame of the audio track comprise program code instructions configured to:

apply a regression coefficient to each of the at least two frequency band energy values; and

combine the regression coefficient modified at least two frequency band audio signals to generate a first frame loudness estimate.

Assignments (12)