Method and apparatus for improving detection of audio signatures

Abstract

Methods, apparatuses, and articles of manufacture are disclosed that improve detection of audio signatures. An example apparatus includes a TDOA determiner to determine a first time difference of arrival of a first audio sensor of a meter and a second audio sensor of the meter, and a second time difference of arrival of the first audio sensor and a third audio sensor of the meter, a TDOA matcher to determine a match by comparing the first time difference of arrival to a first virtual source time difference of arrival and a second virtual source time difference of arrival, identify a first virtual source location as a location of a media presentation device in response to determining that the first time difference of arrival matches the first virtual source time difference of arrival, and remove an audio recording of the second audio sensor to reduce a computational burden of a processor.

Description

[0001] Related applications

[0002] This patent claims priority and benefit to U.S. Patent Application No. 16 / 455,025, filed June 27, 2019, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This invention relates generally to media surveillance, and more specifically to methods and apparatus for improving the detection of audio signatures. Background Technology

[0004] Media monitoring companies want to understand how users interact with media devices such as smartphones, tablets, laptops, and smart TVs. To facilitate this monitoring, companies recruit teams and install meters at the media presentation locations of these teams. These meters monitor media presentations and transmit the media monitoring information to the monitoring company's central facility. This media monitoring information enables media monitoring companies to monitor ad play rates, determine ad effectiveness, determine user behavior, and identify purchasing behaviors associated with various demographic information, among other things. Attached Figure Description

[0005] Figure 1 This is a block diagram of an example system for improving the detection of audio signatures, constructed according to the teachings of the present invention.

[0006] Figure 2 yes Figure 1 A block diagram of an example meter.

[0007] Figure 3 yes Figure 2 A block diagram of an audio analyzer.

[0008] Figure 4 yes Figure 1 and / or Figure 2 Example front view of an example meter.

[0009] Figure 5 yes Figure 1 and / or Figure 2 Example rear view of an example meter.

[0010] Figure 6 yes Figure 1 and / or Figure 2 Example top view of an example meter.

[0011] Figure 7 It indicates that it is used for calibration. Figure 1 and / or Figure 2 A diagram of an example virtual audio source environment for a meter.

[0012] Figure 8 It means Figure 7 A diagram illustrating an example implementation of a virtual audio source environment.

[0013] Figure 9 This indicates that it can be executed to achieve [the desired result]. Figure 1 and / or Figure 2 A flowchart of an example machine-readable instruction to determine the virtual time difference of arrival of an audio sensor pair using a meter.

[0014] Figure 10 This indicates that it can be executed to achieve [the desired result]. Figure 1 and / or Figure 2 The flowchart shows an example meter for improving audio signature detection by removing unnecessary audio recordings.

[0015] Figure 11 It is capable of being implemented according to the teachings of the present invention. Figure 9 and / or Figure 10 A block diagram of an example processor platform with machine-readable instructions.

[0016] These figures are not drawn to scale. Generally, the same reference numerals are used throughout the figures and in the accompanying written description to refer to the same or similar parts.

[0017] When identifying multiple elements or components that can be individually mentioned, this document uses descriptors such as “first,” “second,” “third,” etc. Unless otherwise specified or understood based on the context of their use, such descriptors are not intended to assign any priority, physical order, or listing meaning, or to chronological order, but merely serve as markers to refer to multiple elements or components separately to facilitate understanding of the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in a particular embodiment, while the same element may be referred to in the claims by different descriptors (e.g., “second” or “third”). In such cases, it should be understood that the use of such descriptors is merely for the convenience of referring to multiple elements or components. Detailed Implementation

[0018] Fingerprint- or signature-based media surveillance techniques typically use one or more inherent characteristics of the monitored media to generate a substantially unique proxy for the media during a monitoring time interval. This proxy is called a signature or fingerprint and can take any form (e.g., a series of numerical values, waveforms, etc.) representing any aspect of the media signal(s) (e.g., audio and / or video signals forming the monitored media presentation). A signature can be a series of signatures collected consecutively within a timer interval. A good signature is repeatable when processing the same media presentation, but unique relative to other (e.g., different) presentations of other (e.g., different) media. Therefore, the terms “fingerprint” and “signature” are used interchangeably herein and are defined herein as meaning a proxy for the media generated by one or more inherent characteristics of the media.

[0019] Signature-based media surveillance typically involves determining (e.g., generating and / or collecting) one or more signatures representing media signals (e.g., audio and / or video signals) output by a monitored media device, and comparing the monitored signatures with one or more reference signatures corresponding to a known (e.g., reference) media source. Various comparison criteria, such as cross-correlation values, Hamming distance, etc., can be evaluated to determine whether a monitored signature matches a particular reference signature. When a match is found between a monitored signature and one of the reference signatures, the monitored media can be identified as corresponding to a specific reference media represented by the reference signature that matches the monitored signature. Because attributes such as media identifiers, presentation time, broadcast channel, etc., are collected for the reference signature, these attributes can then be associated with monitored media whose monitored signature matches the reference signature. Example systems for identifying media based on codes and / or signatures are already known and were first disclosed in Thomas's U.S. Patent 5,481,294, the entire contents of which are incorporated herein by reference.

[0020] Historically, audio fingerprinting technology has used the loudest parts of an audio signal (e.g., the parts with the most energy) to create fingerprints within a time segment. However, this approach has several significant limitations in certain situations. In some examples, the loudest parts of the audio signal may be associated with noise (e.g., unwanted audio) rather than the audio of interest. For instance, attempting to fingerprint media from a noisy area (e.g., a room with a group of people watching television), the loudest part of the captured audio signal might be the conversation between the group, rather than the audio signal itself. In this example, many samples of the audio signal would be background noise rather than the media, reducing the usefulness of the generated fingerprint. Therefore, fingerprints generated using existing methods often do not include samples in the higher frequency range.

[0021] The example methods and apparatuses disclosed herein overcome the aforementioned problems by reducing the computational burden on the processor by removing audio signals (e.g., audio recordings) from fingerprint processing based on the phase difference between transformed audio signals. The examples disclosed herein remove audio signals based on the phase difference between transformed audio signals, thereby increasing the accuracy of identifying media associated with fingerprints. Furthermore, the examples disclosed herein utilize transformed audio signals to generate fingerprints. Similarly, the examples disclosed herein utilize the peak values ​​of various portions of the transformed audio signal, thereby reducing the amount of audio to be processed during fingerprint recognition computation (e.g., the processor does not need to process the entire audio signal).

[0022] As used herein, “virtual source location” and “virtual audio source location” refer to the virtual (e.g., computer-generated) location of an audio source that generates virtual (e.g., computer-generated) audio. That is, “virtual audio source location” refers to the location of a computer-generated audio source based on known principles and properties of audio (e.g., speed of sound, etc.). As used herein, “media” refers to audio and / or visual (static or moving) content and / or advertising. In some examples, to identify watermarked media, one or more watermarks are extracted and used to access a reference watermark table mapped to media identification information.

[0023] Figure 1 This is a diagram illustrating an example audience measurement system for improving audio signature detection, constructed according to the teachings of the present invention. Figure 1 In the example shown, the example media presentation environment 102 includes example team members 104 and 106, an example media presentation device 110 that receives media from an example media source 112, and an example meter 114. The meter 114 identifies the media presented by the media presentation device 110 and reports media monitoring information to the example central facility 190 of the example audience measurement entity via the example gateway 140 and the example network 180.

[0024] exist Figure 1 In the example shown, the example media presentation environment 102 is a room in a family home (e.g., a room in a group member's residence, such as a room in the "Nelson Family" residence). Figure 1In the example shown, sample group members 104 and 106 from households have been statistically selected to form media viewership data (e.g., television viewership data) for the target audience / demographic information. People become group members, for example, through a user interface presented on a media device (e.g., through media presentation device 110, through a website, etc.). People become group members additionally or alternatively, such as through telephone interviews, completing online surveys, etc. Additionally or alternatively, people can be contacted and / or recruited using any desired method (e.g., random selection, statistical selection, telephone solicitation, internet advertising, questionnaires, mall advertising, product packaging, etc.). In some examples, the entire household can be registered as a group member. That is, while the mother, father, son, and daughter can each be identified as individual group members, their viewing activity typically occurs within a single household.

[0025] exist Figure 1 In the example shown, one or more group members 104, 106 of the household have registered with the audience measurement entity (e.g., by consenting to become group members) and have provided their demographic information as part of the registration process to the audience measurement entity so that the demographic information can be associated with media broadcasting activities (e.g., television broadcasting, radio broadcasting, internet broadcasting, etc.). Demographic data includes, for example, the group members' age, gender, income level, education level, marital status, geographic location, ethnicity, etc. Although the example media presentation environment 102 is in Figure 1 The example shown is a home, but the example media presentation environment 102 may additionally or alternatively be any other type of environment, such as a theater, restaurant, pub, retail space, arena, etc.

[0026] exist Figure 1 In the example shown, the example media presentation device 110 is a television set. However, the example media presentation device 110 can correspond to any type of audio, video, and / or multimedia presentation device capable of presenting media in an audible and / or visual manner. Figure 1 In the example shown, media presentation device 110 communicates with example audio / video receiver 118. In some examples, media presentation device 110 (e.g., a television) can transmit audio to another media presentation device (e.g., audio / video receiver 118) for output by one or more speakers (e.g., surround sound speakers, soundbars, etc.). As another example, media presentation device 110 may correspond to a multimedia computer system, personal digital assistant, cellular / mobile smartphone, radio, home theater system, audio and / or video stored and played back from memory (e.g., digital video recorder or digital multifunction reel), web pages, and / or any other communication device capable of presenting media to an audience (e.g., group members 104, 106).

[0027] Media presentation device 110 receives media from media source 112. Media source 112 can be any type of media provider(s), such as, but not limited to, cable media service providers, radio frequency (RF) media providers, Internet-based providers (e.g., IPTV), satellite media service providers, and / or any combination thereof. Media can be radio media, television media, pay-per-view media, movies, Internet Protocol Television (IPTV), satellite television (TV), Internet radio, satellite radio, digital television, digital radio, stored media (e.g., optical disc (CD), digital multi-disc (DVD), Blu-ray disc, etc.), any other type of broadcast, multicast and / or unicast media, audio and / or video media presented over the Internet (e.g., streaming), video games, directional broadcasting, satellite broadcasting, video on demand, etc. For example, media presentation device 110 may correspond to television and / or display devices that support standards such as: National Television Standards Committee (NTSC) standards, Phase Alternating Line (PAL) standards, Sequential Storage Color Television (SECAM) standards, standards developed by the Advanced Television Systems Committee (ATSC), such as High Definition Television (HDTV), standards developed by the Digital Video Broadcasting (DVB) project, etc. Advertising, such as advertisements and / or previews of other programs provided or to be provided by media source 112, is also typically included in the media.

[0028] In the examples disclosed herein, an audience measurement entity provides meter 114 to group members 104, 106 (or a group of household members), allowing group members 104, 106 to install meter 114 simply by powering it on and placing it in the media presentation environment 102 and / or near the media presentation device 110 (e.g., near a television set). In some examples, meter 114 may be provided to group members 104, 106 by an entity other than the audience measurement entity. In some examples, more complex installation activities may be performed, such as fixing meter 114 to media presentation device 110, electronically connecting meter 114 to media presentation device 110, etc. Example meter 114 detects media playback and electronically stores monitoring information of the presented media (e.g., codes detected by the presented media, signatures of the presented media, identifiers of group members present at the time of presentation, timestamps of the presentation time). The stored monitoring information is then sent back to central facility 190 via gateway 140 and network 180. Although in Figure 1 In the example shown, media monitoring information is transmitted electronically, but media monitoring information may additionally or alternatively be transmitted in any other way, such as via physical mail meter 114, via the memory of physical mail meter 114, etc.

[0029] The illustrated meter 114 combines audience measurement data and people measurement data. For example, audience measurement data is determined by monitoring the media output from media presentation device 110 and / or other media presentation devices(s), and audience identification data (also referred to as demographic data, people monitoring data, etc.) is determined from the people monitoring data provided to meter 114. Thus, the example meter 114 provides the dual functionality of an audience measurement meter that collects audience measurement data and a people meter that collects and / or associates demographic information corresponding to the collected audience measurement data.

[0030] For example, the meter 114 in the illustrated example collects media identification information and / or data (e.g., one or more signatures, one or more fingerprints, one or more codes, tuning channel identification information, playback time information, etc.) and personnel data (e.g., user identifiers, demographic data associated with audience members, etc.). Media identification information and personnel data can be combined to generate, for example, media playback data (e.g., viewership data), indicating the number and / or type of one or more people watching a specific segment of media distributed via media presentation device 110. To extract media identification data, Figure 1 The meter 114 shown in the example monitors a signature (sometimes called a fingerprint) included in the presentation media.

[0031] In the examples disclosed herein, in order to monitor media presented by media presentation device 110, the meter 114 of the illustrated example senses the media presented by media presentation device 110 and / or some other audio presentation system (e.g., Figure 1 The meter 114 outputs (e.g., transmits) audio (e.g., acoustic signals or ambient audio) from the audio / video receiver 118. For example, the meter 114 processes signals obtained from the media presentation device 110 to detect media and / or source identification signals (e.g., audio signatures) embedded in one or more portions (e.g., audio portions) of the media presented by the media presentation device 110. For example, to sense ambient audio output by the media presentation device 110, the meter 114 of the illustrated example includes multiple example audio sensors (e.g., one or more microphones and / or other acoustic sensors). In some examples, the meter 114 may process audio signals obtained from the media presentation device 110 via a direct cable connection to detect media and / or source identification audio watermarks embedded in these audio signals.

[0032] In some examples, the media presentation device 110 utilizes a rear-facing speaker. When a rear-facing speaker is used, using a forward-facing audio sensor in meter 114 to receive the audio output from the rear-facing speaker typically does not facilitate good recognition of the signature(s). Conversely, when the rear-facing audio sensor of meter 114 is used in conjunction with the rear-facing speaker, better recognition of the signature included in the audio output from the media presentation device can be achieved. In the examples disclosed herein, audio recordings from one or more audio sensors of meter 114 are used to facilitate the best possible signature recognition. For example, when the media presentation device uses rear speakers, audio recordings from one or more rear audio sensors of meter 114 can be used; furthermore, different configurations of the audio sensors of meter 114 can be used, for example, to take into account different acoustic environments resulting in different levels of signature recognition, to take into account different configurations of audio systems (e.g., soundbar systems, 5.1 surround sound systems, 7.1 surround sound systems, etc.), or to use different configurations based on the selected input of the media presentation device 110 (e.g., surround sound speakers can be used when presenting a movie, while rear speakers can be used when presenting broadcast television).

[0033] In some examples, meter 114 may be physically coupled to media presentation device 110 and may be configured to capture audio emitted externally by media presentation device 110 (e.g., free-field audio), thus eliminating the need for a direct physical connection to the audio output of media presentation device 110. For example, the meter 114 in the illustrated example may employ non-intrusive monitoring that does not involve any physical connection to media presentation device 110 (e.g., via...). connect, Intrusive monitoring (e.g., via USB connection, high-definition media interface (HDMI) connection, Ethernet cable connection, etc.) and / or involving one or more physical connections to the media presentation device 110. In some examples, intrusive monitoring can be used to facilitate the determination of which audio sensors the meter 114 should use. For example, the meter 114 can be connected to the media presentation device using a Universal Serial Bus (USB) cable, allowing the meter 114 to identify the speaker configuration of the media presentation device 110. Based on this information, the meter 114 can select the appropriate audio sensor(s) best suited for monitoring the audio output by the media presentation device 110. For example, if the media presentation device 110 indicates that front speakers are being used, the meter 114 can select one or more front audio sensors(s) to monitor the output audio.

[0034] To generate media playback data, the identifiers of one or more media played to the audience are associated with personnel data (e.g., presence information) collected by meter 114. In the illustrated example, meter 114 collects input (e.g., audience identification data) representing the identity of one or more audience members (e.g., group members 104, 106). In some examples, meter 114 collects audience identification data by periodically or non-periodically prompting audience members in media presentation environment 102 to identify themselves as present in the audience. In some examples, meter 114 responds to predetermined events (e.g., when media presentation device 110 is turned on, a channel changes, an infrared control signal is detected, etc.) by prompting one or more audience members to self-identify. The audience identification data and playback data can then be matched with demographic data collected from audience members (e.g., group members 104, 106) during registration to form a metric reflecting, for example, the demographic composition of the audience. Demographic data includes, for example, group members' age, gender, income level, education level, marital status, geographic location, ethnicity, etc.

[0035] In some examples, meter 114 can be configured to receive group member information via an input device, such as a remote control, Mobile phones, etc. In such examples, meter 114 prompts audience members to indicate their presence by pressing the appropriate input key on the input device. The meter 114 in the illustrated example can also determine the time at which audience members are prompted to input information into meter 114. In some examples, Figure 1 The meter 114 supports audio signatures for personnel monitoring, enabling it to detect the presence of a group member identification metering device in the vicinity of the media presentation device 110 (e.g., within the media presentation environment 102). For example, one or more audio sensors of the meter 114 can sense sample audio output (e.g., emitted) by a sample group member identification metering device (e.g., a wristband, mobile phone, etc.) uniquely associated with a specific group member. The audio output by the sample group member identification metering device may include, for example, one or more audio watermarks to facilitate the identification of the group member identification metering device and / or the group member 104 associated with it.

[0036] The illustrated meter 114 communicates with a central facility 190 that is remotely located for the audience measurement entity. Figure 1 In the example shown, example meter 114 communicates with central facility 190 via gateway 140 and network 180. Figure 1 Example metering device 114 periodically, non-periodically, and / or upon request from central facility 190, sends media identification data and / or audience identification data to central facility 190.

[0037] Figure 1 The example gateway 140 shown is a router that enables meter 114 and / or other devices in the media presentation environment (e.g., media presentation device 110) to communicate with network 180 (e.g., the Internet).

[0038] In some examples, example gateway 140 facilitates the delivery of media from one or more media sources 112 over the Internet to media presentation device 110. In some examples, example gateway 140 includes gateway functionality such as modem capabilities. In some other examples, example gateway 140 is implemented in two or more devices (e.g., routers, modems, switches, firewalls, etc.). The example gateway 140 shown can communicate with network 126 via Ethernet, Digital Subscriber Line (DSL), telephone line, coaxial cable, USB connection, Bluetooth connection, any wireless connection, etc.

[0039] In some examples, example gateway 140 hosts a local area network (LAN) for media presentation environment 102. In the illustrated example, the LAN is a wireless local area network (WLAN) and allows meter 114, media presentation device 110, etc., to send and / or receive data over the Internet. Alternatively, gateway 140 may be coupled to such a LAN. In some examples, example gateway 140 is implemented using a cellular communication system and may, for example, enable meter 114 to send information to central facility 190 using a cellular connection.

[0040] The example network 180 shown is a wide area network (WAN) such as the Internet. However, in some examples, a local network may be used additionally or alternatively. Furthermore, the example network 180 can be implemented using any type of public or private network, such as, but not limited to, the Internet, telephone networks, local area networks (LANs), wired networks and / or wireless networks, or any combination thereof.

[0041] The central facility 190 in the example shown is implemented by one or more servers. The central facility 190 processes and stores data received from one or more meters 114. For example, Figure 1 Example Central Facility 190 combines audience identification data and program identification data from multiple households to generate aggregated media surveillance information. Central Facility 190 generates one or more reports for advertisers, program producers, and / or other stakeholders based on the aggregated statistics. Such reports include inferences about the size and demographic composition of the audience for content, channels, and / or advertising based on demographic information and behavior of members of the monitored group.

[0042] As described above, the meter 114 in the example shown provides a combination of media metering and personnel metering. Figure 1The meter 114 includes its own housing, processor, memory and / or software to perform desired media monitoring and / or personnel monitoring functions. Figure 1 Example meter 114 is a fixed device disposed on or near media presentation device 110. To identify and / or confirm the presence of group members in media presentation environment 102, example meter 114 includes a display. For example, the display provides identification of group members 104, 106 present in media presentation environment 102. For example, in the illustrated example, meter 114 displays markings (e.g., illuminated numerical numbers 1, 2, 3, etc.) that identify and / or confirm the presence of first group member 104, second group member 106, etc. In the illustrated example, meter 114 is fixed to the top of media presentation device 110. However, meter 114 may be fixed to any other location on media presentation device (e.g., on the side of media presentation device 110, on the bottom of media presentation device 110), and / or may not be fixed to media presentation device 110. For example, meter 114 may be placed in a location near media presentation device 110.

[0043] Figure 2 It is shown Figure 1 A block diagram of an example implementation of the example meter 114. Figure 2 Example meter 114 includes example audio sensors 202, 204, 206, 208, example audio sensor selector 210, example configuration memory 220, example media recognizer 230, example audio analyzer 240, example configuration interface 245, example audience measurement data controller 250, example data memory 255, example network communicator 260, example person recognizer 270, example power receiver 280, and example battery 285.

[0044] Figure 2The example audio sensors 202, 204, 206, and 208 shown are implemented using microphones and / or other acoustic sensors. Each of the example audio sensors 202, 204, 206, and 208 receives ambient sound (e.g., free-field audio), including audible media presented in the vicinity of meter 114. Alternatively, one or more of the audio sensors 202, 204, 206, and 208 may be implemented via a line-in connection. A line-in connection may allow one or more external microphones to be used with meter 114, and / or in some examples may allow one or more of the audio sensors 202, 204, 206, and 208 to be directly connected to the output of a media presentation device (e.g., an auxiliary output of a television, an auxiliary output of an audio / video receiver for a home entertainment system, etc.). Advantageously, meter 114 is positioned such that audio sensors 202, 204, 206, 208 receive ambient audio generated by the television and / or other devices of the home entertainment system with sufficient quality to identify media presented by media presentation device 110 and / or other devices of media presentation environment 102 (e.g., audio / video receiver 118). For example, in the examples disclosed herein, meter 120 may be placed on top of the television, fixed to the bottom of the television, etc.

[0045] exist Figure 2 The example shown illustrates four audio sensors 202, 204, 206, and 208. Each of the four audio sensors 202, 204, 206, and 208 corresponds to the front right microphone, front left microphone, rear right microphone, and rear left microphone, respectively. Although in Figure 2 Four audio sensors are used in the example shown, but any number of audio sensors can be used additionally or alternatively. Example locations of example audio sensors 202, 204, 206, and 208 on meter 114 are described below. Figure 4 , Figure 5 and / or Figure 6 The example shown is illustrated.

[0046] Figure 2The example audio sensor selector 210, as shown in the example, combines the audio received by audio sensors 202, 204, 206, and 208 to prepare a combined audio signal for analysis by media recognizer 230. In some examples, the example audio sensor selector 210 combines the audio received by audio sensors 202, 204, 206, and 208 by mixing the audio. In the examples disclosed herein, the example audio sensor selector 210 consults example configuration memory 220 to identify which audio sensors 202, 204, 206, and 208 should allow their respective received audio signals to pass through media recognizer 230. Conversely, in some examples, the example audio sensor selector 210 may identify which audio sensors 202, 204, 206, and 208 should not pass (e.g., should be blocked), and accordingly block those audio sensors 202, 204, 206, and 208. In some examples, the audio sensor selector 210 negotiates with the audio analyzer 240 to identify which audio sensors 202, 204, 206, 208 should pass their respective audio signals through the media recognizer 230.

[0047] Figure 2 The example configuration memory 220 shown stores audio sensor configurations that identify which of the audio sensor selectors 210 should select among the audio sensors 202, 204, 206, and 208 to form the audio signal to be processed by the media recognizer 230. However, any additional configuration and / or operational information may be stored additionally or alternatively. For example, WiFi credentials used by network communicator 260, group members, and / or home recognizers may be stored in the configuration memory 220. The example configuration memory 220 can be updated via, for example, a configuration interface 245 and / or an audio analyzer 240. Figure 2 The example configuration memory 220 shown can be implemented by any device used for storing data, such as flash memory, magnetic media, optical media, etc. Furthermore, the data stored in the example configuration memory 220 can be in any data format, such as binary data, comma-separated data, tab-separated data, Structured Query Language (SQL) structures, etc.

[0048] Figure 2 The example media recognizer 230 shown analyzes audio received through one or more of the audio sensors 202, 204, 206, and 208 and identifies the media being presented. The example media recognizer 230 outputs an identifier for the media (e.g., media identification information) to the audience measurement data controller 250. Figure 2In the example shown, the example media recognizer 230 outputs a quality metric for the media identifier. As used herein, a quality metric is defined as any value representing the strength and / or quality of the detected signature / fingerprint. In the examples disclosed herein, the quality metric can be a score, bit error rate (BER), volume level, etc. Furthermore, in some examples, different values ​​representing the strength and / or quality of the detected signature / fingerprint can be combined to form a quality metric.

[0049] In some examples, media identifier 230 may utilize signature-based media identification techniques. Unlike media monitoring techniques based on codes and / or watermarks contained in and / or embedded in the monitored media, fingerprint- or signature-based media monitoring techniques typically use one or more inherent characteristics of the monitored media during a monitoring time interval to generate a substantially unique proxy for the media. This proxy is referred to as a signature or fingerprint and can take any form (e.g., a series of numerical values, waveforms, etc.) representing any aspect of (one or more) the media signal (e.g., the monitored audio and / or video signals forming the media presentation). A signature can be a series of signatures collected consecutively over a time interval. A good signature is repeatable when processing the same media presentation but unique relative to other (e.g., different) presentations of other (e.g., different) media. Therefore, the terms “fingerprint” and “signature” are used interchangeably herein and are defined herein as meaning a proxy for identifying the media generated by one or more inherent characteristics of the media.

[0050] Signature-based media surveillance typically involves determining (e.g., generating and / or collecting) one or more signatures representing media signals (e.g., audio and / or video signals) output by a monitored media device, and comparing one or more monitored signatures with one or more reference signatures corresponding to a known (e.g., reference) media source. Various comparison criteria, such as cross-correlation values, Hamming distance, etc., can be evaluated to determine whether a monitored signature matches a particular reference signature. When a match is found between a monitored signature and one of the reference signatures, the monitored media can be identified as corresponding to a specific reference media represented by the reference signature that matches the monitored signature. Because attributes such as media identifiers, presentation time, broadcast channel, etc., are collected for the reference signature, these attributes can then be associated with monitored media whose monitored signature matches the reference signature. Example systems for identifying media based on codes and / or signatures are already known and were first disclosed in Thomas's U.S. Patent 5,481,294, the entire contents of which are incorporated herein by reference. In some examples, media recognizer 230 analyzes the peak values ​​of transformed audio signals from one or more of the audio sensors 202, 204, 206, and 208 identified by audio sensor selector 210. For example, audio sensor selector 210 may identify audio recordings from the first audio sensor 202 and the second audio sensor 204 to be analyzed. Therefore, media recognizer 230 can perform fingerprinting on the peak values ​​of the transformed audio recordings to reduce the computational burden on the processor, as discussed in more detail below.

[0051] Turn Figure 3The example audio analyzer 240 includes an example virtual source determiner 300, an example audio retrieval unit 302, an example audio converter 304, an example Time Difference of Arrival (TDOA) determiner 306, and an example TDOA matcher 308. The example virtual source determiner 300 analyzes the configuration of audio sensors 202, 204, 206, and 208 (e.g., the positioning of the audio sensors on or within the meter 114). For example, the virtual source determiner 300 identifies the configuration of audio sensors 202, 204, 206, and 208 and radially positions eight virtual audio sources around the meter 114 at different angles. In some examples, the virtual sources may be positioned at different distances from the meter 114. For example, a first virtual source may be virtually positioned three feet from the meter, while a second virtual source may be virtually positioned five feet from the meter 114. In some examples, virtual sources can be positioned at different angles and locations (e.g., a first virtual source is positioned at a first angle at a first distance from meter 114, a second virtual source is positioned at a second angle at a second distance from meter 114, etc.). Although eight virtual audio sources are described in the examples disclosed herein, any number of virtual sources can be used (e.g., nine, 15, 30, etc.). Example virtual source determiner 300 generates a graph to identify virtual sources, the angle at which virtual sources are radially positioned around meter 114, the TDOA between the first audio sensor 202 and the second audio sensor 204, the TDOA between the third audio sensor 206 and the fourth audio sensor 208, the TDOA between the second audio sensor 204 and the fourth audio sensor 208, and the TDOA between the first audio sensor 202 and the third audio sensor 206. In some examples, the graph generated by virtual source determiner 300 includes the distance between the virtual source and meter 114 when the virtual source is positioned at different distances from meter 114. The graph generated by virtual source determiner 300 will be combined below. Figure 7 and Figure 8 A more detailed description follows. In some examples, the virtual source determiner 300 can determine the TDOA of an audio sensor. For example, the virtual source determiner 300 can determine the TDOA between the first audio sensor 202 and the second audio sensor 204, between the first audio sensor 202 and the third audio sensor 206, and between the first audio sensor 202 and the fourth audio sensor 208.

[0052] Audio retrieval unit 302 retrieves audio records from audio sensors 202, 204, 206, 208 and / or from data storage 255. In some examples, audio retrieval unit 302 may retrieve a first audio record generated by the first audio sensor 202, a second audio record generated by the second audio sensor 204, and a third audio record generated by the third audio sensor 206. Although the illustrated example is described with reference to only four audio records and audio sensors, any number of audio records and / or audio sensors may be used. For example, audio retrieval unit 302 may obtain multiple audio records from the first audio sensor 202, the second audio sensor 204, the third audio sensor 206, and the fourth audio sensor 208.

[0053] Example audio transformer 304 transforms an audio signal into a time-frequency bin and / or frequency components of the audio signal. For example, audio transformer 304 may perform a short-time Fourier transform on the audio signal to transform the audio signal to the frequency domain. Additionally, example audio transformer 304 may divide the transformed audio signal into two or more frequency bins (e.g., using Hamming functions, Hann functions, etc.). Alternatively or additionally, audio transformer 304 may aggregate the audio signal into one or more time segments (e.g., the duration of the audio, six-second segments, one-second segments, etc.). In other examples, audio transformer 304 may use any suitable technique to transform the audio signal (e.g., Fourier transform, discrete Fourier transform, sliding-window Fourier transform, wavelet transform, discrete Hadamard transform, discrete Walsh-Hadamard transform, discrete cosine transform, etc.). In some examples, example audio transformer 304 processes a first audio record using short-time Fourier transform to obtain a first audio transform with a first time-frequency band, processes a second audio record using short-time Fourier transform to obtain a second audio transform with a second time-frequency band, and processes a third audio record using short-time Fourier transform to obtain a third audio transform with a third time-frequency band.

[0054] To calibrate meter 114, example TDOA determiner 306 determines the time difference of arrival between the time it takes for a virtual audio signal to reach the first audio sensor and the time it takes for the same virtual audio signal to reach the second audio sensor when the virtual audio source is located. For example, TDOA determiner 306 calculates a first time the first virtual signal from the virtual source arrives at the first audio sensor 202 based on distance and / or angle and the speed of sound. In some examples, TDOA determiner 306 calculates a second time the first virtual signal from the virtual source arrives at the second audio sensor 204 based on distance and / or angle and the speed of sound. In some examples, TDOA determiner 306 determines the first virtual source arrival time difference based on the difference between the first and second times. TDOA determiner 306 completes this process for the remaining audio sensor pairs and source locations, as described below. Figure 7 and Figure 8 For more details, the results from the TDOA determiner 306 can be stored in the data memory 255, the configuration memory 220, and / or sent to the TDOA matcher 308 for further processing.

[0055] To determine the time difference of arrival (TDOA) of audio recordings from audio sensors 202, 204, 206, and 208, the TDOA determiner 306 determines audio characteristics of a portion of the audio signal (e.g., audio signal frequency components, audio regions surrounding time-frequency segments, etc.). For example, the TDOA determiner 306 can determine the phase value of one or more time-frequency segments of the audio signal frequency components based on the audio recordings generated by audio sensors 202, 204, 206, and 208. In some examples, the TDOA determiner 306 determines a first phase value from a first audio recording from the first audio sensor 202 and identifies the first phase value in a second audio recording from the second audio sensor 204. Furthermore, in this example, the TDOA determiner 306 can determine the time difference of arrival between the first phase value from the first audio recording and the first phase value in the second audio recording to determine the TDOA (e.g., TDOA time difference) between the first audio sensor 202 and the second audio sensor 204. 12In some examples, example TDOA determiner 306 calculates the inter-channel time difference (ITD) of the transformed audio from audio transducer 304. For example, TDOA determiner 306 calculates a first ITDD between the phase value of a first transducer corresponding to the first audio sensor 202 and the phase value of a second transducer corresponding to the second audio sensor 204. In such an example, the first ITDD represents a first arrival time difference. TDOA determiner 306 calculates a second ITDD between the phase value of the first transducer and the phase value of a third transducer corresponding to the third audio sensor 206. In the example shown, the second ITDD represents a second arrival time difference. TDOA determiner 306 completes this process for all audio recordings and audio sensor configurations. TDOA determiner 306 sends all virtual TDOA values ​​and all TDOA values ​​from the audio recordings to TDOA matcher 308.

[0056] Example TDOA matcher 308 matches the inter-channel time difference (e.g., the difference in phase values) between audio recordings and compares them with the virtual source time difference, as combined below. Figure 7 and Figure 8 This is discussed in more detail. For example, the TDOA matcher 308 determines the Euclidean distance between the TDOA of the audio recording and the TDOA of the virtual source location. In some examples, the TDOA matcher 308 determines that the audio is being generated by the presentation device from two virtual sources (two of eight virtual sources in this example). In some examples, one of the sources could be an individual watching a presentation device (e.g., a television) that generates audio from another source. Therefore, the TDOA matcher 308 can identify audio recordings from audio sensors 202, 204 that should be removed from further processing because they are generating background noise that negatively impacts the audio of the media being presented. In some examples, the TDOA matcher 308 clusters time-frequency segments into their own representations, which correspond to virtual sources (e.g., virtual source 6) identified by matching inter-channel time differences with virtual TDOAs to extract estimated spatial sources. In some examples, the estimated spatial sources are used by the media identifier 230 and / or the central facility 190 to calculate a fingerprint with less noise. The TDOA matcher 308 can transmit the results to the media recognizer 230, and the media recognizer 230 further analyzes the first and third audio recordings to determine, for example, the media presented by the media presentation device. The results from the audio analyzer 240 are sent to the media recognizer 230 for further processing.

[0057] Back Figure 2 , Figure 2The example configuration interface 245 shown receives configuration input from the user and / or installer of meter 114. In some examples, configuration interface 245 enables the user and / or installer to instruct that audio sensor configuration is stored in configuration memory 220 and used by audio sensor selector 210. In some examples, configuration interface 245 enables the user and / or installer to control other operating parameters of meter 114 (e.g., WiFi credentials used by network communicator 260), set home and / or (one or more) group member identifiers, etc. Figure 2 In the example shown, configuration interface 245 is implemented using a Bluetooth Low Energy radio. However, configuration interface 245 can be implemented in any other way, such as infrared input, Universal Serial Bus (USB) connection, serial connection, Ethernet connection, etc. In some examples, configuration interface 245 enables meter 114 to be communicatively coupled to a media device, such as media presentation device 110. Such a communicative connection enables configuration interface 245 to, for example, detect the audio configuration of media presentation device 110, allowing configuration memory 220 to be updated to select audio sensors 202, 204, 206, 208 corresponding to the selected audio configuration of media presentation device 110. For example, if the media presentation device is using rear speakers, one or more audio sensors corresponding to the rear microphone(s) can be identified in configuration memory 220.

[0058] Figure 2 The example audience measurement data controller 250 shown receives media identification information (e.g., code, signature, etc.) from media identifier 230 and audience identification data from person identifier 270, and stores the received information in data memory 255. The example audience measurement data controller 250 periodically and / or non-periodically transmits the audience measurement information stored in data memory 255 to central facility 190 via network communicator 260 for aggregation and / or preparation of media monitoring reports.

[0059] Figure 2 The example data storage 255 shown can be implemented by any device used for storing data, such as flash memory, magnetic media, optical media, etc. Furthermore, the data stored in the example data storage 255 can be in any data format, such as binary data, comma-separated data, tab-separated data, Structured Query Language (SQL) structures, etc. In the example shown, the example data storage 255 stores media identification information collected by media identifier 230 and audience identification data collected by personnel identifier 270. In some examples, the example data storage 255 also stores group member demographic information, such that the user identifier of the received audience measurement data can be converted into demographic information before being sent to the central facility 190.

[0060] Figure 2 The example person identifier 270 shown determines audience identification data representing the identity of one or more audience members (e.g., group members) present in the media presentation environment 102. In some examples, the person identifier 270 collects audience identification data by periodically or non-periodically prompting audience members in the media presentation environment 102 to identify itself as present in the audience. Group members can identify themselves by, for example, pressing a button on a remote control, saying their names, etc. In some examples, the person identifier 270 prompts one or more audience members to self-identify in response to one or more predetermined events (e.g., when the media presentation device 110 is turned on, a channel is changed, an infrared control signal is detected, etc.). The person identifier 270 provides the audience identification data to the audience measurement data controller, allowing the audience measurement data to be correlated with media identification data to facilitate identification of which media is presented to which audience member.

[0061] Figure 2 The example network communicator 260 shown transmits audience measurement information (e.g., data stored in data memory 255) provided by audience measurement data controller 250 to the central facility 190 of the audience measurement entity. In the example shown, network communicator 260 is connected to... Figure 1 Example gateway 140 hosts a WiFi antenna for WiFi network communication. However, in some examples, the network communicator may be additionally or alternatively implemented by an Ethernet port that communicates via Ethernet (e.g., a local area network (LAN)). While example meter 114 is implemented via... Figure 1 The example gateway 140 in the illustrated example transmits data to the central facility 190, but data can be sent to the central facility 190 in any other way. For example, the network communicator 260 can be implemented by a cellular radio, and the example gateway 140 can be a cellular base station. In some examples, the example gateway 140 can be omitted, and the example network communicator 260 can send data directly to the central facility 190.

[0062] Figure 2The example power receiver 280 shown is implemented as a Universal Serial Bus (USB) socket and enables the meter 114 to be connected to a power source via a cable (e.g., a USB cable). In the examples disclosed herein, the media presentation device 110 has a USB port, which, for example, provides power to an external device such as the meter 114. In some examples, the media presentation device 110 may provide power to an external device via a different type of port (e.g., a High Definition Media Interface (HDMI) port, an Ethernet port, etc.). The example power receiver 280 may be implemented in any manner to facilitate receiving power from the media presentation device 110 or any other power source (e.g., a wall socket). In some examples, the power receiver 280 may additionally or alternatively facilitate diagnostic communication with the media presentation device 110. For example, a configuration interface 245 may communicate with the media presentation device 110 via the connection provided by the power receiver 280 (e.g., the USB port) to, for example, determine whether the media presentation device 110 is powered on, determine which input is being presented through the media presentation device 110, and determine which speakers are being used by the media presentation device 110. In some examples, the connection is an HDMI connection, and the configuration interface 245 communicates with the media presentation device 110 using the HDMI Consumer Electronics Control (CEC) protocol.

[0063] Figure 2 The example battery 285 shown stores power for the meter 114. When no power is supplied to the meter 114 via the power receiver 280, the example battery 285 enables the meter 114 to operate. Figure 2 In the example shown, the sample battery is implemented using a lithium-ion battery. However, any other type of battery can be used additionally or alternatively. Figure 2 In the example shown, the example battery 285 is rechargeable. Therefore, the example battery 285 can be recharged when the meter 114 receives power through the power receiver 280 (e.g., when the media presentation device 110 is powered on) to facilitate the operation of the meter 114 when it does not receive power through the power receiver 280 (e.g., when the media presentation device 110 is powered off). However, in some examples, the example battery 285 may be non-rechargeable.

[0064] Figure 4 yes Figure 1 and / or Figure 2 Example front view of example meter 114. Figure 4In the example shown, the example meter 114 includes a housing 410. In the examples disclosed herein, the housing 410 will be attached to the media presentation device 110. For example, the housing 410 may be attached to the top of the media presentation device 110, to the bottom of the media presentation device 110, to the side of the media presentation device 110, etc. In some examples, the housing 410 of the meter 114 is not attached to the media presentation device 110. For example, the housing 410 may be placed in any other location within the media presentation environment 102 where audio can be received by the meter 114.

[0065] exist Figure 4 In the example shown, the example housing 410 includes an example left front opening 420 and an example right front opening 425. The front openings 420 and 425 allow free-field audio at the front of the housing to be received by example audio sensors 202, 204, 206, and 208 positioned towards the front of the meter 114. Figure 4 In the example shown, the front openings 420 and 425 are symmetrically arranged on the surface of the meter 114. However, the front openings 420 and 425 can be arranged in any other manner. Furthermore, although in Figure 4 The example shown illustrates two front openings 420, 425, but any other number of openings may be used additionally or alternatively. For example, the front of housing 410 may have three openings, the front of housing 410 may have four openings, and so on.

[0066] Figure 5 yes Figure 1 and / or Figure 2 Example rear view of example meter 114. Figure 5 In the example shown, the example housing 410 includes an example left rear opening 530 and an example right rear opening 535. The rear openings 530 and 535 allow free-field audio in the rear of housing 410 to be received by example audio sensors 202, 204, 206, and 208 positioned towards the rear of meter 114. Figure 5 In the example shown, the rear openings 530 and 535 are symmetrically arranged behind the meter 114. However, the rear openings 530 and 535 can be arranged in any other manner. Furthermore, although in Figure 5 The example shown illustrates two rear openings 530, 535, but any other number of openings may be used additionally or alternatively.

[0067] exist Figure 5 In the example shown, housing 410 includes a USB port 540. Figure 5 In the example shown, USB port 540 enables USB cable 545 to connect... Figure 2The power receiver 280 is connected to an external power source (e.g., power supplied by the media presentation device 110). However, any other type and / or number of ports, cables, power supplies, etc., may be used additionally or alternatively.

[0068] Figure 6 yes Figure 1 and / or Figure 2 Example top view of example meter 114. Figure 6 In the example shown, front openings 420, 425 and rear openings 530, 535 are shown along the contour of housing 410. Inside housing 410, an example circuit board 605 is shown (e.g., carrying...). Figure 2 (Example circuit board of example component of example meter 114). Figure 6 In the example shown, a single circuit board 605 is illustrated. However, the example meter 114 may include multiple circuit boards, allowing components to be positioned to make more efficient use of the available space within the housing 410.

[0069] exist Figure 6 In the example shown, the first audio sensor 202 is connected to the circuit board 605 and positioned near the left front opening 420. The second audio sensor 204 is connected to the circuit board 605 and positioned near the right front opening 425. The third audio sensor 206 is connected to the circuit board 605 and positioned near the left rear opening 530. The fourth audio sensor 208 is connected to the circuit board 605 and positioned near the right rear opening 535. Therefore, free-field audio through the openings 420, 425, 530, and 535 of the housing 410 is picked up by the corresponding audio sensors 202, 204, 206, and 208.

[0070] exist Figure 6 In the example shown, the first audio sensor 202 is connected to the circuit board 605 and positioned adjacent to the left front opening 420. In the example shown, the second audio sensor 204 is connected to the circuit board 605 and positioned at a first distance 610 from the first audio sensor 202. In the example shown, the third audio sensor 206 is connected to the circuit board 605 and positioned at a second distance 615 from the first audio sensor 202. In the example shown, the fourth audio sensor 208 is connected to the circuit board 605 and positioned at a third distance 620 from the first audio sensor 202. However, audio sensors 202, 204, 206, and 208 can be configured to, except for... Figure 6 Any alternative configuration other than the one shown.

[0071] Figure 7An example virtual source location environment 700 is shown that can be used to implement the examples disclosed herein. In the illustrated example, the virtual source location environment 700 includes an example meter 114, which includes audio sensors 202, 204, 206, 208 and virtual source locations 1-8. To determine the virtual source location environment 700, an example virtual source determiner 300 analyzes the configuration of the audio sensors 202, 204, 206, 208 (e.g., the positioning of the audio sensors on or within the meter 114). For example, the virtual source determiner 300 identifies the configuration of the audio sensors 202, 204, 206, 208 and radially positions eight virtual audio sources around the meter 114 at different angles. Although eight virtual audio sources are described in the examples disclosed herein, any number of virtual sources (e.g., nine, 15, 30, etc.) can be used.

[0072] In the illustrated example, the virtual source determiner 300 generates a graph 702, which includes a first column 704 for identifying virtual sources, a second column 706 for identifying the angles at which virtual sources are radially positioned around the meter 114, a third column 708 for identifying the TDOA between the first audio sensor 202 and the second audio sensor 204, a fourth column 710 for identifying the TDOA between the third audio sensor 206 and the fourth audio sensor 208, a fifth column 712 for identifying the TDOA between the second audio sensor 204 and the fourth audio sensor 208, and a sixth column 714 for identifying the TDOA between the first audio sensor 202 and the third audio sensor 206. In the illustrated example, the virtual source determiner 300 fills the first column 704 and the second column 706.

[0073] To fill in the remainder of graph 702 (e.g., columns 708-714), TDOA determiner 306 determines the time difference of arrival between the time it takes for a virtual audio signal to reach the first audio sensor and the time it takes for the same virtual audio signal to reach the second audio sensor when the virtual audio source is located. For example, TDOA determiner 306 calculates the first time the first virtual signal from virtual source 1 arrives at the first audio sensor 202 based on the distance between virtual source 1 and the first audio sensor 202 and / or the angle in the second column 706, as well as the speed of sound. In some examples, TDOA determiner 306 calculates the second time the first virtual signal from virtual source 1 arrives at the second audio sensor 204 based on the distance between virtual source 1 and the second audio sensor 204 and / or the angle in the second column 706, as well as the speed of sound. In some examples, TDOA determiner 306 determines the first virtual source arrival time difference (e.g., TDOA) based on the difference between the first and second times. 34The TDOA determiner 306 then fills row 716 with the corresponding TDOA value. The TDOA determiner 306 completes this process for the remaining audio sensor pairs and source locations to fill the rest of chart 702. The completed chart 702 can be stored in data memory 255, configuration memory 220, and / or sent to TDOA matcher 308 for further processing.

[0074] Figure 8 An example implementation of the example disclosed in this article is shown. Figure 8 The examples shown include example virtual source location environment 700, and Figure 7 Chart 702. However, with Figure 7 In contrast, audio analyzer 240 has already analyzed the audio recordings and determined the locations of potential audio sources, as identified in the first row 802 and the second row 804. In the illustrated example, audio retrieval unit 302 retrieves a first audio recording generated by the first audio sensor 202, a second audio recording generated by the second audio sensor 204, and a third audio recording generated by the third audio sensor 206. Although the illustrated example is described with reference to only three audio recordings and audio sensors, any number of audio recordings and / or audio sensors can be used. For example, audio retrieval unit 302 could obtain multiple audio recordings from the first audio sensor 202, the second audio sensor 204, the third audio sensor 206, and the fourth audio sensor 208.

[0075] In the illustrated example, audio transformer 304 processes a first audio record using a short-time Fourier transform algorithm to obtain a first audio transform with a first time-frequency band. The audio transformer 304 in the illustrated example processes a second audio record using a short-time Fourier transform algorithm to obtain a second audio transform with a second time-frequency band. The audio transformer 304 in the illustrated example processes a third audio record using a short-time Fourier transform algorithm to obtain a third audio transform with a third time-frequency band. Although the examples disclosed herein are described using short-time Fourier transform algorithms, audio transformer 304 can perform any number of transforms to transform an audio record (e.g., an audio signal) to the frequency domain. Subsequently, TDOA determiner 306 calculates a first inter-channel time difference between the phase values ​​of the first transform and the phase values ​​of the second transform. In the illustrated example, the first inter-channel time difference represents a first time difference of arrival (e.g., TDOA). 34 The TDOA determiner 306 calculates the second inter-channel time difference between the phase values ​​of the first transform and the third transform. In the example shown, the second inter-channel time difference represents the second time difference of arrival (e.g., TDOA). 12 The TDOA determiner 306 completes this process by configuring all audio recordings and audio sensors.

[0076] TDOA matcher 308 matches the inter-channel time differences (e.g., phase value differences) between audio recordings and compares them to the virtual source time differences in graph 702. For example, TDOA matcher 308 determines the Euclidean distance between the TDOA of the audio recording and the TDOA of the virtual source location. In the example shown, TDOA matcher 308 determines that the audio is being generated by presentation devices from sources 3 and 6. In some examples, source 3 could be an individual watching a presentation device (e.g., a television) that is generating audio from source 6. Therefore, TDOA matcher 308 can identify audio recordings from audio sensors 202, 204 that should be removed from further processing because they are generating background noise that negatively impacts the audio of the media being presented. In some examples, TDOA matcher 308 clusters the time-frequency segments corresponding to virtual source 6 to extract estimated spatial sources. In some examples, the estimated spatial sources are used by media identifier 230 and / or central facility 190 to calculate a fingerprint with less noise. The TDOA matcher 308 can transmit the results to the media recognizer 230, and the media recognizer 230 can further analyze the first and third audio records to determine, for example, the media presented by the media presentation device.

[0077] Although Figure 2 The implementation is shown in the figure. Figure 1 Example meter 114 example method, but Figure 2 One or more of the elements, processes, and / or devices shown may be combined, divided, rearranged, omitted, eliminated, and / or implemented in any other way. Furthermore, Figure 1 and / or Figure 2 Example audio sensors 202, 204, 206, 208, example audio sensor selector 210, example configuration memory 220, example media recognizer 230, example audio analyzer 240, example configuration interface 245, example audience measurement data controller 250, example data memory 255, example network communicator 260, example person recognizer 270, example power receiver 280, example battery 285, and / or more generally, example meter 114, can be implemented by hardware, software, firmware, and / or any combination of hardware, software, and / or firmware. Thus, for example, Figure 1 and / or Figure 2Example audio sensors 202, 204, 206, 208, example audio sensor selector 210, example configuration memory 220, example media recognizer 230, example audio analyzer 240, example configuration interface 245, example audience measurement data controller 250, example data memory 255, example network communicator 260, example person recognizer 270, example power receiver 280, example battery 285, and / or more generally, any of the example meters 114 may be implemented by one or more analog or digital circuits, logic circuits, one or more programmable processors, one or more programmable controllers, one or more graphics processing units (GPUs), one or more digital signal processors (DSPs), one or more application-specific integrated circuits (ASICs), one or more programmable logic devices (PLDs), and / or one or more field-programmable logic devices (FPLDs). When reading any of the device or system claims of this patent to cover purely software and / or firmware implementations, at least one of the following: example audio sensors 202, 204, 206, 208; example audio sensor selector 210; example configuration memory 220; example media recognizer 230; example audio analyzer 240; example configuration interface 245; example audience measurement data controller 250; example data memory 255; example network communicator 260; example person recognizer 270; example power receiver 280; and example battery 285, is hereby expressly defined as including non-transitory computer-readable storage devices or storage disks, such as memory including software and / or firmware, digital multifunction discs (DVDs), optical discs (CDs), Blu-ray discs, etc. Furthermore, Figure 2 Example meter 114, except Figure 2 Other than or in place of those shown Figure 2 Those shown may include one or more elements, processes, and / or devices, and / or may include more than one of any or all of the elements, processes, and devices shown. As used herein, the phrase “in communication” (including variations thereof) includes direct communication and / or indirect communication via one or more intermediate components, and does not require direct physical (e.g., wired) communication and / or continuous communication, but additionally includes selective communication at periodic intervals, pre-arranged intervals, non-periodic intervals, and / or one-off events.

[0078] Used to implement Figure 2 The flow chart of meter 114, representing example hardware logic, machine-readable instructions, hardware-implemented state machines and / or any combination thereof, is in Figure 9 and / or Figure 10As shown below. Machine-readable instructions can be one or more executable programs or portions thereof for execution by a computer processor, such as those described below in conjunction with... Figure 11 The processor 1112 shown in the example processor platform 1100 discussed here. Programs may be embodied in software stored on non-transitory computer-readable storage media (e.g., CD-ROM, floppy disk, hard disk drive, DVD, Blu-ray disc, or memory associated with the processor 1112), but the entire program and / or portions thereof may alternatively be executed by a device other than the processor 1112 and / or embodied in firmware or dedicated hardware. Furthermore, although the example program is for reference only... Figure 9 and Figure 10 The flowchart shown is used to describe the implementation of example meter 114, but many other methods can be used alternatively. For example, the execution order of the boxes can be changed, and / or some of the boxes described can be changed, eliminated, or combined. Additionally or alternatively, any or all boxes can be implemented by one or more hardware circuits (e.g., discrete and / or integrated analog and / or digital circuits, FPGAs, ASICs, comparators, operational amplifiers, logic circuits, etc.) configured to perform the corresponding operations without running software or firmware.

[0079] The machine-readable instructions described herein can be stored in one or more of the following formats: compressed format, encrypted format, distributed format, compiled format, executable format, packaged format, etc. As described herein, machine-readable instructions can be stored as data (e.g., partial instructions, code, code representations, etc.) that can be used to create, manufacture, and / or produce machine-executable instructions. For example, machine-readable instructions can be distributed and stored on one or more storage devices and / or computing devices (e.g., servers). Machine-readable instructions may require one or more of the following to be installed, modified, adapted, updated, combined, supplemented, configured, decrypted, decompressed, unpacked, distributed, redistributed, compiled, etc., to make them directly readable, interpretable, and / or executable by computing devices and / or other machines. For example, machine-readable instructions can be stored in multiple parts that are individually compressed, encrypted, and stored on separate computing devices, wherein these parts, when decrypted, decompressed, and combined, form a set of executable instructions that implement, for example, the program described herein.

[0080] In another example, machine-readable instructions may be stored in a computer-readable state, but require the addition of libraries (e.g., dynamic link libraries (DLLs), software development kits (SDKs), application programming interfaces (APIs), etc.) to execute the instructions on a specific computing device or other device. In yet another example, the machine-readable instructions may need to be configured (e.g., stored settings, data input, recorded network addresses, etc.) before they can be fully or partially executed. Therefore, the disclosed machine-readable instructions and / or / one or more corresponding programs are intended to cover such machine-readable instructions and / or / one or more programs, regardless of their specific format or state at storage or at rest or in transit.

[0081] The machine-readable instructions described in this article can be represented using any past, present, or future instruction language, scripting language, programming language, etc. For example, machine-readable instructions can be represented using any of the following languages: C, C++, Java, C#, Perl, Python, JavaScript, Hypertext Markup Language (HTML), Structured Query Language (SQL), Swift, etc.

[0082] As mentioned above, Figures 9-10 The example process can be implemented using executable instructions (e.g., computer and / or machine-readable instructions) stored on a non-transitory computer and / or machine-readable medium, such as a hard disk drive, flash memory, read-only memory, optical disk, digital multifunction disk, cache, random access memory, and / or any other storage device or disk in which information can be stored for any duration (e.g., long-term, permanent, transient, temporary buffered, and / or cached information). As used herein, the term non-transitory computer-readable medium is explicitly defined to include any type of computer-readable storage device and / or disk, excluding propagation signals and transmission media.

[0083] "Comprising" and "including" (and all their forms and tenses) are used herein as open-ended terms. Therefore, whenever a claim uses any form of "comprising" or "including" (e.g., including (comprises, include, comprising, including), having, etc.) as a preamble, or in any kind of claim statement, it should be understood that additional elements, terms, etc., may be present without exceeding the scope of the corresponding claim or statement. As used herein, when the phrase "at least" is used as a transitional term, for example, in the preamble of a claim, it is open-ended in the same way as the terms "comprising" and "including" are open-ended. For example, when the term "and / or" is used in the form of A, B, and / or C, it refers to any combination or subset of A, B, C, such as (1) A alone, (2) B alone, (3) C alone, (4) A and B, (5) A and C, (6) B and C, and (7) A and B and C. As used herein in the context of describing structures, components, items, objects, and / or things, the phrase "at least one of A and B" means an implementation that includes (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects, and / or things, the phrase "at least one of A or B" means an implementation that includes (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. As used herein in the context of describing the execution or operation of processes, instructions, actions, activities, and / or steps, the phrase "at least one of A and B" means an implementation that includes (1) at least one A, (2) at least one B, and (3) at least one A and at least one B. Similarly, as used herein in the context of describing the execution or operation of processes, instructions, actions, activities and / or steps, the phrase “at least one of A or B” means an implementation including (1) at least one A, (2) at least one B and (3) any one of at least one A and at least one B.

[0084] As used herein, singular references (e.g., "a ("a" or "an")", "first", "second", etc.) do not exclude plurals. As used herein, the term "a" refers to one or more of those entities. The terms "a ("a" or "an")", "one or more", and "at least one" are used interchangeably herein. Furthermore, although listed separately, multiple means, elements, or method actions can be implemented, for example, by a single unit or processor. Additionally, although individual features may be included in different examples or claims, these individual features may be combined, and inclusion in different examples or claims does not imply that the combination of features is infeasible and / or unadvantageous.

[0085] Figure 9 This indicates that it can be executed to achieve [the desired result]. Figure 1 , Figure 2 and / or Figure 3 The flowchart of example machine-readable instruction 900 for measuring the virtual time difference of arrival of an audio sensor pair. Figure 9 The example program 900 shown begins with the example virtual source determiner 300 determining the first virtual source position of the first virtual audio source and the second virtual source position of the second virtual audio source (box 902). For example, the virtual audio source determiner 300 determines the first virtual source position and the second virtual source position by radially positioning the first virtual source and the second virtual source at different angles around the meter 114.

[0086] TDOA determiner 306 determines a first virtual time difference of arrival (TDOA) for the first pair of audio sensors and a second virtual time difference of arrival (TDOA) for the second pair of audio sensors (box 904). For example, TDOA determiner 306 is based on the distance between virtual source 1 and first audio sensor 202 and / or Figure 7 The angle and speed of sound in the second column 706 are used to calculate the first time when the first virtual signal from virtual source 1 arrives at the first audio sensor 202, based on the distance and / or distance between virtual source 1 and the second audio sensor 204. Figure 7 The angle and speed of sound in the second column 706 are used to calculate the second time when the first virtual signal from virtual source 1 arrives at the second audio sensor 204, and the arrival time difference of the first virtual source is determined based on the difference between the first time and the second time (e.g., TDOA). 34 Then fill line 716 with the corresponding TDOA value. Example program ends at 900.

[0087] Figure 10 This indicates that it can be executed to achieve [the desired result]. Figure 1 , Figure 2 and / or Figure 3 The flowchart of example machine-readable instructions 1000 for improving audio signature detection by removing unnecessary audio recordings. Figure 10The example program 1000 shown begins with example audio retrieval unit 302 retrieving a first audio record from a first audio sensor, a second audio record from a second audio sensor, and a third audio record from a third audio sensor (box 1002). For example, audio retrieval unit 302 retrieves a first audio record generated by first audio sensor 202, a second audio record generated by second audio sensor 204, and a third audio record generated by third audio sensor 206. In some examples, audio transducer 304 processes 1) the first audio record using short-time Fourier transform to obtain a first audio transducer with a first time-frequency band, processes 2) the second audio record using short-time Fourier transform to obtain a second audio transducer with a second time-frequency band, and processes 3) the third audio record using short-time Fourier transform to obtain a third audio transducer with a third time-frequency band.

[0088] TDOA determiner 306 determines a first time-of-arrival difference (TOA) between the first and second audio sensors, and a second TOA between the first and third audio sensors (box 1004). For example, TDOA determiner 306 calculates a first inter-channel time difference between the phase values ​​of the first and second transforms. In the illustrated example, the first inter-channel time difference represents the first TOA. TDOA determiner 306 calculates a second inter-channel time difference between the phase values ​​of the first and third transforms. In the illustrated example, the second inter-channel time difference represents the second TOA.

[0089] Next, the TDOA matcher 308 determines whether a match has been identified (box 1006). For example, the TDOA matcher 308 compares the virtual time difference of arrival with the inter-channel time difference (e.g., the value of the time difference of arrival of the audio recordings) to determine the shortest Euclidean distance between these values. If the TDOA matcher 308 does not identify a match, procedure 1000 returns to box 1002. If the TDOA matcher 308 identifies a match, the TDOA matcher 308 identifies the first virtual source location as the location of the media presentation device presenting the media (box 1008). For example, the TDOA matcher 308 may identify the first virtual source 1 of FIG702 as matching the inter-channel time difference of the audio recordings from the first frequency sensor 202 and the third audio sensor 206.

[0090] TDOA matcher 308 removes the second audio record (block 1010). For example, TDOA matcher 308 removes the second audio record to reduce the computational burden on the processor.

[0091] Figure 11 It is constructed to execute Figures 9-10 Instructions to achieve Figure 1 and / or Figure 2A block diagram of an example processor platform 1100 for the meter 114. The processor platform 1100 can be, for example, a server, personal computer, workstation, self-learning machine (e.g., neural network), mobile device (e.g., mobile phone, smartphone, tablet computer (e.g., iPad)). TM Personal digital assistants (PDAs), internet devices, DVD players, CD players, digital video recorders, Blu-ray players, game consoles, personal video recorders, set-top boxes, headsets or other wearable devices, or any other type of computing device.

[0092] The processor platform 1100 shown in the example includes a processor 1112. The processor 1112 shown in the example is hardware. For example, the processor 1112 can be implemented by one or more integrated circuits, logic circuits, microprocessors, GPUs, DSPs, or controllers from any desired family or manufacturer. The hardware processor can be a semiconductor-based (e.g., silicon-based) device. In this example, the processor implements example audio sensors 202, 204, 206, 208, example audio sensor selector 210, example configuration memory 220, example media recognizer 230, example audio analyzer 240, example configuration interface 245, example audience measurement data controller 250, example data memory 255, example network communicator 260, example people recognizer 270, example power receiver 280, example battery 285, and / or more generally, example meter 114.

[0093] The processor 1112 of the illustrated example includes local memory 1113 (e.g., cache). The processor 1112 of the illustrated example communicates via bus 1118 with main memory, which includes volatile memory 1114 and non-volatile memory 1116. The volatile memory 1114 may be synchronous dynamic random access memory (SDRAM), dynamic random access memory (DRAM), etc. Dynamic Random Access Memory And / or any other type of random access storage device. The non-volatile memory 1116 can be implemented by flash memory and / or any other desired type of storage device. Access to the main memory 1114, 1116 is controlled by the memory controller.

[0094] The processor platform 1100 shown in the example also includes interface circuitry 1120. Interface circuitry 1120 can be implemented using any type of interface standard, such as an Ethernet interface, Universal Serial Bus (USB), etc. Interfaces include Near Field Communication (NFC) interfaces and / or PCI Express (serial bus) interfaces.

[0095] In the example shown, one or more input devices 1122 are connected to interface circuitry 1120. The input devices 1122 allow the user to input data and / or commands into processor 1012. These input devices can be implemented, for example, audio sensors, microphones, cameras (still or video), keyboards, buttons, mice, touchscreens, trackpads, trackballs, isotopes, and / or voice recognition systems. Figure 11 In the example shown, one or more example input devices 1122 implement example audio sensors 202, 204, 206, and 208.

[0096] One or more output devices 1124 are also connected to the interface circuitry 1120 of the illustrated example. The output devices 1124 may be implemented, for example, by display devices (e.g., light-emitting diodes (LEDs), organic light-emitting diodes (OLEDs), liquid crystal displays (LCDs), cathode ray tube displays (CRTs), in-situ IPS displays, touchscreens, etc.), haptic output devices, printers, and / or speakers. Therefore, the interface circuitry 1120 of the illustrated example typically includes a graphics driver card, a graphics driver chip, and / or a graphics driver processor.

[0097] The interface circuit 1120 of the example shown also includes communication devices, such as a transmitter, receiver, transceiver, modem, residential gateway, wireless access point, and / or network interface, to facilitate data exchange with external machines (e.g., any kind of computing device) via network 1126. Communication can be conducted via, for example, Ethernet connections, digital subscriber line (DSL) connections, telephone line connections, coaxial cable systems, satellite systems, field wireless systems, cellular telephone systems, etc.

[0098] The processor platform 1100 shown in the example also includes one or more mass storage devices 1128 for storing software and / or data. Examples of such mass storage devices 1128 include floppy disk drives, hard disk drives, optical disk drives, Blu-ray disc drives, redundant array of independent disks (RAID) systems, and digital multifunction disc (DVD) drives.

[0099] Figure 9 and Figure 10 The machine-executable instructions 1132 may be stored in a mass storage device 1128, volatile memory 1114, non-volatile memory 1116 and / or a removable non-transitory computer-readable storage medium (such as a CD or DVD).

[0100] As will be understood from the foregoing, example methods, apparatuses, and articles of art for improving audio signature detection have been disclosed. The examples disclosed herein overcome the aforementioned problems by reducing the computational burden on the processor by removing audio signals (e.g., audio recordings) from fingerprint processing based on the phase difference between transformed audio signals. The examples disclosed herein remove audio signals based on the phase difference between transformed audio signals, thereby increasing the accuracy of identifying media associated with fingerprints. Furthermore, the examples disclosed herein utilize transformed audio signals to generate fingerprints with less noise, thereby improving audio signature detection. Similarly, the examples disclosed herein utilize the peak values ​​of different portions of the transformed audio signal, thereby reducing the amount of audio to be processed during fingerprint recognition computation (e.g., the processor does not need to process the entire audio signal). The disclosed methods, apparatuses, and articles of art improve the efficiency of using computing devices by removing audio signals (e.g., audio recordings) from fingerprint processing based on the phase difference between transformed audio signals. Therefore, the disclosed methods, apparatuses, and articles of art represent one or more improvements to computer functionality.

[0101] The following paragraphs provide various examples of the examples disclosed in this article.

[0102] Example 1 includes an apparatus comprising: a TDOA determiner configured to determine a first arrival time difference between the first audio sensor and the second audio sensor based on a first audio record from a first audio sensor of a meter and a second audio record from a second audio sensor of the meter, and to determine a second arrival time difference between the first audio sensor and the third audio sensor based on the first audio record and a third audio record from a third audio sensor of the meter; and a TDOA matcher configured to determine a match by comparing the first arrival time difference with i) a first virtual source arrival time difference and ii) a second virtual source arrival time difference, in response to determining that the first arrival time difference matches the first virtual source arrival time difference, identifying the location of the first virtual source as the location of a media presentation device presenting the media, and removing the second audio record to reduce the computational burden on the processor.

[0103] Example 2 includes the apparatus described in Example 1, wherein the meter includes a first audio sensor located at a first position on the meter, a second audio sensor located at a second position on the meter, a third audio sensor located at a third position on the meter, and a fourth audio sensor located at a fourth position on the meter.

[0104] Example 3 includes the apparatus described in Example 2, wherein the first virtual source position and the second virtual source position are determined based on the first position, the second position, the third position, and the fourth position.

[0105] Example 4 includes the apparatus described in Example 1, and further includes a virtual source determiner configured to: determine a first virtual source position of a first virtual audio source radially positioned around the meter and a second virtual source position of a second virtual audio source radially positioned around the meter, the first virtual source position being different from the second virtual source position; and determine the first virtual source arrival time difference of a first pair of audio sensors of the meter based on the first virtual source position, and determine the second virtual source arrival time difference of a second pair of audio sensors of the meter based on the second virtual source position.

[0106] Example 5 includes the apparatus described in Example 4, wherein the first pair of audio sensors includes the first audio sensor and the second audio sensor, and the second pair of audio sensors includes the first audio sensor and the third audio sensor.

[0107] Example 6 includes the apparatus described in Example 5, wherein the virtual source determiner is configured to: calculate a first time when a first virtual signal arrives at the first audio sensor based on the distance between the first virtual source and the first audio sensor and the speed of sound; calculate a second time when the first virtual signal arrives at the second audio sensor based on the distance between the first virtual source and the second audio sensor and the speed of sound; and determine the arrival time difference of the first virtual source based on the difference between the first time and the second time.

[0108] Example 7 includes the apparatus described in Example 5, wherein the virtual source determiner is configured to: calculate a third time when a second virtual signal arrives at the first audio sensor based on the distance between the second virtual source and the first audio sensor and the speed of sound; calculate a fourth time when the second virtual signal arrives at the third audio sensor based on the distance between the second virtual source and the third audio sensor and the speed of sound; and determine the arrival time difference of the second virtual source based on the difference between the third time and the fourth time.

[0109] Example 8 includes the apparatus described in Example 1, and further includes an audio converter configured to: process the first audio record using a short-time Fourier transform to obtain a first audio transform having a first time-frequency band, process the second audio record using the short-time Fourier transform to obtain a second audio transform having a second time-frequency band, and process the third audio record using the short-time Fourier transform to obtain a third audio transform having a third time-frequency band.

[0110] Example 9 includes the apparatus of Example 8, wherein the TDOA determiner is configured to: calculate a first inter-channel time difference between the first audio transform and the second audio transform, the first inter-channel time difference representing the first arrival time difference; and calculate a second inter-channel time difference between the first audio transform and the third audio transform, the second inter-channel time difference representing the second arrival time difference.

[0111] Example 10 includes a method comprising: determining a first arrival time difference between the first audio sensor and the second audio sensor based on a first audio recording from a first audio sensor of a meter and a second audio recording from a second audio sensor of the meter, and determining a second arrival time difference between the first audio sensor and the third audio sensor based on the first audio recording and a third audio recording from a third audio sensor of the meter; determining a matching by comparing the first arrival time difference with i) a first virtual source arrival time difference and ii) a second virtual source arrival time difference, in response to determining that the first arrival time difference matches the first virtual source arrival time difference, identifying the first virtual source location as the location of a media presentation device presenting the media by executing instructions using the processor, and removing the second audio recording by executing instructions using the processor to reduce the computational burden on the processor.

[0112] Example 11 includes the method of Example 10, wherein the meter includes a first audio sensor located at a first position on the meter, a second audio sensor located at a second position on the meter, a third audio sensor located at a third position on the meter, and a fourth audio sensor located at a fourth position on the meter, wherein the first virtual source position and the second virtual source position are determined based on the first position, the second position, the third position, and the fourth position.

[0113] Example 12 includes the method of Example 10, further comprising: determining a first virtual source position of a first virtual audio source radially positioned around the meter and a second virtual source position of a second virtual audio source radially positioned around the meter, the first virtual source position being different from the second virtual source position; and determining the first virtual source arrival time difference of a first pair of audio sensors of the meter based on the first virtual source position, and determining the second virtual source arrival time difference of a second pair of audio sensors of the meter based on the second virtual source position.

[0114] Example 13 includes the method of Example 10, wherein determining the arrival time difference of the first virtual source based on the location of the first virtual source includes: calculating a first time when the first virtual signal arrives at the first audio sensor based on the distance between the first virtual source and the first audio sensor and the speed of sound; calculating a second time when the first virtual signal arrives at the second audio sensor based on the distance between the first virtual source and the second audio sensor and the speed of sound; and determining the arrival time difference of the first virtual source based on the difference between the first time and the second time.

[0115] Example 14 includes the method described in Example 10, wherein determining the arrival time difference of the second virtual source based on the location of the second virtual source includes: calculating a third time for the second virtual signal to arrive at the first audio sensor based on the distance between the second virtual source and the first audio sensor and the speed of sound; calculating a fourth time for the second virtual signal to arrive at the third audio sensor based on the distance between the second virtual source and the third audio sensor and the speed of sound; and determining the arrival time difference of the second virtual source based on the difference between the third time and the fourth time.

[0116] Example 15 includes the method of Example 10, further comprising: processing the first audio record using a short-time Fourier transform to obtain a first audio transform having a first time-frequency band, processing the second audio record using the short-time Fourier transform to obtain a second audio transform having a second time-frequency band, and processing the third audio record using the short-time Fourier transform to obtain a third audio transform having a third time-frequency band.

[0117] Example 16 includes the method described in Example 15, further comprising: calculating a first inter-channel time difference between the first audio transform and the second audio transform, the first inter-channel time difference representing the first arrival time difference; and calculating a second inter-channel time difference between the first audio transform and the third audio transform, the second inter-channel time difference representing the second arrival time difference.

[0118] Example 17 includes a non-transitory computer-readable medium comprising instructions that, when executed, cause a machine to at least: determine a first arrival time difference between the first audio sensor and the second audio sensor based on a first audio recording from a first audio sensor of a meter and a second audio recording from a second audio sensor of the meter; and determine a second arrival time difference between the first audio sensor and the third audio sensor based on the first audio recording and a third audio recording from a third audio sensor of the meter; determine a match by comparing the first arrival time difference with i) a first virtual source arrival time difference and ii) a second virtual source arrival time difference; in response to determining that the first arrival time difference matches the first virtual source arrival time difference, identify the location of the first virtual source as the location of a media presentation device presenting the media; and remove the second audio recording to reduce the computational burden on the processor.

[0119] Example 18 includes the non-transitory computer-readable medium of Example 17, wherein the instructions further cause the machine to perform: determining a first virtual source position of a first virtual audio source radially positioned around the meter and a second virtual source position of a second virtual audio source radially positioned around the meter, the first virtual source position being different from the second virtual source position; and determining, based on the first virtual source position, the first virtual source arrival time difference of a first pair of audio sensors of the meter, and determining, based on the second virtual source position, the second virtual source arrival time difference of a second pair of audio sensors of the meter, the first pair of audio sensors including the first audio sensor and the second audio sensor, and the second pair of audio sensors including the first audio sensor and the third audio sensor.

[0120] Example 19 includes the non-transitory computer-readable medium of Example 18, wherein the instructions further cause the machine to perform: calculating a first time when a first virtual signal arrives at the first audio sensor based on the distance between the first virtual source and the first audio sensor and the speed of sound; calculating a second time when the first virtual signal arrives at the second audio sensor based on the distance between the first virtual source and the second audio sensor and the speed of sound; determining a time difference of arrival of the first virtual source based on the difference between the first time and the second time; calculating a third time when a second virtual signal arrives at the first audio sensor based on the distance between the second virtual source and the first audio sensor and the speed of sound; calculating a fourth time when the second virtual signal arrives at the third audio sensor based on the distance between the second virtual source and the third audio sensor and the speed of sound; and determining a time difference of arrival of the second virtual source based on the difference between the third time and the fourth time.

[0121] Example 20 includes the non-transitory computer-readable medium of Example 17, wherein the instructions further cause the machine to perform: processing the first audio record using a short-time Fourier transform to obtain a first audio transform having a first time-frequency band; processing the second audio record using the short-time Fourier transform to obtain a second audio transform having a second time-frequency band; processing the third audio record using the short-time Fourier transform to obtain a third audio transform having a third time-frequency band; calculating a first inter-channel time difference between the first transform and the second transform, the first inter-channel time difference representing the first arrival time difference; and calculating a second inter-channel time difference between the first transform and the third transform, the second inter-channel time difference representing the second arrival time difference.

[0122] Although certain example methods, apparatuses, and articles of manufacture are disclosed herein, the scope of this patent is not limited thereto. Rather, this patent covers all methods, apparatuses, and articles of manufacture that fall fully within the scope of the claims of this patent.

[0123] The appended claims are incorporated herein by reference, wherein each claim is considered as a separate embodiment of the invention.

Claims

1. An apparatus for improving the detection of audio signatures, comprising: A TDOA determiner is configured to determine a first time difference of arrival between the first audio sensor and the second audio sensor based on a first audio record from a first audio sensor of the meter and a second audio record from a second audio sensor of the meter, and to determine a second time difference of arrival between the first audio sensor and the third audio sensor based on the first audio record and a third audio record from a third audio sensor of the meter; as well as TDOA matcher, the TDOA matcher is used for: Matching is determined by comparing the first arrival time difference with i) the arrival time difference of the first virtual source and ii) the arrival time difference of the second virtual source; In response to determining that the first arrival time difference matches the first virtual source arrival time difference, the location of the first virtual source is identified as the location of the media presentation device that presents the media. as well as In response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce the computational burden on the processor. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

2. The apparatus of claim 1, wherein, The meter includes a first audio sensor located at a first position on the meter, a second audio sensor located at a second position on the meter, a third audio sensor located at a third position on the meter, and a fourth audio sensor located at a fourth position on the meter.

3. The apparatus of claim 2, wherein, The first virtual source position and the second virtual source position are determined based on the first position, the second position, the third position, and the fourth position.

4. The apparatus of claim 1, further comprising a virtual source determiner, the virtual source determiner being configured to: Determine the first virtual source position of a first virtual audio source radially positioned around the meter, and the second virtual source position of a second virtual audio source radially positioned around the meter, wherein the first virtual source position is different from the second virtual source position; and The arrival time difference of the first virtual source of the first pair of audio sensors of the meter is determined based on the location of the first virtual source, and the arrival time difference of the second virtual source of the second pair of audio sensors of the meter is determined based on the location of the second virtual source.

5. The apparatus of claim 4, wherein, The first pair of audio sensors includes the first audio sensor and the second audio sensor, and the second pair of audio sensors includes the first audio sensor and the third audio sensor.

6. The apparatus of claim 5, wherein, The virtual source determiner is used for: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; as well as The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time.

7. The apparatus of claim 5, wherein, The virtual source determiner is used for: The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the second virtual source and the first audio sensor. The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; as well as The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

8. The apparatus of claim 1, further comprising an audio converter, the audio converter being used for: The first audio record is processed using a short-time Fourier transform to obtain a first audio transform with a first time-frequency band; The second audio record is processed using the short-time Fourier transform to obtain a second audio transform with a second time-frequency band; and The third audio recording is processed using the short-time Fourier transform to obtain a third audio transform with a third time-frequency band.

9. The apparatus of claim 8, wherein, The TDOA determiner is used for: Calculate the first inter-channel time difference between the first audio transform and the second audio transform, where the first inter-channel time difference represents the first arrival time difference; and Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

10. The apparatus of claim 1, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

11. A method for improving the detection of audio signatures, comprising: By using processor execution instructions, a first arrival time difference between the first audio sensor and the second audio sensor is determined based on a first audio record from a first audio sensor of the meter and a second audio record from a second audio sensor of the meter, and a second arrival time difference between the first audio sensor and the third audio sensor is determined based on the first audio record and a third audio record from a third audio sensor of the meter; The matching is determined by comparing the first arrival time difference with i) the first virtual source arrival time difference and ii) the second virtual source arrival time difference using instructions executed by the processor; In response to determining that the first arrival time difference matches the first virtual source arrival time difference, the processor executes instructions to identify the location of the first virtual source as the location of the media presentation device that presents the media. as well as By using the processor to execute instructions, in response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce the computational burden on the processor. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

12. The method of claim 11, wherein, The meter includes a first audio sensor located at a first position on the meter, a second audio sensor located at a second position on the meter, a third audio sensor located at a third position on the meter, and a fourth audio sensor located at a fourth position on the meter, wherein the first virtual source position and the second virtual source position are determined based on the first position, the second position, the third position, and the fourth position.

13. The method of claim 11, further comprising: Determine the first virtual source position of a first virtual audio source radially positioned around the meter, and the second virtual source position of a second virtual audio source radially positioned around the meter, wherein the first virtual source position is different from the second virtual source position; as well as The arrival time difference of the first virtual source of the first pair of audio sensors of the meter is determined based on the location of the first virtual source, and the arrival time difference of the second virtual source of the second pair of audio sensors of the meter is determined based on the location of the second virtual source.

14. The method of claim 11, wherein, Determining the arrival time difference of the first virtual source based on its location includes: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; and The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time.

15. The method of claim 11, wherein, Determining the arrival time difference of the second virtual source based on its location includes: The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the second virtual source and the first audio sensor. The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; and The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

16. The method of claim 11, further comprising: The first audio record is processed using a short-time Fourier transform to obtain a first audio transform with a first time-frequency band; The second audio recording is processed using the short-time Fourier transform to obtain a second audio transform with a second time-frequency band; as well as The third audio recording is processed using the short-time Fourier transform to obtain a third audio transform with a third time-frequency band.

17. The method of claim 16, further comprising: Calculate the first inter-channel time difference between the first audio transformation and the second audio transformation, where the first inter-channel time difference represents the first arrival time difference; as well as Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

18. The method of claim 11, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

19. A non-transitory computer-readable medium comprising instructions that, when executed, cause a machine to perform at least the following: A first arrival time difference between the first audio sensor and the second audio sensor is determined based on a first audio record from a first audio sensor of the meter and a second audio record from a second audio sensor of the meter, and a second arrival time difference between the first audio sensor and the third audio sensor is determined based on the first audio record and a third audio record from a third audio sensor of the meter; Matching is determined by comparing the first arrival time difference with i) the arrival time difference of the first virtual source and ii) the arrival time difference of the second virtual source; In response to determining that the first arrival time difference matches the first virtual source arrival time difference, the location of the first virtual source is identified as the location of the media presentation device that presents the media. as well as In response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce the computational burden on the processor. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

20. The non-transitory computer-readable medium of claim 19, wherein, The instructions also cause the machine to perform: Determine the first virtual source position of a first virtual audio source radially positioned around the meter and the second virtual source position of a second virtual audio source radially positioned around the meter, wherein the first virtual source position is different from the second virtual source position; as well as The first virtual source arrival time difference of the first pair of audio sensors of the meter is determined based on the first virtual source location, and the second virtual source arrival time difference of the second pair of audio sensors of the meter is determined based on the second virtual source location. The first pair of audio sensors includes the first audio sensor and the second audio sensor, and the second pair of audio sensors includes the first audio sensor and the third audio sensor.

21. The non-transitory computer-readable medium of claim 20, wherein, The instructions also cause the machine to perform: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time; The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance between the second virtual source and the first audio sensor and the speed of sound; The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; as well as The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

22. The non-transitory computer-readable medium of claim 19, wherein, The instructions also cause the machine to perform: The first audio record is processed using a short-time Fourier transform to obtain a first audio transform with a first time-frequency band; The second audio recording is processed using the short-time Fourier transform to obtain a second audio transform with a second time-frequency band; The third audio record is processed using the short-time Fourier transform to obtain a third audio transform with a third time-frequency band; Calculate the first inter-channel time difference between the first audio transform and the second audio transform, where the first inter-channel time difference represents the first arrival time difference; and Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

23. The non-transitory computer-readable medium of claim 19, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

24. An apparatus for improving the detection of audio signatures, comprising: At least one memory, wherein the at least one memory stores instructions; and Processor circuitry, the processor circuitry being configured to execute the instructions to: The first arrival time difference between the first audio sensor and the second audio sensor is determined based on the first audio record from the first audio sensor of the meter and the second audio record from the second audio sensor of the meter. The second arrival time difference between the first audio sensor and the third audio sensor is determined based on the first audio recording and the third audio recording from the third audio sensor of the meter; as well as Matching is determined by comparing the first arrival time difference with i) the arrival time difference of the first virtual source and ii) the arrival time difference of the second virtual source; In response to determining that the first arrival time difference matches the first virtual source arrival time difference, the location of the first virtual source is identified as the location of the media presentation device that presents the media. as well as In response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce the computational burden on the processor. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

25. The apparatus of claim 24, wherein, The meter includes a first audio sensor located at a first position on the meter, a second audio sensor located at a second position on the meter, a third audio sensor located at a third position on the meter, and a fourth audio sensor located at a fourth position on the meter, wherein the processor circuit determines the first virtual source position and the second virtual source position based on the first position, the second position, the third position, and the fourth position.

26. The apparatus of claim 25, wherein, The processor circuit is used for: Determine the first virtual source position of a first virtual audio source radially positioned around the meter, and the second virtual source position of a second virtual audio source radially positioned around the meter, wherein the first virtual source position is different from the second virtual source position; as well as The arrival time difference of the first virtual source of the first pair of audio sensors of the meter is determined based on the location of the first virtual source, and the arrival time difference of the second virtual source of the second pair of audio sensors of the meter is determined based on the location of the second virtual source.

27. The apparatus of claim 26, wherein, The first pair of audio sensors includes the first audio sensor and the second audio sensor, and the second pair of audio sensors includes the first audio sensor and the third audio sensor.

28. The apparatus of claim 27, wherein, The processor circuit is used for: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; as well as The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time.

29. The apparatus of claim 28, wherein, The processor circuit is used for: The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance between the second virtual source and the first audio sensor and the speed of sound; The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; as well as The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

30. The apparatus of claim 24, wherein, The processor circuit is used for: The first audio record is processed using a short-time Fourier transform to obtain a first audio transform with a first time frequency band; The second audio record is processed using the short-time Fourier transform to obtain a second audio transform with a second time frequency band; as well as The third audio record is processed using the short-time Fourier transform to obtain a third audio transform with a third time frequency band.

31. The apparatus of claim 30, wherein, The processor circuit is used for: Calculate the first inter-channel time difference between the first audio transform and the second audio transform, where the first inter-channel time difference represents the first arrival time difference; and Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

32. The apparatus of claim 24, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

33. A measuring instrument, comprising: Multiple audio sensors; At least one memory, wherein the at least one memory stores instructions; and Processor circuitry, the processor circuitry being configured to execute the instructions to: Generate multiple virtual sources that are radially spaced at different angles relative to the meter; Determine the arrival time difference of each virtual source; Access the audio recording from the audio sensor of the meter; Determine the first arrival time difference of the first audio record associated with the first pair of audio sensors of the meter; Determine the second arrival time difference of the second audio record associated with the second pair of audio sensors of the meter; Compare the first arrival time difference and the second arrival time difference with the virtual source arrival time difference; In response to determining that the first arrival time difference is consistent with the arrival time difference of a corresponding virtual source: The location of the first virtual source, associated with the arrival time difference of the corresponding virtual source, is identified as the location of the media presentation device that presents the media captured by the first pair of audio sensors; and In response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce noise. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

34. The measuring instrument as claimed in claim 33, wherein, The first pair of audio sensors includes a first audio sensor and a second audio sensor, and the second pair of audio sensors includes a third audio sensor and a fourth audio sensor.

35. The measuring instrument as claimed in claim 33, wherein, The first pair of audio sensors includes a first audio sensor and a second audio sensor, and the second pair of audio sensors includes the first audio sensor and a third audio sensor.

36. The measuring instrument as claimed in claim 33, wherein, To determine the arrival time difference of each virtual source, the processor circuitry is used to: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; as well as The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time.

37. The measuring instrument as claimed in claim 36, wherein, The processor circuit is further used for: The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance between the second virtual source and the first audio sensor and the speed of sound; The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; and The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

38. The measuring instrument as claimed in claim 33, wherein, The processor circuitry is used to process the audio recordings from the audio sensor using a short-time Fourier transform to obtain an audio transformation with time-frequency bands.

39. The measuring instrument as claimed in claim 38, wherein, The processor circuit is used to calculate a first inter-channel time difference between a first audio conversion and a second audio conversion, wherein the first inter-channel time difference represents the first arrival time difference; and Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

40. The measuring instrument as claimed in claim 33, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

41. A non-transitory computer-readable medium comprising instructions that, when executed, cause a machine to perform at least: Generate multiple virtual sources that are radially spaced at different angles relative to the meter; Determine the arrival time difference of each virtual source; Access the audio recording from the audio sensor of the meter; Determine the first arrival time difference of the first audio record associated with the first pair of audio sensors of the meter; Determine the second arrival time difference of the second audio record associated with the second pair of audio sensors of the meter; Compare the first arrival time difference and the second arrival time difference with the virtual source arrival time difference; In response to determining that the first arrival time difference is consistent with the arrival time difference of a corresponding virtual source: The location of the first virtual source, associated with the arrival time difference of the corresponding virtual source, is identified as the location of the media presentation device that presents the media captured by the first pair of audio sensors; and In response to determining that the second audio sensor is generating background noise that negatively impacts the audio of the media being presented by the media presentation device, the second audio recording is removed to reduce noise. The meter collects audience measurement data, and The meter is capable of detecting the audio configuration of the media presentation device to select an audio sensor of the meter corresponding to the audio configuration of the media presentation device, thereby facilitating the recognition of signatures included in the audio output by the media presentation device.

42. The non-transitory computer-readable medium of claim 41, wherein, To determine the arrival time difference of each virtual source, the instruction also causes the machine to execute: The first time when the first virtual signal arrives at the first audio sensor is calculated based on the distance and speed of sound between the first virtual source and the first audio sensor. The second time when the first virtual signal arrives at the second audio sensor is calculated based on the distance between the first virtual source and the second audio sensor and the speed of sound; The arrival time difference of the first virtual source is determined based on the difference between the first time and the second time.

43. The non-transitory computer-readable medium as claimed in claim 42, wherein, The instructions also cause the machine to perform: The third time when the second virtual signal arrives at the first audio sensor is calculated based on the distance between the second virtual source and the first audio sensor and the speed of sound; The fourth time when the second virtual signal arrives at the third audio sensor is calculated based on the distance between the second virtual source and the third audio sensor and the speed of sound; and The arrival time difference of the second virtual source is determined based on the difference between the third time and the fourth time.

44. The non-transitory computer-readable medium of claim 41, wherein, The instructions also cause the machine to process the audio recording from the audio sensor using a short-time Fourier transform to obtain an audio transformation with time-frequency bands.

45. The non-transitory computer-readable medium of claim 44, wherein, The instructions also cause the machine to perform: Calculate the first inter-channel time difference between the first audio transformation and the second audio transformation, where the first inter-channel time difference represents the first arrival time difference; and Calculate the second inter-channel time difference between the first audio transformation and the third audio transformation, where the second inter-channel time difference represents the second arrival time difference.

46. ​​The non-transitory computer-readable medium of claim 41, wherein, The meter provides a combination of audience measurement data and personnel measurement data.

Images