Monitoring loudness level during media replacement event using shorter time constant
In one aspect, an example method includes (i) determining, by a playback device, a first loudness level of a first portion of first media content from a first source while the playback device presents the first media content, with the first portion having a first length; (ii) switching, by the playback device, from presenting the first media content from the first source to presenting second media content from a second source; (iii) based on the switching, determining, by the playback device, second loudness levels of second portions of the first media content while the playback device presents the second media content, with the second portions having a second length that is shorter than the first length; and (iv) while the playback device presents the second media content, adjusting, by the playback device, a volume of the playback device based on one or more of the second loudness levels.
1. A method comprising:
switching, by a playback device, from presenting first media content from a first source to presenting second media content from a second source;
while the playback device presents the second media content from the second source, receiving, by the playback device, an audio signal corresponding to the first media content;
determining, by the playback device, loudness levels of portions of the first media content using the audio signal corresponding to the first media content while the playback device presents the second media content; and
while the playback device presents the second media content, adjusting, by the playback device, a volume of the playback device based on one or more of the loudness levels.
2. The method of claim 1 , wherein:
the loudness levels comprise a loudness level of a given portion and a loudness level of a subsequent portion,
the method further comprises determining that the loudness level of the given portion differs from the loudness level of the subsequent portion by more than a threshold amount, and
adjusting the volume of the playback device comprises adjusting the volume based on a difference between the loudness level of the given portion and the loudness level of the subsequent portion.
3. The method of claim 2 , further comprising determining a target volume based on the difference,
wherein adjusting the volume comprises adjusting the volume toward the target volume.
4. The method of claim 1 , wherein adjusting the volume based on one or more of the loudness levels comprises:
providing the loudness levels as input to a neural network, wherein the neural network is trained using audio samples that include volume adjustments;
detecting a volume adjustment to the first media content based on an output of the neural network; and
adjusting the volume based on the volume adjustment.
5. The method of claim 4 , wherein the output of the neural network comprises data classifying the loudness levels as being characteristic of a volume adjustment rather than a fluctuation that is inherent to the first media content.
6. The method of claim 4 , further comprising identifying a trend in the loudness levels,
wherein the adjusting the volume is further based on the identified trend.
7. The method of claim 1 , wherein adjusting the volume comprises adjusting the volume using a smoothing filter.
8. The method of claim 1 , wherein:
the playback device comprises a television, and
presenting the second media content comprises presenting a replacement advertisement using a media player.
9. The method of claim 1 , wherein:
the second media content is stored in a local cache of the playback device, and
the method further comprises retrieving the second media content from the local cache prior to presenting the second media content.
10. A non-transitory computer-readable medium having stored thereon program instructions that upon execution by a processor, cause performance of a set of acts comprising:
switching from presenting first media content from a first source using a playback device to presenting second media content from a second source using the playback device;
while the playback device presents the second media content from the second source, receiving an audio signal corresponding to the first media content;
determining loudness levels of second portions of the first media content using the audio signal corresponding to the first media content while the playback device presents the second media content; and
while the playback device presents the second media content, adjusting a volume of the playback device based on one or more of the loudness levels.
11. The non-transitory computer-readable medium of claim 10 , wherein:
the loudness levels comprise a loudness level of a given portion and a loudness level of a subsequent portion,
the set of acts further comprises determining that the loudness level of the given portion differs from the loudness level of the subsequent portion by more than a threshold amount, and
adjusting the volume of the playback device comprises adjusting the volume based on a difference between the loudness level of the given portion and the loudness level of the subsequent portion.
12. The non-transitory computer-readable medium of claim 11 , wherein:
the set of acts further comprises determining a target volume based on the difference, and
adjusting the volume comprises adjusting the volume toward the target volume.
13. The non-transitory computer-readable medium of claim 10 , wherein adjusting the volume based on one or more of the loudness levels comprises:
providing the loudness levels as input to a neural network, wherein the neural network is trained using audio samples that include volume adjustments;
detecting a volume adjustment to the first media content based on an output of the neural network; and
adjusting the volume based on the volume adjustment.
14. The non-transitory computer-readable medium of claim 13 , wherein the output of the neural network comprises data classifying the loudness levels as being characteristic of a volume adjustment rather than a fluctuation that is inherent to the first media content.
15. The non-transitory computer-readable medium of claim 10 , wherein:
the second media content is stored in a local cache of the playback device, and
the set of acts further comprises retrieving the second media content from the local cache prior to presenting the second media content.
16. A computing system configured for performing a set of acts comprising:
switching from presenting first media content from a first source using a playback device to presenting second media content from a second source using the playback device;
while the playback device presents the second media content from the second source, receiving an audio signal corresponding to the first media content;
determining loudness levels of second portions of the first media content using the audio signal corresponding to the first media content while the playback device presents the second media content; and
while the playback device presents the second media content, adjusting a volume of the playback device based on one or more of the loudness levels.
17. The computing system of claim 16 , wherein:
the loudness levels comprise a loudness level of a given portion and a loudness level of a subsequent portion,
the set of acts further comprises determining that the loudness level of the given portion differs from the loudness level of the subsequent portion by more than a threshold amount, and
adjusting the volume of the playback device comprises adjusting the volume based on a difference between the loudness level of the given portion and the loudness level of the subsequent portion.
18. The computing system of claim 17 , wherein:
the set of acts further comprises determining a target volume based on the difference, and
adjusting the volume comprises adjusting the volume toward the target volume.
19. The computing system of claim 16 , wherein adjusting the volume based on one or more of the loudness levels comprises:
providing the loudness levels as input to a neural network, wherein the neural network is trained using audio samples that include volume adjustments;
detecting a volume adjustment to the first media content based on an output of the neural network; and
adjusting the volume based on the volume adjustment.
20. The computing system of claim 16 , wherein:
the second media content is stored in a local cache of the playback device, and
the set of acts further comprises retrieving the second media content from the local cache prior to presenting the second media content.