Method and system for audio compression for contextual volume control
A method that includes receiving audio content; receiving a volume level at which the audio content is to be played back through a speaker of the headset that is worn by a user; determining a noise level of environmental noise at the headset; determining a noise exposure level by the user based on the volume level, a content level of the audio content, and the noise level; determining whether a content level of the audio content is less than a threshold that is based on the noise exposure level; and responsive to determining that the content level is less than the threshold, adjusting the content level by applying a gain to the audio content such that the gain-adjusted content level of the audio content is at least equal to the noise exposure level.
1 . A method comprising:
receiving audio content;
receiving a volume level at which the audio content is to be played back through a speaker of a headset that is worn by a user;
determining a sensitivity of the headset;
determining a sound output level of the headset based on the sensitivity of the headset and a content level of the audio content;
determining a noise level of environmental noise at the headset;
producing an adjusted noise level based on the sound output level and the noise level;
determining a noise exposure level by the user based on the volume level and the adjusted noise level;
determining whether the content level of the audio content is less than a threshold dependent on the noise exposure level; and
responsive to determining that the content level is less than the threshold, adjusting the content level by applying a gain to the audio content such that a gain-adjusted content level of the audio content is at least equal to the noise exposure level.
2 . The method of claim 1 ,
wherein producing the adjusted noise level comprises subtracting the sound output level from the noise level.
3 . The method of claim 2 , wherein determining the noise exposure level further comprises subtracting the volume level from the adjusted noise level.
4 . The method of claim 1 , wherein the content level is a first content level of a first portion of the audio content, wherein the method further comprises:
receiving a second portion, subsequent to the first portion, of the audio content;
determining that a second content level of the second portion is greater than the noise exposure level; and
ceasing to apply the gain.
5 . The method of claim 1 , wherein adjusting the content level comprises applying upwards compression to the audio content.
6 . The method of claim 5 further comprising determining the threshold as a compression threshold of the upwards compression based on the noise exposure level, wherein the upwards compression is applied responsive to the content level being less than the compression threshold.
7 . The method of claim 6 further comprising determining a minimum signal-to-noise ratio (SNR) of the content level with respect to the noise exposure level, wherein the compression threshold is a combination of the noise exposure level and the minimum SNR.
8 . An electronic device, comprising:
at least one processor; and
memory having instructions stored therein which when executed by the at least one processor causes the electronic device to:
determine a volume level at which audio content is to be played back through a speaker of a headset that is to be worn by a user;
determine a sound output level of the headset based on a sensitivity of the headset and a content level of the audio content;
determine a noise level of environmental noise at the headset;
produce an adjusted noise level based on the sound output level and the noise level;
determine a noise exposure level by the user while wearing the headset based on the volume level and the adjusted noise level;
determine whether a content level of the audio content is less than a threshold that is based on the noise exposure level; and
responsive to the content level being less than the noise exposure level, apply audio compression upon the audio content such that the content level is at least equal to the noise exposure level.
9 . The electronic device of claim 8 , wherein the instructions to determine the noise level comprises instructions to receive, over a wireless connection, the noise level from the headset.
10 . The electronic device of claim 8 , wherein the memory has further instructions to receive, over a wireless connection, a volume setting of the headset, wherein the volume level is determined according to the volume setting, wherein the volume setting is produced by the headset based on a user-adjustment of a volume control of the headset or a volume machine learning model that outputs the volume setting responsive to the noise level as input.
11 . The electronic device of claim 8 , wherein the instructions to apply the audio compression comprises instructions to produce a gain-adjusted audio signal by applying one or more gains to an audio signal of the audio content, wherein the memory has further instructions to transmit, over a wireless connection, the gain-adjusted audio signal to the headset to drive the speaker of the headset.
12 . The electronic device of claim 8 ,
wherein the instructions to produce the adjusted noise level comprises instructions to subtract the sound output level from the noise level,
wherein the instructions to determine the noise exposure level comprises instructions to subtract the volume level from the adjusted noise level.
13 . The electronic device of claim 8 comprises a portable electronic device that is in wireless communication with the headset.
14 . The electronic device of claim 13 , wherein the portable electronic device comprises a smartphone, a laptop computer, or a smartwatch.
15 . A non-transitory machine-readable medium having instructions which when executed by at least one processor of an output device, causes the output device to:
receive audio content;
receive a volume level at which the audio content is to be played back through a speaker of the output device;
determine a sensitivity of the output device;
determine a sound output level of the output device based on the sensitivity and a content level of the audio content;
determine a noise level of environmental noise;
produce an adjusted noise level based on the sound output level and the noise level;
determine a noise exposure level by a user of the output device based on at least one of the volume level and the adjusted noise level;
determine whether the content level of the audio content is less than a threshold based on the noise exposure level; and
responsive to a determination that the content level is less than the threshold, apply audio compression upon the audio content such that the content level of the audio content is at least equal to the noise exposure level.
16 . The non-transitory machine-readable medium of claim 15 ,
wherein the instructions to produce the adjusted noise level comprises instructions to subtract the sound output level from the noise level.
17 . The non-transitory machine-readable medium of claim 16 , wherein the instructions to determine the noise exposure level comprises instructions to subtract the volume level from the adjusted noise level.
18 . The non-transitory machine-readable medium of claim 15 comprises further instructions to determine the threshold as a compression threshold based on the noise exposure level, wherein the audio compression is applied responsive to the content level being less than the compression threshold.
19 . The non-transitory machine-readable medium of claim 18 comprises further instructions to determine a minimum signal-to-noise ratio (SNR) of the content level, wherein the compression threshold is a combination of the noise exposure level and the minimum SNR.
20 . The non-transitory machine-readable medium of claim 15 , wherein the instructions to receive the noise level comprises instructions to:
receive a plurality of noise levels, each a loudness, k-weighted, full-scale (LKFS) value, wherein the non-transitory machine-readable medium; and
apply a smoothing filter to the plurality of noise levels to filter out one or more outlier LKFS values.