Audio normalization for digital video broadcasts
View Patent ↗Systems and methods of this disclosure can operate to provide audio level normalization in multimedia streams. In some implementations, audio level normalization can include partially decoding audio frames included in a multimedia stream, calculating an average energy level of the partially decoded frames, comparing the average energy level of the partially decoded frames to a nominal energy level, modifying a dialnorm setting associated with the multimedia stream, and reencoding the partially decoded audio frames for inclusion as a modified audio stream corresponding to the video stream.
1. A computer-implemented method, comprising:
receiving a multimedia stream comprising an audio stream and a video stream;
separating the audio stream from the video stream of the multimedia stream;
partially decoding audio frames included in the audio stream of the multimedia stream, wherein partially decoding the audio frames comprises discontinuing the decoding after obtaining frequency band coefficients associated with the audio frames;
calculating an average energy level of the partially decoded frames across a sampling period;
comparing the average energy level of the partially decoded frames to a nominal energy level;
modifying a dialnorm setting associated with the audio stream based upon the comparison; and
reencoding the partially decoded audio frames for inclusion as a modified audio stream corresponding to the video stream.
2. The computer-implemented method of claim 1 , wherein pulse code modulation values are not obtained prior to calculating an average energy level of the partially decoded frames.
3. The computer-implemented method of claim 1 , wherein obtaining pulse code modulation values comprises a full decoding of the audio frames.
4. The computer-implemented method of claim 1 , wherein partial decoding the audio frames is operable to conserve substantial processing overhead associated with decoding the audio frames.
5. The computer-implemented method of claim 1 , wherein partially decoding the audio frames comprises partially decoding of an AC-3 frame structure to yield frequency band coefficients without pulse code modulation values.
6. The computer-implemented method of claim 5 , wherein calculating an average energy level comprises calculating a loudness value based on the frequency band coefficients.
7. The computer-implemented method of claim 1 , wherein the computer-implemented method is operable to normalize segments of the audio stream to conform to a constant overall television station audio level.
8. The computer-implemented method of claim 7 , wherein audio segments associated with an advertisement are normalized to audio segments associated with a main program segment.
9. The computer-implemented method of claim 1 , wherein the computer-implemented method is operable to normalize segments of audio streams from different channels in a channel lineup to conform to a normalized audio level.
10. A system, comprising:
a transport stream conversion module operable to receive audio packets and produce an AC-3 elementary stream associated with the audio packets;
an AC-3 dialnorm calculation module operable to analyze the AC-3 elementary stream, partially decode the AC-3 elementary stream to derive frequency coefficients associated with the AC-3 elementary stream, and to calculate a loudness value associated with the AC-3 elementary stream based on the frequency coefficients, wherein pulse code modulation values are not obtained prior to calculating a loudness level of the partially decoded frames;
an audio level adjustment module operable to receive the calculated loudness factor and compare the calculated loudness factor to an encoded dialnorm level, to determine whether the calculated loudness value is diverging from the encoded dialnorm level, and to adjust the encoded dialnorm level if the calculated loudness value diverges from the encoded dialnorm level by more than a predetermined amount.
11. The system of claim 10 , wherein the partially decode of AC-3 elementary stream comprises discontinuing the decoding after obtaining frequency band coefficients associated with the audio frames.
12. The system of claim 10 , wherein partially decoding the AC-3 elementary stream to derive frequency coefficients associated with the AC-3 elementary stream is operable to conserve substantial processing overhead associated with fully decoding the AC-3 elementary stream.
13. The system of claim 10 , wherein the system comprises a program audio chain associated with a program audio stream and an ad audio chain associated with an advertisement audio stream, and wherein the audio level adjustment module is further operable to splice advertisements into the program audio stream to maintain a relatively constant loudness value.
14. A computer-implemented method, comprising:
receiving a multimedia stream comprising an audio stream and a video stream;
separating the audio stream from the video stream of the multimedia stream;
partially decoding audio frames included in the audio stream of the multimedia stream, wherein pulse code modulation values are not obtained as a result of the partial decoding of the audio frames;
calculating an average energy level of the partially decoded frames across a sampling period;
comparing the average energy level of the partially decoded frames to a nominal energy level;
modifying a dialnorm setting associated with the audio stream based upon the comparison; and
reencoding the partially decoded audio frames for inclusion as a modified audio stream corresponding to the video stream.
15. The computer-implemented method of claim 14 , wherein obtaining pulse code modulation values comprises a full decoding of the audio frames.
16. The computer-implemented method of claim 14 , wherein partial decoding the audio frames is operable to conserve substantial processing overhead associated with decoding the audio frames.
17. The computer-implemented method of claim 14 , wherein partially decoding the audio frames comprises partially decoding of an AC-3 frame structure to yield frequency band coefficients without pulse code modulation values.
18. The computer-implemented method of claim 17 , wherein calculating an average energy level comprises calculating a loudness value based on the frequency band coefficients.
19. The computer-implemented method of claim 14 , wherein the computer-implemented method is operable to normalize segments of the audio stream to conform to a constant overall television station audio level.
20. The computer-implemented method of claim 19 , wherein audio segments associated with an advertisement are normalized to audio segments associated with a main program segment.
21. The computer-implemented method of claim 14 , wherein the computer-implemented method is operable to normalize segments of audio streams from different channels in a channel lineup to conform to a normalized audio level.