Estimation of true audience size for digital content

Abstract

A content server system provides a client device with content, such as an audio stream. Using various techniques, an estimate is made of the actual size of an audience associated with the provided content, rather than assuming that the audience is limited to a single user of the client device. The estimate may be made by the content server system, the client device, or the content server system and the client device collectively. Using the estimate of actual size of the audience, the content server system can take actions appropriate for the audience, such as providing advertisements appropriate for the size of the audience and for the collective characteristics of the audience members. The estimate of the actual audience size additionally allows the content server system to be compensated more precisely for any advertisements provided to that audience.

What is claimed is: 
     
         1 . A computer-implemented method, comprising:
 identifying a streaming client device receiving streamed audio;   determining a number of inferred listeners that are listening to the streamed audio when output by the streaming client device, the inferred listeners being in addition to a user using of the streaming client device;   including an advertisement in the streamed audio provided to the streaming client device; and   logging an impression count responsive to the inclusion of the advertisement, the impression count reflecting the determined number of inferred listeners.   
     
     
         2 . The computer-implemented method of  claim 1 , wherein determining the number of inferred listeners comprises:
 receiving first location-related data associated with the streaming client device;   receiving additional location-related data associated with additional client devices other than the streaming client device;   determining, based on the first location-related data and the additional location-related data, that a user of the streaming client device and users of the additional client devices can all hear audio output of the streaming client device; and   responsive to determining that the user of the streaming client device and users of the additional client devices can all hear audio output of the streaming client device, determining the number of inferred listeners based on a number of the additional client devices.   
     
     
         3 . The computer-implemented method of  claim 1 , wherein determining the number of inferred listeners comprises:
 receiving first data derived from sensors of the streaming client device;   receiving additional data derived from sensors of additional client devices other than the streaming client devices;   determining, based on the first data of the streaming client device and of the additional client devices, that a user of the streaming client device and users of the additional client devices are behaving in a similar manner; and   responsive to determining that the user of the streaming client device and the users of the additional client devices are behaving in a similar manner, determining the number of inferred listeners based on a number of the additional client devices.   
     
     
         4 . The computer-implemented method of  claim 1 , wherein determining the number of inferred listeners comprises:
 providing a stimulant message to the streaming client device, the stimulant message prompting users to react in a specified manner;   receiving, from the streaming client device, reaction data measuring user reaction to the stimulant message;   estimating a number of distinct people represented by the reaction data; and   responsive to the estimation, determining the number of inferred listeners based on the estimated number of distinct people.   
     
     
         5 . The computer-implemented method of  claim 1 , wherein determining the number of inferred listeners comprises identifying a number of distinct human voices in audio input data of the streaming client device. 
     
     
         6 . The computer-implemented method of  claim 1 , further comprising:
 selecting the advertisement that is included in the streamed audio based at least in part on the inferred listeners.   
     
     
         7 . The computer-implemented method of  claim 6 , further comprising:
 determining that there is a plurality of inferred users;   identifying a plurality of languages spoken among the inferred users and the user using the streaming client device;   identifying a primary language of a country in which the streaming client device is located; and   responsive to identifying the plurality of languages, selecting, as the advertisement that is included in the streamed audio, an advertisement in the primary language.   
     
     
         8 . The computer-implemented method of  claim 1 , wherein determining the number of inferred listeners comprises:
 determining whether audio output of the streaming client device is being provided via headphones; and   responsive to determining that the audio output of the streaming client device is being provided via headphones, determining that there are no inferred listeners.   
     
     
         9 . A computer-implemented method performed by a client device, comprising:
 receiving streamed audio from a content server system;   outputting the streamed audio;   determining a number of inferred listeners that are listening to the streamed audio output by the client device, the inferred listeners being in addition to a user using the client device; and   sending, to the content server system, an audience count that is based on the inferred number of listeners.   
     
     
         10 . The computer-implemented method of  claim 9 , comprising:
 receiving an advertisement from the content server system that is selected based at least in part on the inferred listeners.   
     
     
         11 . The computer-implemented method of  claim 9 , comprising:
 receiving location-related data associated with additional client devices;   determining, based on location-related data of the client device, and on the location-related data associated with the additional client devices, that a user of the client device and users of the additional client devices can all hear audio output of the client device; and   responsive to determining that the user of the client device and users of the additional client devices can all hear audio output of the client device, notifying the content server system of a presence of inferred listeners.   
     
     
         12 . The computer-implemented method of  claim 9 , comprising:
 receiving a request from a user of the client device to share the streamed audio with nearby users; and   responsive to receiving a notification that a nearby user has joined the audio stream that was shared, notifying the content server system of a presence of an additional inferred listener.   
     
     
         13 . The computer-implemented method of  claim 9 , comprising:
 receiving, from additional client devices via short-range wireless communications, data derived from sensors of the additional client devices;   determining, based on data derived from sensors of the client device and on the data derived from sensors of the additional client devices, that a user of the client device and users of the additional client devices are behaving in a similar manner; and   responsive to the determination, notifying the content server system of a presence of inferred listeners.   
     
     
         14 . The computer-implemented method of  claim 9 , comprising:
 receiving, within the streamed audio, a stimulant message prompting users to react in a specified manner;   obtaining reaction data by measuring user reaction to the stimulant message on the client device;   estimating a number of distinct people represented by the reaction data; and   responsive to the number being greater than 1, notifying the content server system of a presence of inferred listeners.   
     
     
         15 . A computer-readable storage medium comprising computer program instructions executable by a processor, the instructions comprising:
 instructions for identifying a streaming client device receiving streamed audio;   instructions for determining a number of inferred listeners that are listening to the streamed audio when output by the streaming client device, the inferred listeners being in addition to a user using of the streaming client device;   instructions for including an advertisement in the streamed audio provided to the streaming client device; and   instructions for logging an impression count responsive to the inclusion of the advertisement, the impression count reflecting the determined number of inferred listeners.   
     
     
         16 . The computer-readable storage medium of  claim 15 , wherein determining the number of inferred listeners comprises:
 receiving first location-related data associated with the streaming client device;   receiving additional location-related data associated with additional client devices other than the streaming client device;   determining, based on the first location-related data and the additional location-related data, that a user of the streaming client device and users of the additional client devices can all hear audio output of the streaming client device; and   responsive to determining that the user of the streaming client device and users of the additional client devices can all hear audio output of the streaming client device, determining the number of inferred listeners based on a number of the additional client devices.   
     
     
         17 . The computer-readable storage medium of  claim 15 , wherein determining the number of inferred listeners comprises:
 receiving first data derived from sensors of the streaming client device;   receiving additional data derived from sensors of additional client devices other than the streaming client devices;   determining, based on the first data of the streaming client device and of the additional client devices, that a user of the streaming client device and users of the additional client devices are behaving in a similar manner; and   responsive to determining that the user of the streaming client device and the users of the additional client devices are behaving in a similar manner, determining the number of inferred listeners based on a number of the additional client devices.   
     
     
         18 . The computer-readable storage medium of  claim 15 , wherein determining the number of inferred listeners comprises:
 providing a stimulant message to the streaming client device, the stimulant message prompting users to react in a specified manner;   receiving, from the streaming client device, reaction data measuring user reaction to the stimulant message;   estimating a number of distinct people represented by the reaction data; and   responsive to the estimation, determining the number of inferred listeners based on the estimated number of distinct people.   
     
     
         19 . The computer-readable storage medium of  claim 15 , further comprising:
 selecting the advertisement that is included in the streamed audio based at least in part on the inferred listeners.   
     
     
         20 . The computer-readable storage medium of  claim 15 , wherein determining the number of inferred listeners comprises:
 determining whether audio output of the streaming client device is being provided via headphones; and   responsive to determining that the audio output of the streaming client device is being provided via headphones, determining that there are no inferred listeners.