HK1219011B - Storage medium, apparatus and method to distinguish local media from spillover media - Google Patents

Description

Method, device and storage medium for distinguishing local media from overflow media

Technical Field

The present disclosure relates generally to media monitoring and, more particularly, to methods and apparatus for detecting overflow in audience monitoring systems.

Background Art

Consuming media presentations typically includes listening to audio information and/or viewing visual information such as, for example, radio programs, music, television programs, movies, still images, etc. Media-centric companies such as, for example, advertising agencies, broadcast networks, etc. are often interested in the viewing and listening interests of their audiences in order to better market products and/or services.

Summary of the Invention

The present disclosure relates to a method for distinguishing local media from overflow media, the method comprising the following steps: sampling a first audio signal from a first source received by a first microphone, the sampling of the first audio signal generating a first audio sample; sampling a second audio signal from the first source received by a second microphone, wherein the second microphone is separated from the first microphone by a first distance, the sampling of the second audio signal generating a second audio sample; selecting a first offset value based on a first correlation value between the first audio sample and the second audio sample; determining a count of the first offset value; associating the first offset value with the cluster if the first offset value is within a first number of samples of second offset values in the cluster; calculating a weighted average of the first offset value and the second offset value; calculating an origin direction of the first source based on the weighted average; when the origin direction is within a threshold angle, detecting an audio code embedded in at least one of the first audio signal or the second audio signal, the audio code identifying the media associated with the first source; and ignoring the media associated with the first source when the origin direction exceeds the threshold angle.

BRIEF DESCRIPTION OF THE DRAWINGS

1 is a block diagram of an example environment in which an example audience measurement system constructed according to the teachings of the present disclosure may be operated to collect audience measurement data.

FIG. 2 is an enlarged illustration of a portion of the example measurement system of FIG. 1 .

FIG3 is a block diagram of an example implementation of the meter of FIG1.

FIG. 4 is an enlarged illustration of a portion of the example measurement system of FIG. 1 .

5 is a block diagram of an example implementation of the direction detector of FIG. 2 .

6 is a block diagram of an example implementation of the post-processor of FIG. 5 .

7 is an illustration of the example meter of FIG. 1 in operation.

8A and 8B are illustrations of example tables used in connection with the example measurement system of FIG. 1 .

9 and 10 are flow diagrams representative of example machine readable instructions that may be executed to implement the example meter of FIGs. 1-3.

11 is a flow diagram representative of example machine readable instructions that may be executed to implement the example direction detector of FIGS. 3 and 4 .

12 is a flow diagram representative of example machine readable instructions that may be executed to implement the example post-processor of FIGS. 5 and 6 .

FIG. 13 illustrates an example table used in connection with the example post-processor of FIG. 4 and FIG. 5 .

14 is a block diagram of an example processing system capable of executing the example machine readable instructions of FIGS. 9-11 and/or 12 to implement the example meter 112 of FIGS. 1-3 , the example direction detector of FIGS. 3 and 5 , and/or the example post-processor 520 of FIGS. 5 and 6 .

DETAILED DESCRIPTION

Techniques for measuring the exposure and/or number of audience members exposed to media include awarding media exposure credits for a media presentation when an audience member is exposed to the media presentation. The awarding of media exposure credits can be determined by the media exposed to the audience member and by detecting the identity and/or number of people in the audience. A meter placed near a media presentation device such as a television can also be used to monitor the media usage identification activities of audience members. The meter can be configured to monitor the media exposure (e.g., viewing activity and/or listening activity) of one or more audience members using one or more techniques. For example, one technique for monitoring media exposure includes detecting and/or collecting information (e.g., codes, signatures, etc.) from an audio signal emitted or presented by a media delivery device (e.g., a television, stereo system, speakers, computer, etc.).

When an audience member is exposed to media presented by a media presentation device, the meter can detect audio associated with the media and generate monitoring data therefrom. Generally speaking, the monitoring data can include any information representing (or associated with) and/or that can be used to identify a particular media presentation (e.g., content, advertisement, song, television program, movie, video game, etc.). For example, the monitoring data can include a signature collected or generated by the meter based on audio codes broadcast using (e.g., embedded in) the media.

Unfortunately, the typical home presents unique monitoring challenges for meters. For example, a typical home includes multiple media delivery devices, each of which is configured to deliver media to a specific viewing and/or listening area located within the home. A meter located in a viewing area (e.g., a room in the home) can be configured to detect any media being delivered to the detection area by a specific media presentation device, and to trust programming associated with the media that has been exposed to audience members. Thus, the meter operates under the premise that any media detected by the meter is associated with programming originating from a specific media presentation device in the monitored area. However, in some cases, the meter may detect media content emitted by one or more different media delivery devices that are not located in the viewing area, resulting in the detected programming being inappropriately trusted to the wrong device and/or audience member.

The ability of a meter to detect audio being delivered outside of the monitored area is an effect known as "spillover," as media being delivered outside of the monitored area is said to "spill over" into the viewing area occupied by the meter and the specific media presentation device that the meter is intended to measure. Spillover can, for example, cause the meter to inappropriately credit the media as having originated from the bedroom TV and being exposed to people in the bedroom, if a meter associated with and proximate to a television in a bedroom detects audio being delivered by a television in an adjacent living room.

The meter can also be used to determine whether the media presentation device to be monitored is turned on. This determination can be made by monitoring the sound level detected by the meter and determining that the media presentation device is turned on if the detected sound level is above a threshold (e.g., a threshold audio energy value in dB). However, if ambient room noise (e.g., from a person talking near the meter) or sounds from another media presentation device causes the meter to detect sounds above the threshold, the meter may erroneously determine that the media presentation device to be monitored is turned on, even if it is actually turned off.

Example methods, apparatuses, systems, and/or articles disclosed herein include a meter that receives audio from a media presentation device and monitors audio codes embedded in the audio received from the media presentation device via two spaced-apart microphones located on either side of the meter. In some examples disclosed herein, the meter determines the direction of the received audio by measuring the time difference between when the audio is received by each of the two microphones. In other examples disclosed herein, the meter determines whether the received audio originates from the media presentation device being monitored based on the determined direction of the sound.

FIG1 is a block diagram of an example environment in which an example audience measurement system constructed in accordance with the teachings of the present disclosure can operate to collect audience measurement data. For purposes of clarity and efficiency, the example system and corresponding methods, systems, devices, and/or articles of manufacture are described herein with respect to an example area/environment 102 for use. In the illustrated example of FIG1 , area 102 is a home. However, in other examples, area 102 can be a store (e.g., a retail store), a shopping mall, an amusement park, and/or other area. Example environment 102 includes rooms 104 and 106, a media device 108, one or more media device speakers 110, an overflow media device 122, one or more overflow media device speakers 124, and one or more audience members 118, 120, and 126. The illustrated example environment is monitored by rooms 112, 114, and 116. Data collected in the environment is delivered via a network 128 to a data collection facility 130 having a server 132 and a database 134.

Rooms 104 and 106 of the illustrated example are rooms located within home 102 (eg, bedrooms, kitchen, living room, etc.).

An example media device 108 delivers media (e.g., content and/or advertisements), and one or more speakers 110 emit audio signals that propagate throughout the room 104. Audience members 118, 120 in the room 104 are exposed to the media delivered by the media device 108. The media device 108 may be represented by a television, a radio, a computer, etc.

In the example illustrated in FIG1 , a meter 112 monitors the media delivered by a media device 108 by detecting an audio signal from a media device speaker 110. The example meter 112 detects one or more audio codes embedded in the detected audio signal. The audio code identifies a specific media presentation (e.g., a television program) being delivered by the media device 108. This identification can be by identifying the media presentation by name, or by identifying a station and/or signature and a timestamp that enables a lookup of the program in a table. The example meter 112 of the example illustrated is placed in a room 104 near the media device 108 and detects audio signals emitted from one or more speakers 110 via a first microphone 114 and a second microphone 116. In the example illustrated in FIG1 , the first microphone 114 and the second microphone 116 are separated by a certain distance (e.g., based on the size of the housing of the meter 112 (e.g., three inches)). Because the example first microphone 114 is spatially closer to the speaker 110 than to the example second microphone 116, the sound detected by the example meter 112 will arrive at the first microphone before arriving at the second microphone 116. The time difference between when the two microphones detect the audio can be used to determine the direction of origin (i.e., the direction from which the sound originates).

In the example illustrated in FIG1 , audio signals emitted by the media device speaker 110, the overflow media device speaker 124, and/or other sources propagate throughout the room 104 at the speed of sound. Generally speaking, the speed of sound depends on atmospheric conditions, including air temperature and humidity, and will be assumed herein to propagate at a speed of 13,041.6 inches per second (331.25 meters per second). However, one or more alternative environmental conditions may result in one or more alternative propagation speeds. If the example media device speaker 110 emits a sound at time zero (t ₀ ), the emitted sound will arrive at the first microphone 114 at a first time (t ₁ ) and at the second microphone 116 at a second time (t ₂ ). The example meter 112 may calculate the difference between the time the audio signal arrives at the first microphone 114 (t ₁ ) and the time the audio signal arrives at the second microphone 116 (t ₂ ). Additionally, because the propagation speed of sound is known and the distance between the first microphone 114 and the second microphone 116 is known, the angle between the example meter 112 and the source of the audio signal can be determined by the example meter 112 .

1 , overflow media device 122 is a media delivery device (e.g., a television, radio, etc.) located in room 106. Example overflow media device 122 delivers media audio via one or more overflow media device speakers 124 intended to deliver media to audience members 126 within room 106. However, audio emitted by example overflow media device speakers 124 may overflow into another room 104 and be detected by meter 112. This may cause meter 112 to erroneously record media delivered by overflow media device 122 as being delivered by media device 108. This may result in errors in reporting media presented by media devices 108, 122 in household 102.

The example data collection facility of the example illustrated in FIG1 receives data collected by the example meter 112 (e.g., detected audio codes, detected origination direction of received audio). In the example illustrated in FIG1 , the data collection facility 130 includes a server 132 and a database 134. The example database 134 stores the data received by the data collection facility 130 from the meter 112 and can be implemented using any suitable memory and/or data storage devices and/or technologies. The example server 132 analyzes the information stored in the database 134 to, for example, determine one or more media usage activities of the example household 102.

In the example illustrated in FIG1 , the meter 112 is capable of communicating with the data collection facility 130, and vice versa, via a network 128. The example network 128 of FIG1 enables selective establishment and/or disconnection of a connection between the example meter 112 and the example data collection facility 130. The example network 128 can be implemented using any type of public or private network, such as, for example, the Internet, a telephone network, a local area network (LAN), a cable network, and/or a wireless network. To enable communication via the example network 128, the example meter 112 and the example data collection facility 130 of the illustrated example of FIG1 include communication interfaces that enable connection to, for example, Ethernet, a digital subscriber line (DSL), a telephone line, a coaxial cable, and/or a wireless connection.

FIG2 illustrates an example implementation of detecting overflow in an environment having one or more specific locations of a media device speaker 110 and an overflow media device speaker 124. The first microphone 114 may receive an audio signal before or after the second microphone 116 receives the same audio signal. The time delay between when the audio is received by the two microphones 114, 116 depends on the distance between the two microphones 114, 116 and the angle at which the sound is received. Because the distance between the microphones 114, 116 is fixed in the example illustrated in FIG2 , the time delay between when the audio signal is received by the two microphones 114, 116 depends on the angle of the source of the audio signal.

In the example illustrated in FIG2 , the media device speaker 110 and the overflow media device speaker 124 are positioned at different angles relative to the meter 112. Sound from the example media device speaker 110 is received by the meter 112 from within angle (A) 202, indicating that the sound source is located within area 204. Additionally, sound from the example overflow media device speaker 124 is received by the example meter 112 from outside angle (A) 202, indicating that the sound source is located within area 206. By using the techniques described herein to detect the direction of origin of the sound, the meter 112 is able to accept audio codes received from within area 204 and ignore audio codes received from area 206. This enables the meter 112 to distinguish local media (e.g., audio emitted by the media device 108) from overflow media (e.g., audio emitted by the overflow media device 122), ignore audio from the overflow media device speaker 124, and only accept audio codes received from the media device speaker 110.

In the example illustrated in FIG2 , the angle of the received audio is determined by measuring the time difference between when the audio is received by first microphone 114 and second microphone 116. Because the distance from media device speaker 110 to first microphone 114 is the same as the distance from media device speaker 110 to second microphone 116, audio received from the example media device speaker 110 in FIG1 will be received by example first microphone 114 and example second microphone 116 at approximately the same time. However, the distance from first microphone 114 to point 208 in the example of FIG2 is the same as the distance from second microphone 116 to point 208. Therefore, first microphone 114 and second microphone 116 will simultaneously receive audio received by meter 112 from the sound source at point 208. Thus, when first microphone 114 and second microphone 116 receive audio simultaneously, meter 112 determines whether the audio is coming from the direction of media device speaker 110 in the example of FIG2 or from the direction of point 208. However, meter 112 cannot determine which of these two directions the audio is coming from. This is referred to as front-back ambiguity and is caused by the fact that only two microphones are used in the meter 112. As described in further detail below, example methods, systems, apparatuses, and/or articles of manufacture disclosed herein facilitate resolving the front-back ambiguity via predetermined, periodic, and/or non-periodic rotation of the first microphone 114 and the second microphone 116.

Figure 3 is a block diagram of an implementation of the example meter 112 of Figure 1. In the example illustrated in Figure 3, the meter 112 includes a first microphone 114, a second microphone 116, a direction detector 300, a filter 302, a media device database 304, a code reader 306, a memory 308, and a data transmitter 310.

3 , the first microphone 114 and the second microphone 116 are located on different sides of the housing of the meter 112 and are separated by a distance (e.g., three inches). In some examples, the meter 112 is rotated 90 degrees (via a motor or other device) to change the orientation of the first microphone 114 and the second microphone 116. In other examples, the meter 112 remains stationary while the first microphone 114 and the second microphone 116 rotate within the meter's housing.

FIG4 illustrates one example implementation of a meter 112 in which the meter 112 is capable of rotating. In the example of FIG4 , the meter 112 is rotated so that the first microphone 114 and the second microphone 116 are oriented so that they are 'pointed' toward the media device 108 as illustrated in FIG4 . This can increase the accuracy of measurements taken by the example meter 112. As described below, in other examples, the meter 112 is rotated to address uncertainty.

As described above, front-to-back ambiguity may occur when the first microphone 114 and the second microphone 116 are oriented on a left-to-right line relative to each other, and the example meter 112 cannot determine whether the sound source is in front of or behind the line occupied by the first microphone 114 and the second microphone 116 (e.g., as shown in FIG. 2 for the first microphone 114 and the second microphone 116 and the sound sources 110 and 208). Side-to-side (e.g., left-right) ambiguity may occur when the first microphone 114 and the second microphone 116 are oriented on a front-to-back line relative to each other, and the example meter 112 cannot determine whether the sound source is to the right or left of the line occupied by the first microphone 114 and the second microphone 116. Thus, one or more ambiguity issues may be overcome when the example first microphone 114 and the second microphone 116 are rotated, for example, 90 degrees, on a regular, predetermined, non-regular, and/or manual basis.

For example, if the meter 112 is oriented as shown in FIG2 , this creates a front-to-back ambiguity as explained in conjunction with FIG2 . However, if the example first and second microphones 114, 116 or the example meter 112 are rotated 90 degrees, the change in orientation of the first and second microphones 114, 116 will create a left-right ambiguity. Using two different orientations to measure audio direction eliminates any ambiguity.

The example direction detector 300 illustrated in FIG3 detects the direction of origin of an audio signal received by a meter 112. An example implementation of the direction detector 300 is further discussed in conjunction with FIG5. The example filter 302 of FIG3 filters out audio signals received from one or more directions of origin (e.g., the filter 302 accepts audio signals received from within angle (A) 202 and ignores audio signals received from outside angle (A) 202 as shown in FIG2). The example filter 302 of FIG3 determines which audio signal origin directions to filter by accessing an example media device database 304, which stores information regarding the location of the meter 112 in the home 102, the orientation of the meter 112, and the location of the media device 108 in the home 102. The example filter 302 uses the information stored in the media device database 304 to determine the direction from the meter 112 where the media device 108 is located and what angle of the sound origin direction should be filtered out.

3 detects audio codes embedded in received audio signals that are not filtered out by the filter 302. One or more detected audio codes may be stored in the example memory 308 to facilitate periodic, scheduled, non-periodic, and/or manual transmission to the example data collection facility 130 via the example data transmitter 310. The embedded audio code may identify a television program or station being presented by the example media device 108.

In some examples, a signature generator is used to generate media signatures instead of or in addition to code reader 306. A media signature is a representation of some characteristic of a media signal (e.g., a characteristic of the signal's spectrum). Signatures can be thought of as fingerprints. They generally do not rely on the insertion of an identification code into the media, but rather preferably reflect intrinsic characteristics of the media and/or media signal. Systems utilizing codes and/or signatures for audience measurement are already known. See, for example, U.S. Pat. No. 5,481,294 to Thomas, which is incorporated herein by reference in its entirety.

5 is a block diagram of an implementation of the example direction detector 300 of FIG 3. The example direction detector 300 includes a first microphone receiver 500, a second microphone receiver 502, an audio sampler 504, an absolute value calculator 506, a threshold processor 508, a moving average calculator 510, an audio sample selector 512, an audio signal delayer 514, a correlation engine 516, an offset selector 518, and a post-processor 520.

In operation, the example direction detector 300 determines, for an audio signal received by the meter 112 from the audio source, a direction of origin of the audio source relative to the example meter 112. The example direction detector 300 of Figures 3 and 4 determines the audio signal origin direction by determining a time difference between when the audio signal is received by the example first microphone 114 and when the same audio signal is received by the example second microphone 116.

For example, FIG7 illustrates sound propagating toward example meter 112 from a first source 700, a second source 702, and a third source 704. In the example illustrated in FIG7 , first source 700 is equidistant from first microphone 114 and second microphone 116. Therefore, sound emanating from example first source 700 arrives at both example first microphone 114 and example second microphone 116 simultaneously. On the other hand, example second source 702 is farther from example first microphone 114 than from example second microphone 116. Therefore, sound emanating from second source 702 arrives at example second microphone 116 before arriving at example first microphone 114. In the example of FIG7 , third source 704 is also farther from first microphone 114 than from second microphone 116, but the distance difference is not as great as that of second source 702. Therefore, sound emanating from third source 704 arrives at second microphone 116 before arriving at first microphone 114, but the time difference is not as great as that of sound emanating from second source 702.

7 , assume that the speed of sound is 13,041.6 inches per second, that the first microphone 114 and the second microphone 116 are separated by three inches, and that the audio received by the meter 112 is sampled at 48,000 Hz (i.e., 48,000 samples per second). Therefore, in the example illustrated in FIG7 , the sound emanating from the second source 702 will have to travel three inches further to reach the first microphone 114 than to reach the second microphone 116. The amount of time it takes for the sound to travel three inches can be calculated by dividing three inches by 13,041.6 inches per second, which is approximately equal to 0.00023 seconds. Because the example meter 112 samples the received audio at 48,000 samples per second, multiplying 0.00023 seconds by 48,000 samples per second yields approximately 11 samples that will be received by the second microphone 116 from the second source 702 before the first sample from the second source 702 is received by the first microphone 114. Therefore, the audio signal received by the second microphone 116 from the second source 702, delayed by eleven samples, should match the audio signal received by the first microphone 114 from the second source 702.

Similar delays in audio samples between audio signals received by the example first microphone 114 and the example second microphone 116 can be determined for any other resource locations by determining the distances between the respective sources and the first microphone 114 and the second microphone 116 and performing similar calculations as disclosed above.

FIG8A is a graph 800 illustrating example delays, over multiple samples, between when an audio signal is received by an example first microphone 114 and when the audio signal is received by an example second microphone 116 for one or more placements of an audio source within a room. In the example graph 800 of FIG8A , the room is 12 feet long by 24 feet wide, and the example meter 24 is placed in the center of the room. Row 802 of the example graph 800 lists how far the audio source is from the left side of the room, with the example meter 24 placed between 5.2 and 5.7 feet from the left side of the room. Column 804 of the example graph 800 lists how far the audio source is from the front or back of the center of the room, where positive numbers represent audio sources located in front of the center of the room and negative numbers represent audio sources located behind the center of the room.

The values in the example graph 800 of FIG8A list the number of samples of the audio signal from the audio source that will arrive at the first microphone 114 before the audio signal arrives at the second microphone 116. A positive number indicates that the audio signal will arrive at the first microphone 114 before the second microphone 116. A negative number indicates that the audio signal will arrive at the second microphone 116 before the first microphone 114. For example, an audio source located 1.6 feet from the left side of the room and 4.8 feet in front of or behind the center of the room will send seven samples to the first microphone 114 (as shown at points 806 and 808 in FIG8 ) before the first sample of the audio signal arrives at the second microphone 116. Because the first and second microphones 114 and 116 are oriented from left to right in the example of FIG8 , the same result of 7 samples is obtained regardless of whether the audio source is located 4.8 feet in front of or behind the example meter 112. This illustrates an example of front-to-back uncertainty.

As shown in FIG8B , the direction detector 300 can generate a graph such as graph 800 to determine the angle of the sound source. In the example illustrated in FIG8B , if the direction detector 300 detects that the audio signal received by the second microphone 116 is delayed by seven samples from the audio signal received by the first microphone 114, the direction detector 300 determines that the audio is received from either angle (A) or angle (−A), because audio received from any other angle would have resulted in a different sample delay between the audio received by the first microphone 114 and the second microphone 116. Additionally, if the example direction detector 300 detects that the audio signal received by the second microphone 116 is delayed by seven or more samples from the audio signal received by the first microphone 114, the direction detector 300 determines that the audio signal is received from within angle (B), because audio sources received from outside angle (B) would have a delay of less than seven samples between the audio received by the first microphone 114 and the second microphone 116. If, for example, the example media device 108 is located within angle (B), the example filter 302 may filter out audio received from angles outside of angle (B).

5 , the illustrated example first microphone receiver 500 receives an audio signal detected by the first microphone 114. The second microphone receiver 502 of FIG5 receives an audio signal detected by the second microphone 116. The audio signals received by the first microphone receiver 500 and the second microphone receiver 502 are analog audio signals emitted by the media device speakers 110, overflow media device speakers 124, audience members 118, 120, 126, or other audio sources in the home 102.

The audio sampler 504 of the illustrated example converts the analog audio signals received by the first microphone receiver 500 and the second microphone receiver 502 into digital audio signals by sampling the received analog audio signals. In the illustrated example, the sampling rate is 48,000 Hz (i.e., the analog audio signals are sampled 48,000 times per second). In other examples, other sampling rates can be used. The absolute value calculator 506 of the illustrated example calculates the absolute value of the audio signal.

The threshold processor 508 of the illustrated example determines a decimation threshold for the audio signal and eliminates audio samples with audio intensity values below the determined decimation threshold. In the example illustrated in FIG5 , the threshold processor 508 determines one decimation threshold for the audio signal received by the first microphone receiver 500 and another decimation threshold for the audio signal received by the second microphone receiver 502. Alternatively, the threshold processor 508 can determine the same decimation threshold for the audio received by the first microphone receiver 500 and the second microphone 502. In the illustrated example, the threshold processor 508 determines the decimation threshold for the audio signal so that 98.5% of the samples of the audio signal are below the decimation threshold and 1.5% of the samples of the audio signal are above the decimation threshold. In other examples, other threshold levels can be determined.

The illustrated example moving average calculator 510 calculates a moving average of an audio signal. The example moving average calculator 510, acting as a low-pass filter, can implement the moving average by taking the average of an audio sample and one or more adjacent audio samples for each sample in the audio signal. In the example illustrated in FIG5 , the moving average calculator 510 takes a moving average of two samples. In other examples, the moving average calculator 510 can take a moving average of a different number of samples.

The illustrated example audio sample selector 512 selects a plurality of samples (e.g., 1,000 samples) that are equally spaced across the audio signal (e.g., a 36-second audio segment received by the example first microphone receiver 500). In some examples, the audio sample selector 512 selects a plurality of samples that are unequally spaced across the audio signal (e.g., more samples may be selected from a particular portion of the audio signal, and fewer samples may be selected from other portions of the audio signal). In the example illustrated in FIG. 5 , the audio sample selector 512 selects 1,000 samples. In other examples, other numbers of samples may be selected. The example audio sample selector 512 selects two or more sets of 1,000 samples, where each set of samples is offset by a different length of time. For example, the audio sample selector 512 first selects 1,000 samples from the audio signal, where the selected audio samples are equally spaced between a first sample of the audio signal and another later sample of the audio signal (e.g., 36 seconds into the audio signal). The example audio sample selector 512 may later select 1,000 samples from the audio signal, wherein the selected audio samples are equally spaced between a sample after the first sample of the audio signal (e.g., a sample 0.5001 seconds into the audio signal) and another later sample (e.g., a sample 36.5 seconds into the audio signal). The example audio sample selector 512 may make additional selections of 1,000 samples, wherein each subsequent selection of 1,000 samples is offset from the previously selected 1,000 samples by an additional time (e.g., 0.5001 seconds).

The illustrated example audio signal delayer 514 selects audio samples from one audio signal that are delayed by a plurality of samples of another audio signal. For example, if the example audio sample selector 512 selects 1,000 samples from the audio signal received by the first microphone receiver 500, where the 1,000 samples are received at 1,000 different time points, the example audio signal delayer 514 may select 1,000 samples from the audio signal received by the second microphone 502 at the same 1,000 time points. The example audio signal delayer 514 may later select 1,000 samples from the audio signal received by the second microphone 502 that are one sample later than each of the 1,000 samples previously selected by the audio signal delayer 514 (i.e., the audio signal is delayed by one sample). The example audio signal delayer 514 may later select 1,000 samples from the audio signal received by the second microphone 502 that are one sample before each of the 1,000 samples initially selected by the audio signal delayer 514 (ie, the audio signal is delayed by -1 sample).

The illustrated example correlation engine 516 determines a correlation between two audio signals. In the illustrated example, the correlation engine 516 determines a correlation between 1,000 samples of the audio signal received by the first microphone receiver 500 and 1,000 samples of the audio signal received by the second microphone receiver 502 that are offset by a number of samples. In the illustrated example, the correlation engine 516 determines multiple correlations between samples selected by the audio sample selector 512 from the first microphone receiver 500 and samples selected by the audio signal delayer 514 from the second microphone receiver 502, wherein the audio signal delayer 514 delays the audio signal received by the second microphone receiver 502 before each correlation is calculated. The example correlation engine 516 can employ any type of statistical and/or correlation algorithm on the received audio signals, such as, but not limited to, normalized correlation, Pearson correlation coefficient, and/or rank correlation coefficient.

The offset selector 518 of the illustrated example determines which of the correlations between the set of samples selected by the audio sample selector 512 and the plurality of sets of samples selected by the audio signal delay 514, as determined by the correlation engine 516, has a highest correlation value. That is, the offset selector 518 of the illustrated example determines the sample offset of the audio signal received by the example second microphone receiver 502 that most closely correlates with the audio signal received by the example first microphone receiver 500.

The illustrated example post-processor 520 processes the output of the offset selector 518. An example implementation of the post-processor 520 is further discussed in conjunction with FIG.

6 is a block diagram of an implementation of the example post-processor 520 of FIG 5. The example post-processor 520 includes an excursion analyzer 600, a cluster analyzer 602, and a weighted average 604.

The example offset analyzer 600 of the illustrated example counts the number of times each offset in the samples is determined by the offset selector 518 to determine an offset count for each determined offset (e.g., three times for determining an offset of 7 samples, twice for determining an offset of 3 samples, once for determining an offset of -4 samples, etc.). One or more of these offset counts are then sent to the example cluster analyzer 602. In the illustrated example, counts for the three most frequent offsets (i.e., the three offsets most frequently determined by the example offset selector 518) are sent to the cluster analyzer 602. In other examples, a different number of counts may be sent to the cluster analyzer 602.

The illustrated example cluster analyzer 602 determines whether each of the offset counts received from the offset analyzer 600 should be accepted into a cluster. The offset counts sent from the example offset analyzer 600 are received by the example cluster analyzer 602 one at a time. When the first offset count is received by the example cluster analyzer 602, the offset count is accepted into the cluster. When subsequent offset counts are received by the example cluster analyzer 602, the received offset count is accepted into the cluster if and only if the difference between the offset corresponding to the offset count and the offset corresponding to at least one of the other offset counts already in the cluster is less than or equal to two. After the example cluster analyzer 602 analyzes each of the offset counts sent from the example offset analyzer 600, the offset counts accepted into the cluster and the corresponding offsets are sent to the example weighted averager 604.

The illustrated example weighted averager 604 calculates a weighted average of the offsets accepted into the cluster by the cluster analyzer 602, where each offset is weighted by its corresponding offset count. The weighted average output by the example weighted averager 604 can be used to determine the angle of the audio source by, for example, using the example table 800 disclosed in conjunction with FIG. 8A .

Although Figures 1 to 7 illustrate example ways of implementing the audience measurement system of Figure 1, one or more of the elements, processes and/or devices illustrated in Figures 1 to 7 may be combined, divided, rearranged, omitted, eliminated and/or implemented in any other manner. Furthermore, the example meter 112, the example first microphone 114, the example second microphone 116, the example data collection facility 130, the example server 132, the example database 134, the example direction detector 300, the example filter 302, the example media device database 304, the example code reader 306, the example memory 308, the example data transmitter 310, the example first microphone receiver 500, the example second microphone receiver 502, the example audio sampler 504, the example absolute value calculator 506, the example threshold processor 508, the example moving average calculator 510, the example audio sample selector 512, the example audio signal delayer 514, the example correlation engine 516, the example offset selector 518, the example post-processor 520, the example offset analyzer 600, the example cluster analyzer 602, and/or the example weighted averager 604 may be implemented by hardware, software, firmware, and/or any combination of hardware, software, and/or firmware. Thus, for example, any of the following items can be implemented by one or more analog or digital circuits, logic circuits, programmable processors, application specific integrated circuits (ASICs), programmable logic devices (PLDs), and/or field programmable logic devices (FPLDs): the example meter 112, the example first microphone 114, the example second microphone 116, the example data collection facility 130, the example server 132, the example database 134, the example direction detector 300, the example filter 302, the example media device database 304, the example code reader 306, the example storage 308, an example data transmitter 310, an example first microphone receiver 500, an example second microphone receiver 502, an example audio sampler 504, an example absolute value calculator 506, an example threshold processor 508, an example moving average calculator 510, an example audio sample selector 512, an example audio signal delayer 514, an example correlation engine 516, an example offset selector 518, an example post-processor 520, an example offset analyzer 600, an example cluster analyzer 602, and/or an example weighted averager 604, and/or an example audience measurement system more generally. When any of the apparatus claims or system claims of this patent are read in relation to a purely software and/or firmware implementation, the example meter 112, the example first microphone 114, the example second microphone 116, the example data collection facility 130, the example server 132, the example database 134, the example direction detector 300, the example filter 302, the example media device database 304, the example code reader 306, the example memory 308, the example data transmitter 310, the example first microphone receiver 500, the example second microphone receiver 502, the example audio sampler 504, the example absolute value calculator 50 6, at least one of the example threshold processor 508, the example moving average calculator 510, the example audio sample selector 512, the example audio signal delayer 514, the example correlation engine 516, the example offset selector 518, the example post-processor 520, the example offset analyzer 600, the example cluster analyzer 602, and/or the example weighted averager 604, and/or the example audience measurement system more generally is hereby expressly defined as comprising a tangible computer-readable storage device or storage disk storing software and/or firmware, such as a memory, a digital versatile disk (DVD), a compact disk (CD), a Blu-ray disk, etc. Furthermore, the example audience measurement system of FIG. 1 may include one or more elements, processes, and/or devices in addition to or in place of the elements, processes, and/or devices illustrated in FIG. 1 to FIG. 7, and/or may include more than one of any or all of the illustrated elements, processes, and devices.

Flowcharts representing example machine-readable instructions for implementing the example meter 112 of FIG1 through FIG3, the example direction detector 300 of FIG3 and FIG5, and the example post-processor 520 of FIG5 and FIG6 are shown in FIG9 through FIG12. In these examples, the machine-readable instructions comprise a program executed by a processor, such as the processor 1412 shown in the example processor platform 1400 discussed below in conjunction with FIG14. The program may be implemented as software stored on a tangible computer-readable storage medium, such as a CD-ROM, floppy disk, hard drive, digital versatile disk (DVD), Blu-ray disk, or memory associated with the processor 1412, although the entire program and/or portions thereof may alternatively be executed by a device other than the processor 1412 and/or implemented in firmware or dedicated hardware. Furthermore, while the example program is described with reference to the flowcharts illustrated in FIG9 through FIG12, many other methods of implementing the example meter 112 of FIG1 through FIG3, the example direction detector 300 of FIG3 and FIG5, and the example post-processor 520 of FIG5 and FIG6 may alternatively be used. For example, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, or combined.

As described above, the example processes of Figures 9-12 can be implemented using encoded instructions (e.g., computer and/or machine-readable instructions) stored on a tangible computer-readable storage medium such as a hard drive, flash memory, read-only memory (ROM), compact disk (CD), digital versatile disk (DVD), cache memory, random access memory (RAM), and/or any other storage device or storage disk that stores information for any duration (e.g., for an extended period of time, permanently, for a short period of time, temporarily buffered, and/or cached information). As used herein, the term tangible computer-readable storage medium is expressly defined to include any type of computer-readable storage device and/or storage disk, but excludes propagating signals. As used herein, "tangible computer-readable storage medium" and "tangible machine-readable storage medium" are used interchangeably. Additionally or alternatively, the example processes of Figures 9 to 12 may be implemented using encoded instructions (e.g., computer-readable instructions and/or machine-readable instructions) stored on a non-transitory computer- and/or machine-readable medium such as a hard drive, flash memory, read-only memory, compact disk, digital versatile disk, cache memory, random access memory, and/or any other storage device or storage disk that stores information for any duration (e.g., for an extended period of time, permanently, for a short period of time, temporarily buffered, and/or cached information). As used herein, the term non-transitory computer-readable medium is expressly defined to include any type of computer-readable device or disk, but excludes propagating signals. As used herein, when the phrase "at least" is used as a transition term in the preamble of a claim, it is open ended in the same manner as the term "comprising" is open ended.

FIG9 is a flow diagram representing example machine-readable instructions for implementing the example meter 112 of FIG1-3. FIG9 begins when the example meter 112 determines whether the first microphone 114 and the second microphone 116 have received an audio signal (block 900). If the example meter 112 determines that audio has not been received (block 900), the example meter 112 waits until audio has been received. If the example meter 112 determines that audio has been received (block 900), the example direction detector 300 determines the direction (i.e., angle) from which the audio signal was received (block 902). An example method for implementing block 902 of FIG9 is further discussed in conjunction with FIG11.

After the example direction detector 300 determines the direction of origin of the received audio signal (block 902), the example filter 302 determines a threshold angle (e.g., an angle relative to the example meter 112 where the example media device 108 is located) by accessing the example media device database (block 904). The example filter 302 then determines whether the direction of origin determined by the example direction detector 300 is within the threshold angle (block 906).

If the example direction detector 300 determines that the direction of origin determined by the example direction detector 300 is not within the threshold angle (block 906), control passes to block 916. If the example direction detector 300 determines that the direction of origin determined by the example direction detector 300 is within the threshold angle (block 906), the example code reader 306 uses known techniques to detect an audio code embedded in the audio signal received by the example first microphone 114 and the second microphone 116 (block 908). The example code reader 306 then stores the detected audio code in the example memory 308 (block 910).

The example meter 112 then determines whether to send the data to the example data collection facility 130 (block 912). This determination can be based on the time of day (e.g., data is sent every day at 12:00 AM), the time since the last data transmission (e.g., data is sent every hour), the amount of data in the example memory 308 (e.g., data is sent when the example memory 308 is full), or other factors. If the example meter 112 determines that the data is to be sent (block 912), the example data transmitter 310 sends the data in the example memory 308 to the example data collection facility 130 (block 914). If the example meter 112 determines that the data is not to be sent (block 912), control passes to block 916.

After the example data transmitter 310 sends the data in the example memory 308 to the example data collection facility 130 (block 914), or after the example meter 112 determines that the data will not be sent (block 912), or after the example filter 302 determines that the origin direction determined by the example direction detector 300 is not within the threshold angle (block 906), the example meter 112 determines whether to continue operation (block 916). This determination can be based on whether the example media device 108 is turned off or other factors. If the example meter 112 determines to continue operation (block 916), control returns to block 900. If the example meter 112 determines not to continue operation (block 916), the example of FIG. 9 ends.

FIG10 is a flow chart showing an example machine-readable instruction for implementing the example meter 112 of FIG1 to FIG3 to determine whether the example media device 108 of FIG1 is on or off. FIG10 begins (block 1000) when the example meter 112 determines whether the first microphone 114 and the second microphone 116 have received an audio signal. If the example meter 112 determines that audio has not been received (block 1000), the example meter 112 waits until audio has been received. If the example meter 112 determines that audio has been received (block 1000), the example direction detector 300 determines the direction (i.e., angle) of origin of the received audio signal (block 1002). An example method for implementing block 1002 of FIG10 is further discussed in conjunction with FIG11.

After the example direction detector 300 determines the direction of origin of the received audio signal (block 1002), the example filter 302 determines a threshold angle (e.g., an angle relative to the example meter 112 where the example media device 108 is located) by accessing the example media device database (block 1004). The example filter 302 then determines whether the direction of origin determined by the example direction detector 300 is within the threshold angle (block 1006).

If the example direction detector 300 determines that the originating direction determined by the example direction detector 300 is within the threshold angle (block 1006), the example direction detector 300 determines to turn on the media device 108 (block 1008). If the example direction detector 300 determines that the originating direction determined by the example direction detector 300 is not within the threshold angle (block 1006), the example direction detector 300 determines to turn off the media device 108 (block 1010). After the example direction detector 300 determines to turn on the media device 108 (block 1008) or after the direction detector 300 determines to turn off the media device 108 (block 1010), the example of FIG. 10 ends.

FIG11 is a flow diagram representing example machine-readable instructions for implementing the example direction detector 300 of FIG3 and FIG4. FIG11 begins when the example first microphone receiver 500 receives an audio signal (a first audio signal) detected by the first microphone 114 and the second microphone receiver 502 receives an audio signal (a second audio signal) detected by the second microphone 116 (block 1100). The example audio sampler 504 then samples the first audio signal and the second audio signal to convert the received audio signals from analog audio signals to digital audio signals (block 1102). The example absolute value calculator 506 then determines the absolute values of the first audio signal and the second audio signal (block 1104).

The example threshold processor 508 then determines a decimation threshold for the first audio signal received by the first microphone receiver 500 (block 1106). In the illustrated example, the threshold processor 508 determines a decimation threshold audio intensity such that, for example, 98.5% of the audio samples in the first audio signal fall below the decimation threshold. The example threshold processor 508 then determines a decimation threshold audio intensity for the second audio signal received by the second microphone receiver 502 (block 1108). In the illustrated example, the threshold processor 508 determines a decimation threshold audio intensity such that, for example, 98.5% of the audio samples in the second audio signal fall below the decimation threshold. The example threshold processor 508 then removes audio samples having intensities below the respective determined decimation thresholds from the first and second audio signals (block 1110). The example moving average calculator 510 then determines a moving average of the first and second audio signals (block 1112). In the illustrated example, the moving average calculator 510 determines a moving average of the first audio signal and the second audio signal using sample sizes of the first audio signal and the second audio signal to determine the moving average.

The example audio sample selector 512 then selects a subset of samples from the first audio signal (block 1114). In the illustrated example, the audio sample selector 512 selects, for example, 1,000 samples from the first audio signal at 1,000 equally spaced time points over a segment of the first audio signal (e.g., over the first 36 seconds of the received audio signal). The example audio signal delayer 514 then selects the same number of samples from the second audio signal at the same time points but delayed by a certain number of samples (block 1116). In the example illustrated in FIG11 , the audio signal delayer 514 selects 1,000 samples from the second audio signal at the same time points as the 1,000 samples from the first audio signal but delayed by a certain number of samples. In the illustrated example, as shown in FIG8A , the distance between the first microphone 114 and the second microphone 116 is three inches, and depending on the location of the sound source, the delay between the audio signals arriving at the first microphone 114 and the second microphone 116 can be as much as eleven samples. Thus, in the illustrated example, the audio samples selected from the second audio signal by the audio signal delayer 514 are delayed by between -11 and 11 samples compared to the samples selected from the first audio signal by the audio sample selector 512 .

After the example audio sample selector 512 selects a set of samples from the first audio signal and the example audio signal delayer 514 selects a set of samples from the second audio signal, the example correlation engine 516 determines a correlation between the two sets of audio samples (block 1118). The example correlation engine 516 can employ any type of statistical and/or correlation algorithm on the received audio signals, such as, but not limited to, normalized correlation, Pearson correlation coefficient, and/or rank correlation coefficient.

After the example correlation engine 516 determines the correlation between the two sets of audio samples (block 1118), the example offset selector 518 determines whether additional correlations are needed between the first audio signal and the second audio signal that are offset by a different number of samples. In the illustrated example, as described above, correlations are performed between samples from the first audio signal and each of the sets of samples from the second audio signal that are offset by a different number of samples from -11 to 11 (i.e., 23 different sets of samples from the second audio signal). In the illustrated example, in block 1118, the offset selector 518 determines whether the correlation engine 516 has determined correlations between the set of samples from the first audio signal and all 23 sets of samples from the second audio signal. If the example offset selector 518 determines that additional offsets of the second audio signal need to be processed by the example correlation engine 516 (block 1120), control returns to block 1116, and an additional set of samples from the second audio signal that are offset by a different number of samples is selected by the audio signal delay 514. If the example offset selector 518 determines that no additional offsets of the second audio signal need to be processed by the example correlation engine 516 (block 1120 ), control passes to block 1122 .

In block 1122, the example offset selector selects the offset number of samples that produces the highest correlation value with samples from the first audio signal (e.g., an offset of samples between -11 and 11). The example offset selector stores the offset value. As discussed below, the example audio sample selector 512 determines whether additional offsets of the first audio signal need to be considered (block 1124).

In block 1122, the example offset selector 512 selects the offset for the sample of the second audio signal that best correlates with the sample from the first audio signal. However, this correlation only considers a portion of the total samples from the first audio signal. For example, if a 36-second audio segment is sampled at 48,000 samples per second, the audio segment will have 1,728,000 samples, but the correlation may only consider 1,000 of these samples. Therefore, better results may be obtained by considering a different set of samples from the first audio signal (e.g., a different 1,000 samples).

In the illustrated example, twenty such sets of 1,000 samples from the first audio signal are considered, and each of the twenty sample sets is correlated with (i.e., offset between -11 and 11 samples from) a set of 23 samples from the second audio signal. For each of the twenty sample sets from the first audio signal, the best offset value among the 23 sample sets from the second audio signal (e.g., the offset that best correlates with the sample set from the first audio signal) is selected, and the twenty offsets are combined to determine the best overall offset.

To achieve optimal results, the audio sample selector 512 selects twenty different sets of samples from the first audio signal such that there is minimal possible overlap between the samples contained in each of the twenty sample sets. In the illustrated example, each of the sets of 1,000 samples from the first audio signal is offset from one another by 0.5001 seconds. The value of 0.5001 is selected in the illustrated example because offsetting the sets of 1,000 samples by this amount of time produces a relatively small amount of overlap between the samples in the sample sets. In other examples, such as where a different number of samples is selected in each sample set or where a different sampling rate is used, a different offset time value may be selected.

Returning to FIG11 , if the example audio sample selector 512 determines that an additional set of samples from the first audio signal is to be considered (block 1124), control returns to block 1114, and the example audio sample selector 512 selects an additional set of samples from the first audio signal offset by a certain amount (e.g., 0.5001 seconds) from the previously selected set of samples. If the example audio sample selector 512 determines that an additional set of samples from the first audio signal is not to be considered (block 1124), each of the offsets selected by the example offset selector 518 (e.g., each of the 20 offset values) is processed by the example post-processor 520 (block 1126). An example method for implementing block 1126 is further discussed in conjunction with FIG12 . After the example post-processor 520 determines the total sample offset between the first audio signal and the second audio signal (block 1126), the example direction detector 300 determines the direction of origin for the source of the received audio signal by, for example, consulting a table such as table 800 of FIG8A or using an algorithm (block 1128). Then, the example of FIG. 11 ends.

FIG12 is a flow diagram representing example machine-readable instructions for implementing the example post-processor 520 of FIG4 and FIG5. FIG12 begins when the example offset analyzer 600 determines the frequency of each of the sample offsets output by the example offset selector 518 of FIG5 (block 1202). FIG13 illustrates an example of such a determination.

FIG13 illustrates an example table 1300 containing sample data output by the example offset selector 518 of FIG5 . Column 1304 of table 1300 lists the sample offset values output by the example offset selector 518. Column 1302 of table 1300 lists the number of times each of the offsets is output by the example offset selector 518. In the example of table 1300, an offset of -12 samples is selected three times by the example offset selector 518, and an offset of -13 samples is selected twice by the example offset selector 518.

Returning to FIG12 , after the example offset analyzer 600 determines the frequency of each offset output by the example offset selector 518 (block 1202), the example cluster analyzer 602 loads the most frequent offset (e.g., the offset value selected the most times by the example offset selector 518). In the example of FIG13 , the most frequent offset is −12 samples, which is selected three times by the example offset selector. The example cluster analyzer 602 then determines whether to accept the offset into the cluster (block 1206). The first time an offset is considered by the example cluster analyzer 602, the cluster analyzer 602 accepts the offset into the cluster. If the offset value being considered is within two samples of one of the other offset values in the cluster, then subsequent offset values considered by the example cluster analyzer 602 will be accepted into the cluster. In the example table 1300 of FIG13 , the offset value of −12 samples will be accepted into the cluster analyzer because it will be the first value considered. In the example table 1300 of Figure 13, an offset value of -13 samples would be considered next, and because -13 is within two of -12, the offset value of -13 samples would be accepted into the cluster. In the example table 1300 of Figure 13, an offset value of 20 would be considered next, and because 20 is not within two of -13 or -12, the offset value of 20 would be accepted into the cluster.

12 , if the example cluster analyzer 602 determines not to accept the offset value being considered into the cluster (block 1206), control passes to block 1210. If the example cluster analyzer 602 determines to accept the offset value being considered into the cluster (block 1206), the cluster is updated with a new offset value and offset count (e.g., an offset value of -12 and an offset count of 3 in the example table 1300 of FIG. 13 ) (block 1208).

The example cluster analyzer 602 then determines whether to consider additional offsets (block 1210). In the illustrated example, the cluster analyzer 602 determines the three most frequent offset values output by the example offset selector 518. In other examples, a different number of most frequent offset values may be considered. In the illustrated example, if more than three offsets are selected three times most frequently (e.g., if there is a tie in table 1300 of FIG. 13 where -13 samples, 20 samples, -6 samples, and -14 samples are all selected twice by the example offset selector 518), the cluster analyzer 602 randomly selects an offset value to break the tie (e.g., in the example table 1300 of FIG. 13, offset values of -13 samples and 20 samples are selected along with -12 samples). In other examples, when there is a tie among the most frequently selected offset values, additional offset values (e.g., greater than three) may be selected.

If the example cluster analyzer 602 determines that additional offsets are to be considered (block 1210), control returns to block 1204. If the example cluster analyzer 602 determines that additional offsets are not to be considered (block 1210), the example weighted averager 604 takes a weighted average of the offset values in the cluster weighted by the offset count. In the example table 1300 of FIG. 13 , as discussed above, an offset of -12 samples is accepted into the cluster along with an offset count of 3 and a sample of -13 is accepted into the cluster along with an offset count of 2. Therefore, in the example table 1300 of FIG. 13 , the example weighted averager 604 calculates the weighted average as ((3*-12)+(2*-13))/(3+2), which yields a value of -12.4 samples. The example of FIG. 12 then ends.

The example methods, devices, systems, and/or articles disclosed herein enable determining the direction of origin of an audio source using a relatively small number of samples of an audio signal emitted by the audio source. For example, the illustrated example audience measurement system can determine the direction of origin of a 36-second audio segment from an audio source sampled at 48,000 samples per second, resulting in 1,728,000 samples. The illustrated example audience measurement system determines the direction of origin of an audio source by calculating the correlation between 20 sets of audio samples detected by two microphones, where each set of 20 audio samples contains 1,000 audio samples. This results in significantly improved efficiency, reduced processing time, and/or fewer resources required compared to a measurement system that uses the full 1,728,000 samples to calculate the correlation between the audio signals received by the two microphones.

14 is a block diagram of an example processor platform 1400 capable of executing the instructions of FIG9-12 to implement the example meter 112, the examples of FIG1 and FIG3, the example direction detector 300 of FIG3 and FIG5, and the example post-processor 520 of FIG5 and FIG6. The processor platform 1400 can be, for example, a server, a personal computer, a mobile device (e.g., a cell phone, a smartphone, a tablet such as an iPad™), a personal digital assistant (PDA), an internet device, a DVD player, a CD player, a digital video recorder, a Blu-ray player, a game console, a personal video recorder, a set-top box, or any other type of computing device.

The illustrated example processor platform 1400 includes a processor 1412. The illustrated example processor 1412 is hardware. For example, the processor 1412 may be implemented by one or more integrated circuits, logic circuits, microprocessors, or controllers from any desired family or manufacturer.

The processor 1412 of the illustrated example includes a local memory 1413 (e.g., a cache memory). The processor 1412 of the illustrated example communicates with a main memory including a volatile memory 1414 and a non-volatile memory 1416 via a bus 1418. The volatile memory 1414 may be implemented by synchronous dynamic random access memory (SDRAM), dynamic random access memory (DRAM), RAMBUS dynamic random access memory (RDRAM), and/or any other type of random access memory device. The non-volatile memory 1416 may be implemented by flash memory and/or any other desired type of memory device. Access to the main memories 1414 and 1416 is controlled by a memory controller.

The processor platform 1400 of the illustrated example also includes an interface circuit 1420. The interface circuit 1420 may be implemented by any type of interface standard, such as an Ethernet interface, a universal serial bus (USB), and/or a PCI express interface.

In the illustrated example, one or more input devices 1422 are connected to the interface circuitry 1420. The input devices 1422 enable a user to enter data and commands into the processor 1412. The input devices may be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, keys, a mouse, a touch screen, a trackpad, a trackball, isopoint, and/or a voice recognition system.

One or more output devices 1424 are also connected to the illustrated example interface circuit 1420. Output device 1424 can be implemented, for example, by a display device (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display, a cathode ray tube display (CRT), a touch screen, a tactile output device, a light emitting diode (LED), a printer, and/or a speaker). Therefore, the illustrated example interface circuit 1420 typically includes a graphics driver card, a graphics driver chip, or a graphics driver processor.

The illustrated example interface circuitry 1420 also includes communication devices such as transmitters, receivers, transceivers, modems, and/or network interface cards to facilitate exchanging data with an external machine (e.g., any type of computing device) via a network 1426 (e.g., an Ethernet connection, a digital subscriber line (DSL), a telephone line, a coaxial cable, a cellular telephone system, etc.).

The illustrated example processor platform 1400 also includes one or more mass storage devices 1428 for storing software and/or data. Examples of these mass storage devices 1428 include floppy disk drives, hard drive disks, compact disk drives, Blu-ray disk drives, RAID systems, and digital versatile disk (DVD) drives.

The encoded instructions 1432 of Figures 9-12 may be stored in the mass storage device 1428, the volatile memory 1414, the non-volatile memory 1416, and/or on a removable, tangible computer-readable storage medium such as a CD or DVD.

Although certain example methods, apparatus, and articles of manufacture have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus, and articles of manufacture fairly falling within the scope of the claims of this patent.

Claims (39)

1. A method for distinguishing local media from overflow media, the method comprising the following steps: A first audio signal from a first source received from a first microphone is sampled, and the sampling of the first audio signal generates a first audio sample; The second audio signal received from the first source by the second microphone is sampled, wherein the second microphone is separated from the first microphone by a first distance, and the sampling of the second audio signal generates a second audio sample; The first offset value is selected based on the first correlation value between the first audio sample and the second audio sample; Determine the count of the first offset value; If the first offset value is within a first number of samples of the second offset value in the cluster, then the first offset value is associated with the cluster; Calculate the weighted average of the first offset value and the second offset value; The origin direction of the first source is calculated based on the weighted average. When the origin direction is within a threshold angle, detect audio code embedded in at least one of the first audio signal or the second audio signal, the audio code identifying media associated with the first source; and When the origin direction exceeds the threshold angle, the media associated with the first source is ignored. 2. The method of claim 1, wherein the first source is a media device. 3. The method of claim 1, wherein the threshold angle is based on the physical location of the first source associated with the media. 4. The method according to claim 1, further comprising the following steps: Calculate the absolute values of the first audio signal and the second audio signal; and The first threshold and the second threshold are applied to the first audio signal and the second audio signal, respectively. 5. The method according to claim 4, further comprising the following step: Remove samples with intensities below the first threshold from the first audio samples; and Remove samples with an intensity below the second threshold from the second audio samples. 6. The method of claim 5, wherein the first threshold and the second threshold are determined such that the step of removing samples from the first audio sample and the second audio sample removes a majority of the samples from the first audio sample and the second audio sample. 7. The method according to claim 1, further comprising the following steps: Calculate the moving average of the first audio signal and the second audio signal; and Based on the moving average, the first correlation value between the first audio sample and the second audio sample is determined. 8. The method of claim 7, wherein each point of the moving average of the first audio signal and the second audio signal comprises the average of two samples from the first audio signal and the second audio signal, respectively. 9. The method of claim 1, wherein the first audio sample comprises samples from the first audio signal that are equally spaced between a first time point and a second time point. 10. The method of claim 9, wherein the second audio sample comprises samples from the second audio signal that are equally spaced between the first time point and the second time point. 11. The method according to claim 10, further comprising the step of: Select a third audio sample from the second audio signal, wherein the third audio sample is offset from the second audio sample by a second number of samples; and Determine a second correlation value between the first audio sample and the third audio sample. 12. The method according to claim 11, further comprising the following step: The second offset value is selected based on the second correlation value. 13. The method according to claim 9, further comprising the following step: A fourth audio sample is selected from the first audio signal, wherein the fourth audio sample is offset from the first audio sample by a first time. 14. A tangible machine-readable storage medium comprising instructions that, when executed, cause a machine to perform at least the following operations: A first audio signal from a first source received from a first microphone is sampled, and the sampling of the first audio signal generates a first audio sample; The second audio signal received from the first source by the second microphone is sampled, wherein the second microphone is separated from the first microphone by a first distance, and the sampling of the second audio signal generates a second audio sample; The first offset value is selected based on the first correlation value between the first audio sample and the second audio sample; Determine the count of the first offset value; If the first offset value is within a first number of samples of the second offset value in the cluster, then the first offset value is associated with the cluster; Calculate the weighted average of the first offset value and the second offset value; The origin direction of the first source is calculated based on the weighted average. When the origin direction is within a threshold angle, detect audio code embedded in at least one of the first audio signal or the second audio signal, the audio code identifying media associated with the first source; and When the origin direction exceeds the threshold angle, the media associated with the first source is ignored. 15. The machine-readable storage medium of claim 14, wherein the first source is a media device. 16. The machine-readable storage medium of claim 14, wherein the threshold angle is based on the physical location of the first source associated with the medium. 17. The machine-readable storage medium of claim 14, wherein the instructions cause the machine to perform the following operations: Calculate the absolute values of the first audio signal and the second audio signal; and The first threshold and the second threshold are applied to the first audio signal and the second audio signal, respectively. 18. The machine-readable storage medium of claim 17, wherein the instructions cause the machine to perform the following operations: Remove samples from the first audio samples of the first audio signal that have an intensity below the first threshold; and Remove samples from the second audio samples of the second audio signal that have an intensity lower than the second threshold. 19. The machine-readable storage medium of claim 18, wherein the instructions cause the machine to determine the first threshold and the second threshold such that the step of removing samples from the first audio sample of the first audio signal and the second audio sample of the second audio signal removes a majority of samples from the first audio sample of the first audio signal and the second audio sample of the second audio signal. 20. The machine-readable storage medium of claim 14, wherein the instructions cause the machine to perform the following operations: Calculate the moving average of the first audio signal and the second audio signal; and The first correlation value between the first audio sample and the second audio sample is determined based on the moving average. 21. The machine-readable storage medium of claim 20, wherein each point of the moving average of the first audio signal and the second audio signal comprises the average of two samples, respectively, from the first audio signal and the second audio signal. 22. The machine-readable storage medium of claim 14, wherein the first audio sample comprises samples from the first audio signal that are equally spaced between a first time point and a second time point. 23. The machine-readable storage medium of claim 22, wherein the second audio sample comprises samples from the second audio signal that are equally spaced between the first time point and the second time point. 24. The machine-readable storage medium of claim 23, wherein the instructions cause the machine to perform the following operations: Select a third audio sample from the second audio signal, wherein the third audio sample is offset from the second audio sample by a second number of samples; and Determine a second correlation value between the first audio sample and the third audio sample. 25. The machine-readable storage medium of claim 24, wherein the instruction causes the machine to select a second offset value based on the second correlation value. 26. The machine-readable storage medium of claim 22, wherein the instruction causes the machine to select a fourth audio sample from the first audio signal, wherein the fourth audio sample is offset from the first audio sample by a first time. 27. An apparatus for distinguishing local media from overflow media, the apparatus comprising: An audio sampler samples a first audio signal from a first source received by a first microphone to generate a first audio sample, and samples a second audio signal from the first source received by a second microphone to generate a second audio sample, wherein the second microphone is separated from the first microphone by a first distance. An offset selector selects a first offset value based on a first correlation value between the first audio sample and the second audio sample; An offset analyzer that determines a count of the first offset value; A cluster analyzer associates the first offset value with the cluster if the first offset value is within a first number of samples of the second offset value in the cluster. A weighted averager that calculates a weighted average of the first offset value and the second offset value; A direction detector that calculates the origin direction of the first source based on the weighted average; and A code reader that detects audio codes embedded in at least one of the first audio signal or the second audio signal when the origin direction is within a threshold angle, the audio codes identifying media associated with the first source, and ignores the media associated with the first source when the origin direction exceeds the threshold angle. 28. The apparatus of claim 27, wherein the first source is a media device. 29. The apparatus of claim 27, wherein the threshold angle is based on the physical location of the first source associated with the media. 30. The apparatus of claim 27, further comprising: An absolute value calculator that calculates the absolute values of the first audio signal and the second audio signal; and A threshold processor that applies a first threshold and a second threshold to the first audio signal and the second audio signal, respectively. 31. The apparatus of claim 30, wherein the threshold processor is configured to remove samples of the first audio signal having an intensity lower than the first threshold from a first audio sample, and to remove samples of the second audio signal having an intensity lower than the second threshold from a second audio sample. 32. The apparatus of claim 31, wherein the first threshold and the second threshold are determined such that the threshold processor removes a majority of samples from the first audio sample of the first audio signal and the second audio sample of the second audio signal. 33. The apparatus of claim 27, further comprising: A moving average calculator that calculates the moving average of the first audio signal and the second audio signal; and A correlation engine determines a first correlation value between the first audio sample and the second audio sample. 34. The apparatus of claim 33, wherein the moving average calculator is used to calculate the moving average of the first audio signal and the second audio signal, such that each point of the moving average includes the average of two samples from the first audio signal and the second audio signal, respectively. 35. The apparatus of claim 27, wherein the audio sampler is configured to generate a first audio sample such that the first audio sample comprises samples from the first audio signal that are equally spaced between a first time point and a second time point. 36. The apparatus of claim 35, further comprising an audio signal delayer for generating a second audio sample such that the second audio sample comprises samples from the second audio signal that are equally spaced between the first time point and the second time point. 37. The apparatus of claim 36, wherein the audio signal delayer is used to select a third audio sample from the second audio signal that is offset from the second audio sample by a second number of samples, and the apparatus further comprises a correlation engine for determining a second correlation value between the first audio sample and the third audio sample. 38. The apparatus of claim 37, wherein the offset selector is further configured to select a second offset value based on the second correlation value. 39. The apparatus of claim 35, wherein the audio sampler selects a fourth audio sample from the first audio signal that is offset from the first audio sample by a first time.