WO2016086905A1

WO2016086905A1 - Method for measuring audiences

Info

Publication number: WO2016086905A1
Application number: PCT/CR2014/000007
Authority: WO
Inventors: Carlos Luis CORDERO JIMÉNEZ
Original assignee: Monitoreo Tecnológico, S.A
Priority date: 2014-12-05
Filing date: 2014-12-05
Publication date: 2016-06-09

Abstract

The invention relates to a method for measuring audiences using the sound collected by mobile devices carried by people, in order to produce the frequency spectrum thereof, compress same and transmit same to a processing centre. In addition, the frequency spectrum of the sound emitted by multiple radio and television transmitters is collected for the purposes of comparing them and thereby determining the similarities between both sounds, and, in the event that a match is determined, generating a report, and calculating the audience of every electronic medium based on these reports, be it AM radio, FM radio, satellite radio, aerial television, cable television, satellite television, signals transmitted via Internet, telephone, wireless radiofrequency transmissions, transmissions via electrical conductors, light transmissions with or without fibre optics, transmissions via material media using acoustic vibrations, as well as signals recorded in mechanical, magnetic, electrostatic, semiconductor, chemical, biological and graphical printed media.

Description

HEARING MEASUREMENT METHOD

FIELD OF THE INVENTION

The present invention relates to the technological field and consists of a method of audience measurement.

BACKGROUND OF THE INVENTION

Audience measurement is used to know the time and schedule exposure of a group of people to radio, television and internet broadcasts. This information is used to know the preference of consumers and catalog them by age range, gender, income and others.

1. Measurement of audiences. Definitions and foundations:

1.1 Definitions:

We will start by defining audience, which has several meanings in the Dictionary of the Royal Spanish Academy, but which in our case, refers to two very specific:

5. f. Audience who attends radio and television programs, or who attends an act or show.

7. f. Number of people who receive a message through any means of communication.

Another definition refers to the audience as "the fraction of the reference population exposed to a media outlet."

Also, the Dictionary of the Royal Academy defines the "audience index" as the "Number or percentage of people who follow a radio or television station or chain or a particular program."

It can be indicated that the audience measurement consists of indicates the percentage of households or people with the television turned on on a channel, the radio turned on at a station or visits to a website on a specific day and time, in relation to the total of people with television, radio or internet access. Audience measurement is indispensable for modern marketing, advertising and media programming. This measurement of the audience is a statistic that attempts to reflect reality, precisely. It has been seen as an application of surveys and has serious economic implications.

Knowing the audience levels of the different radio stations is very useful, mainly for the different parties involved in the realization of advertising: broadcasters, advertisers and advertisers. Thus, the audience levels per hour can be used by the programmer of a station to know the preferences of the listeners in their different schedules. In addition, for the publicist and the advertiser it is of great value to know the levels of audience and penetration that the different stations have to achieve maximum effectiveness of their advertising strategies focusing them on the target market of advertising.

The behavior of the audience is conditioned by the medium. In the case of the cinema we find a selective spectator, since the election that carries out implies an economic disbursement, a displacement and the programming fan is more restricted than the television. Also influencing factors outside the viewer, such as the capacity of the room, schedules, exhibition dates, etc.

The existence of the "Global Guidelines for Television Audience Measurement (GGTAM)" (1998), which establishes the necessary, desirable and admissible elements for any measurement of international class audiences (World Federation of Advertisers, 2008), should be emphasized. (Look at annex 1).

The following table includes the main audience measurement companies that operated until 2006:

Major to ic a god - 2006

See: http: //www.oocities.orq/vifibio/05PRINCIPIOSENPROCESAMIENTODESENALES.PDF (07.27.2014)

With regard to television, it is a more heterogeneous viewer, since it is a free and domestic medium, therefore, easy and convenient access. This type of audience is less selective than that of cinema, because it basically looks on television for entertainment.

These audiences could be called "passive" since it is believed that the recipient does not interact with the message. Some theories tell us about active audiences: The concept of active audience appears around 1970 criticizing among others the hypodermic theory. It involves changes in the recipient of the communicative activity. It brings a new paradigm that questions that of the linear paradigm-

Taking into account the heterogeneity of television audiences, the new trend of the networks is to stratify the public to attract minorities to specialized programs, as was already the case in the journalistic, film and music industry. 1.2 Non-electronic audience measurement systems:

Questionnaires: they serve to ask the viewer things. They are used in telephone coincidence studies (telephone a sample of people in order to find out what its members are watching on television at the precise moment of the call). They may also require the respondent basic sociodemographic information. They also use population censuses, sample censuses and their updates, to learn about the structural and sociodemographic characteristics of the universe to study. The interview allows to study the preferences of the viewers or listeners in a more direct and sincere way; know the opinion about the programming, also the habits of the public, know the strengths and weaknesses of the competition, if they receive the signal clearly, or any other qualitative or quantitative analysis that is required. Consumption diaries: They serve for the viewer himself to record his television consumption (who, what day, at what time, he saw which channel, for how long), either at the same time he watches television or immediately after.

Statistics: The statistical measurement of audiences dates back to the 1930s, when in the United States, Archibald M. Crossley, with the support of the National Association of Advertisers (ANA) conducted the first studies on radio audience levels.

This was carried out in the United States, initially through telephone interviews, in which the audience of the 24 hours prior to the call was recorded, compiling the record of which programs and stations were heard, by whom and which were preferred.

The methodology used initially underwent several changes to try to solve practical problems of measurement, and by the appearance of new companies providing the service. Among the differences that were appreciated in the techniques used by the different companies were: the geographical coverage area, the measuring instrument used, the method of information gathering and the records made in the interviews, which implied basically different results.

The differences between methodologies originated marked discrepancies ² between the results of the measurements; The main reason was that the comparisons were simply not valid. Something important to note is that, recognizing the differences between existing measurement systems, the reports included warning notes (for example: ³ These results cannot be projected to the entire United States. ² ) which were invariably ignored by users.

1.3 Electronic Audience Measurement Systems: One way to avoid the problems inherent in survey-based methodologies is to measure tuning electronically, to have daily, accurate and continuous records over time. Radio audience levels measured through electronic instruments were carried out in the United States for the first time by AC Nielsen in the mid-1940s. The radio electronic record usage project was abandoned in 1964 to lead to the development and improvement of the "peoplemeter", an instrument used to date to measure the audience on television.

Audimeters: These are electronic devices that keep track of the tuning time of one or the other station, in different homes in the Spanish territory. This system provides accurate estimates at national and regional levels. The most serious drawback of electronic measurement systems is to obtain the audience of local stations

They are generally granted a clear superiority over the methods of declaration because it overcomes the imperfections and limitations of the human being as a declarer of their behavior and because the audimetry operations provide a level of detail and a speed in the delivery of results impossible to match by the technique of surveys. Arbitran and Telecontrol are two of the leading companies in the manufacture of audimeters in Europe.

MobilTRAK: Through their EMU ^' s measuring devices (Electronic Measurement Units) they can detect the emissions of car radio oscillators and determine the frequency of the tuned station. They place the EMU ^' s on highways, communication nodes and other high-volume traffic locations and passively detect how many of the passing cars carry the radio on and the station tuned at the time of passage.

The technology is restricted to the broadcasts in AM and provides the evolution of the share of the different chains throughout the day. The system has statistical representation problems and does not provide the sociodemographic characteristics of the listeners. When several cars pass at once only one of them is detected and registered.

The new technology arrived in Latin American lands in 1988, when in Brazil the measurement of television audiences began through "peoplemeters". Today, more than seventy countries use audimeter technology to measure television audience. Today, more than seventy countries use audimeter technology to measure television audience.

Today's measuring devices were designed under the analog standard, which allows the tuned channel to be collected without problems, considering that its location in the radio spectrum is defined and that it is only possible to send a single signal per channel. This changed with the development of different digital environments. A multimedia card that processes the images was incorporated, identifying the tuned channel. This solution is efficient in configurations with external decoder.

However, these devices present significant difficulties because in the digital television transmission the ability of the meters to identify the tuned channel is invalidated, given the multitude of available signals with no perceived interference between different emitters. Each station has the ability to divide its band, generating secondary channels through which to transmit different content, in the sense of a dual broadcast. Simultaneous transmissions have emerged, involving the transmission of different and alternative content by the same means. The identification of the content viewed through the recognition of some section of the information (for example, the audio pattern of the content) becomes very difficult. The habits and behaviors of the audience in relation to media consumption have varied greatly. People have learned that the contents can be viewed in a deferred way through video recorders or other players that record the programs on hard drives. In addition to this, television consumption is no longer carried out exclusively through the television screen. There are a greater number of different audiences, increasingly smaller and homogeneous.

To overcome these difficulties, a methodology consisting of the identification of the television signal was generated through the addition of specific data in the signal transmitted by the transmitter, which was recognized by an audimeter installed in the houses. The first tests were performed with marks on the video signal, but in analog technology. In today's digital era, these brands have been replaced by inaudible codes in the audio signal.

1.4 Measurement of audiences on the Internet: Here the measurement procedures tend to take advantage of the technological advantages that the characteristics of the medium provide. Although traditional survey and declaration methods are still used, a series of methodologies that are specific to the Internet have been developed. These have been classified into three groups, depending on the typology of the basic unit of study:

- User-based methods (user-centric in Anglo-Saxon terminology), where the unit to study is the person: Surveys in the Network and PC ^' S Panels.

- Methods based on servers (site-centric) where the Web site is the immediate object of the investigation: Analyzers of log files and systems of tags.

- Methods based on the advertising providers (ad-centric) where the measurement is carried out from the servers responsible for managing the advertising of a network of Web sites, providing the visitor to one of those Web sites at any time an element different advertising based on previously agreed conditions.

They usually combine request counting through server log records with the use of coolers to estimate coverage and frequency. These methods focus specifically on the audience measurement of advertising banners (both their mere exposure and their click rate through) and many times these measurements have come into conflict with those made by the sites based on their log files (the latter usually provide higher figures).

The three groups of methodologies have relative advantages and weaknesses and, until a relevant technological change occurs, the study of the Internet must combine, in appropriate proportions, studies of the three mentioned typologies.

2. Electromagnetic signals and communications:

An electromagnetic wave is the form of propagation of electromagnetic radiation through space and is produced by accelerating charges. His theoretical aspects are related to the theories proposed by Maxwell's equations.

Electromagnetic waves propagate through space without the need for a medium, and can therefore propagate in a vacuum. This is because electromagnetic waves are produced by the oscillations of an electric field, in relation to an associated magnetic field.

Electromagnetic waves travel at approximately a very high constant speed, but not infinite of 300,000 km per second. Electromagnetic waves propagate through an oscillation of electric and magnetic fields. They are the support of telecommunications and the world today.

Extensive frequency ranges and wavelengths are not clearly defined. There is no clear division between one type of wave and the next. The names given to each type of wave are simply a convenient way to describe the region of the spectrum in which they are.

The following series of images shows the location of the electromagnetic radio waves of the media, in order to locate the frequencies of interest.

3. Signal Filtering:

Signal processing lets you know what properties a set of acquired data has, in itself, it allows you to take a signal to a way in which it can be interpreted better.

Filtering is done to adapt a received signal to be interpreted in a better way or clean it or separate it from other signals, so that it can be seen more clearly. Filters help in this regard, eliminating unwanted or uninteresting frequency components.

3.1 Filter classification:

According to their effect, filters are divided into two categories:

• Frequency shapers: Those that change the spectrum

· Frequency selective: those that eliminate certain frequencies and let others pass.

• Depending on their implementation, the filters are divided into: • Analogs: Analog filters use physical components (resistors, capacitors, transistors, etc.).

- Liabilities: They do not need food. They are made with resistors, capacitors ¹ and coils. They have high input impedance, which loads the previous stage and the gain depends on the frequency. They have low output impedance and the output loads the filter. They use high inductances to filter at low frequency.

- Assets: They need food. They have large input and very low output impedance. Resistors and capacitors are the constituent elements. The dynamic range of output voltage is limited by the saturation of operational amplifiers. Bandwidth is limited. Operational amplifiers generate noise

^■ Digital:

Digital filters modify a discrete signal through mathematical operations.

- FIR: Finite Impulse Response: They are very stable because they do not have poles, they also have a linear phase under certain conditions. Filters with very steep slopes can be designed, making it almost an ideal filter. - MR: Infinite Impulse Response: It is the type of digital filter resulting from approximating an analog filter. According to their function, filters are classified as:

^■ Low pass:

They are those that allow the passage of zero frequency signals up to a certain predetermined value called the "upper cutoff frequency" of the filter, for which the attenuation produced by the filter is 3 db. This means that in that place half of the input power is eliminated.

• Pass-high:

They are those that allow the passage of signals from a frequency called "lower cutoff frequency", to a higher one, which theoretically in an ideal filter extends to infinity.

^■ Pass-band filter: They allow the passage of signals whose frequencies are between two, called "upper and lower cutoff frequency" respectively. It can be built by combining low pass and high pass.

^■ Band Eliminator filter (band rejects): They do not allow the passage of signals, whose frequencies are included between two others, called "upper and lower cutoff frequencies". It can be constructed by combining low pass and high pass. 3.2. Digital fixtures:

A digital filter is a filter that operates on digital signals. It is a mathematical operation that takes a sequence of numbers (the input signal) and modifies it by producing another sequence of numbers (the output signal) with the aim of highlighting or attenuating certain characteristics.

It can exist as a formula on a paper, a loop in a computer program, as a circuit integrated in a chip.

The applications used are:

- Signal separation

- Recovery of distorted signals:

- Sound synthesis: creation or modification of signals to mold spectra or waveforms and achieve the desired auditory effect.

3.3. sampling:

It is the operation that consists of taking periodic samples of the values of a continuous (analog) signal at regular intervals of time (or of the independent variable). The frequency at which the values are captured is called the sampling frequency.

It is the first process involved in the conversion of an analog signal into digital.

The other mathematical processes are:

- Quantification: assignment of a margin of value to a single output level for example by rounding or truncation over a given precision, and

- Coding: translation of quantified values into a code, usually binary.

Quantification, unlike sampling, is not reversible as there is a loss of information that results in an error called quantization noise. During sampling the signal is still analog (it can take any value), from quantization the signal becomes digital (it already takes finite values).

This transcription to digital signals is done to facilitate its processing and immunize the signal resulting from noise and other interference to which analog signals are more sensitive.

4. Fourier Analysis: The Fourier Transform is an analysis tool widely used in the scientific field (acoustics, biomedical engineering, numerical methods, signal processing, radar, electromagnetism, communications, etc.) Transforms a signal represented in the time domain to the frequency domain, without altering its information content, it is only a different way of representing it.

Fourier analysis allows us to decompose a complex signal into a set of single frequency components, although it does not indicate the moment in which they occurred. This decomposition is useful for stationary signals: the components of the frequencies that make up the complex signal do not change over time.

The mathematical function that defines the transformed is:

4.1 Fourier in telecommunications:

The study of telecommunications is based on the mathematical methods established by Joseph Fourier, initially generated for the study of vibrations. They are used to analyze the behavior of electrical signals that vary in a complicated way as a function of time.

Fourier Analysis is fundamental in this study, as it provides the mathematical techniques with which the engineer can describe signals and systems not only in the time domain but also in the frequency domain. Fourier Analysis is based on the representation of a complicated function as a sum of sinusoidal functions, and its usefulness depends on two important physical facts: the invariance in time and linearity.

In communication systems the signals are magnitudes that vary over time, such as voltages and currents. The functional elements of a system are the electrical circuits and both these and the signals can be represented in the "time domain" if the independent variable is time (t), or in the "frequency domain" if the independent variable is the frequency (f).

In the analysis and study of communication systems it is necessary to describe or represent the signals and systems in the frequency domain, which leads to the concepts of "spectrum" and "bandwidth". The spectrum-temporal representation of signals and systems is possible by Fourier Spectral Analysis: Series and Transformed.

The electrical signals used in communication systems are generally represented in the time domain where the independent variable is t (time).

However, in the analysis of communication systems it is imperative to also describe the signals in the frequency domain where the independent variable is f (frequency). This means that a Temporal signal can be considered as constituted by a number of frequency components, generally sinusoidal signals, with a given amplitude, phase and frequency. That is, although a signal physically exists in the time domain, it can be said that it is formed by a set of components in the frequency domain, called the "spectrum" of the signal.

Fourier series allow to describe a signal, a function of time, as a superposition of simpler (sinusoid) signals of several frequencies multiples of the fundamental frequency. The frequency spectrum is a measure of the distribution of amplitudes or phases of each frequency. The process that quantifies the different intensities of each frequency is known as spectral analysis.

A periodic signal can be represented by a graph of arrows parallel to the axis of the ordinates of height (intensity) in the frequency. A representation of the spectrum of amplitudes in stripes of the signal is thus obtained. These spectra are discrete. In the case that they are real, the signal has only one frequency representation. For the Fourier transform the spectrum will be continuous.

4.2 Fourier in phonetics: Most phonologists use spectrograms to conduct their investigations. However, for non-stationary signals, sections or windows must be taken where it can be considered stationary and thus be able to apply the Fourier Transform. To perform the complete analysis we must take a sequence of windows to observe the evolution of the frequencies of the original signal. Applying the TF to a window results in a set of values that correspond to the amplitudes of the different frequencies. As the difference in amplitudes is. very different, a logarithmic scale is usually applied for coding. In addition, a minimum signal level is established, below which noise is considered and eliminated, and maximum level for color coding. The windows can be overlapping, contiguous or separated, depending on the intensity of information we want to visualize.

5. State of the art

There are also various inventions around audience measurement systems that (which although precedent the present invention) differentiate this invention in different aspects that we will see.

The "Real-time, mobile, audience measurement system" (CA 2441373 A1, PCT / SE2013 / 000006) is a system that identifies, records and reports, and transmits the information to a radio station with the level of rating that is currently encrypting the information. The station and the listening volume information gives an indication about statistical ranges of the popularity of the content either on radio or on television and you can sort the information to order it gradually. Although the device can be used for both radio and television, the system focuses primarily on measuring radio data, specifically vehicle radio and features:

• A unit in a vehicle.

· A remote unit.

• A central server.

The vehicle unit identifies the station and wirelessly transmits this information to the remote unit, which then transmits the information to the central server over the Internet. The central server collects data from all sources and carries out the processing of the transmitted information. However, this background does not have a method of separation and segmentation of the collected audio that determines the quality of the data and instead only collects the volume of the audio without any processing, nor temporal segmentation that is the point that the present invention intends to solve. The "Audience Measurement System" (EP 0669070 A1, PCT / US1994 / 01 1795) aims to measure visual ratings and audio content by the media. It uses a watermark by using a mobile communication device comprising a receiver and a server. In which the communication between the server and the mobile communication device is carried out through a transmission medium. The audience measurement system uses auxiliary codes to identify coded broadcast programs to be selected for viewing or listening, and collects data or signals from programs that can be used later to identify uncoded broadcast programs. It is, however, an invention that does not provide a solution to the global measurement of audiences because it is for use by households only, does not describe how the fingerprint is extracted, does not use synchronization in the collection of audios, the method of extraction and processing of the fingerprint is different from the present invention because it does not contemplate the different methods of audio collection, nor transmission thereof for processing

The "System for measuring the audience rating of a broadcasting station" (EP 0309326 B1) measures audiences by recording on memory cards. The cards with electronic chips, have units of measurement that emit in parallel the moment of transmissions of the station. The parallel transmission may be an encoded broadcast in the radio frequency band used. Preferably, this parallel transmission is carried out by means of a computerized telephone communications server to which the memory cards of the listeners are linked by a reader / recorder connected to a modulator interconnected with the server. The audience measurement indices are shown that allow a correct analysis of the information. However, this invention does not process or compare sounds from any source, so its system and commercial application is different.

The "Audience measurement system and method for digital broadcasts" (WO 2001061892 A3, PCT / US2001 / 005104) describes a method in which digital broadcast audience is measured, and performs an alignment program verification, in addition to the statistical audience samples. They are collected from the data aggregated by the broadcasting body for this purpose. It has an encrypted channel ID and other program identification data, which are placed in the service ID stream and the decoder or receiving device is directed to the output of this data to the communication port of the device. It also has a probe connected to the communication port, in order to extract the data and decoders for the statistical use of digital programming measurement. This invention is also not a solution in the market because it is not able to compare sounds. On the contrary, it decodes digital codes transmitted by the radio station to establish its origin and the moment of transmission, so it does not solve the problem that our invention does.

The "Network resource monitoring and measurement system and method (US 8495198 B2) consists of a method and analysis system for the measurement of multiple data sources through a communications network, in order to determine the information or use of one or more resource servers. Prepares a collection of data and processing media that collect and process the data sources that are sent to a report server at the request of interested parties. The database uses the strengths of the industry Most widely used research media, such as TV data notes, radio ratings data, reader surveys and service use questionnaires.This method is used as a more integrated method of producing television and radio data sets. through cross-references with other media and consumption variables. The television database is periodically updated so that the The television or radio program provides the data that are of interest to the contracting operators, but it is a measurement that has full dependence on that database as a variable that generates the data. This background does not compare sounds or describe sources, but is limited to a simple statistical analysis system.

The "System and method for obtaining comprehensive vehicle radio listener statistics" (WO 2003049339 A3, PCT / US2002 / 038174) uses a computer program to obtain statistics of radio listeners of global vehicles, based on parameters such as radio status (by for example, the status on / off and CD / Tape / AM / FM), the volume of the radio, the information of the memorized station, the current frequency adjustment (i.e. the station identification), and global positioning by satellite (GPS). It has a field unit placed in the vehicle for the collection and transmission of these parameters to a base station. The system monitors store all events related to the interaction of the occupants with the vehicle radio, including automatic detection of the selected radio station through a speaker port. The stored data is transmitted to the central data collection computer of a base station for immediate collection and analysis. The system is capable of producing detailed reports containing error-free measurement statistics, such a system can be made available to subscribers, such as broadcasters, corporate advertisers, advertising agencies and the like. This background does not compare sounds, rather it collects information on the status of a car radio and transmits them, it also injects codes to the tuner of the same radio to determine its tuning but still does not solve the comparison problem that our invention does solve. None of them manages to solve the cost efficiency problem brought by the method object of the present invention, and none of them achieves statistics with such levels

high reliability, since it is a method of obtaining and comparing the data cleanly.

DESCRIPTION OF THE INVENTION

The methods most commonly used so far for audience measurement are: a. Surveys: select a sample of people and households that represent the universe and ask them the preference of selecting electronic media throughout the day. This method has a low sampling rate and the sample quantity is usually very limited, this is the most popular method used in radio. b. "People Meter": It is an electronic method of capturing is a device that is installed next to the television and intercepts the original radio frequency signal so that the user indicates to the device which channel he wishes to watch, so that information is recorded accurately . The user must also indicate by means of a remote control who is the person who is watching television. The information collected is transmitted to a computer where it analyzes and calculates the audience level. This method is the most popular and only serves for television c. "Portable People Meter": from the Arbitran company, it is a system consisting of a portable device that each sample subject carries at all times to collect "water tracks" that are inaudible codes that are transmitted along with the radio sound and television. These codes are generated by a specialized team installed in each of the radio and television stations. The codes collected by the portable device are transmitted to a computer to analyze and calculate the audience level. This method works for radio and television.

The number of people monitored indicates the accuracy of the measure. The high cost of many of these methods in the market limits the sample size thereby increasing the measurement error. The methods left to the discretion of the test subject indicate who is exposed to the transmission causes a decrease in the statistical confidence of the sample therefore the data obtained even if they have low statistical errors will not indicate the reality of the audience's behavior. The method of "People Meter" is considered invasive because it alters the customs of the audience too much when the television reception team intervenes, and the obligation to follow a complex protocol to indicate which individuals are exposed to the transmission. Usually the test subjects leave the suggested protocol, for this reason the level of statistical confidence drops considerably. The technology we present precisely overcomes these obstacles and presents a more reliable and reliable measurement system that allows the interaction of various media.

The present invention consists of a method to statistically measure the audience of transmissions by electronic means, for which sound is collected by mobile, portable, electronic, cloud or fixed devices. The sound is then used to obtain the frequency spectrum that is compressed and transmitted to a remote process center consisting of multiple computers, where the frequency spectrum of the sound emitted by multiple radio and television stations is collected.

Both spectra are compared to determine the similarity between the sound heard by people and the sound emitted by electronic media. When there is a match, a report is generated and the set of matches is used to calculate the audience of each electronic medium.

The electronic media in which the audience is measured includes: AM and FM analog and digital radio, satellite radio, aerial television, cable television, satellite television, terrestrial television, signals transmitted over the Internet, telephone, wireless radio frequency transmissions, transmissions through electrical conductors, light transmissions with and without fiber optics, transmissions by material means using acoustic vibrations, signals recorded in mechanical, magnetic, electrostatic, semiconductor, chemical, biological and printed graphics media.

The process is simple and includes the following steps:

STAGE 1 REGARDING THE DATA CAPTURED

1 The captured sound is compressed using one of the following comprehension algorithms:

Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Loss Method, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation, using a microphone from a portable or stationary electronic device located a short distance from the person who It is part of a statistical sample.

2 Each of the sounds is separated from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes of one at 300 seconds

3 The sound separation is repeated successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds, leaving a record of the exact time of the start of capture.

4 The captured and separated sound is compressed using one of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation.

STAGE 2 WITH REGARD TO THE SIGNAL PROCESSING

1. The sound emitted by electronic means is digitized. 2. Each of the sounds is separated from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes, from one to 300 seconds, always using the same sound separation parameters used for short distance devices of the person who is part of the statistical sample.

3. The separation of sounds in frequency ranges is repeated successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds with record of the exact time of the start of capture.

4. The captured and separated sound is compressed, using a list of algorithms that are selected from the following by any of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 algorithm, Adaptive Differential pulse-code modulation. COMPARISON STAGE 3

One or both sounds captured with or without separation at various frequencies are transmitted to a remote computer where they will be compared.

The frequency spectra of the sound emitted by each electronic medium at a time and the sound received by each person that is part of the statistical sample at the same moment of capture are compared, making continuous and successive comparisons.

If the comparison is positive, the moment of capture of that positive comparison, the electronic media code and the code of the test subject are inserted into a database system.

A statistical projection of the audience is calculated using positive comparisons.

This method can use a wired or wireless connection with the device that is reproducing the sound heard by the person who is part of the statistical sample to capture the data, and the sound can be received by means of a microphone, an accelerometer or directly from other equipment by means of a wired or wireless connection and can be digitized with equipment located in the facilities of the stations or in remote places. The method in question allows the capture, digitization and processing of sound emitted by electronic means such as computers, cell phones, tablets, sound reproduction equipment, video reproduction equipment, sound recording equipment, video recording equipment, equipment with microprocessors, equipment with microcontroiators, videogames equipment, electronic agendas, microprocessors, microcontroiators, digital signal processors, programmable logic devices or other device where the sounds emitted by these means are digitized. This process allows the digitalization of the sound emitted by electronic media to be alternated over time and in fixed or variable periods, and alternates the digitalization and process of the sound received by the people who are part of the statistical sample, with the use normal of the equipment used for digitization, using wired, wireless, fiber optic, Internet, magnetic media, mechanical media, chemical media, biological media, electrostatic media, semiconductor media and printed graphics systems to transmit the collected information.

The sounds captured from both the person who is part of the statistical sample and that transmitted by the electronic medium are transmitted to an intermediate node and retransmitted to various nodes until they reach the remote computer where they will be compared.

The invention also relates to a technological platform for measuring spectra of sound frequencies emitted by electronic and communication means, where:

• the sound captured by a microphone located in a portable or stationary electronic device of short range is digitized, this sound is processed by separating it into several frequencies and transmitting it to a computer system.

• the sound of the electronic media is digitized, this sound is processed by separating it into several frequencies and transmitting it to a computer system.

• Both data captured at the same time are compared to determine the sound matches and determine the time, means and person where the measurement coincides.

• Statistical projections of measurements that are sent electronically to a database are generated.

In summary, the process of measuring audiences by electronic means includes the following steps: a) A signal captured in devices at a short distance from the person who is part of the statistical sample is processed through the following steps: a.1. Using a microphone or accelerometer of a portable or stationary electronic device located a short distance from the person who is part of a statistical sample, the sound is captured and compressed by any of the following compression algorithms: Lempel-Ziy, Lempel- Ziv-Renau, Huffman coding, Psychoacoustic method with loss, Vorbis algorithm, MP3 algorithm, Adaptive Differential pulse-code modulation a.2 Each sound is separated from two to one thousand twenty four frequency ranges using digital filters, analog filters, transformed Fourier or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes from one to 300 seconds. a.3. The separation of sounds is repeated successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds, leaving a record of the exact time of the start of capture. a.4. The captured and separated sound is compressed by one of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau,

Huffman coding, Psychoacoustic method with loss, Vorbis algorithm, MP3 algorithm, Adaptive Differential pulse-code modulation. b) The processing of the signal emitted by electronic means is carried out by the following steps:

b.1. Sound emitted by electronic media is digitized.

b.2. Each of the sounds is separated from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes, of one at 300 seconds, always using the same sound separation parameters used for short distance devices of the person who is part of the statistical sample.

b.3 The separation of sounds in frequency ranges is repeated successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds with record of the exact time of the start of capture.

b.4. The captured and separated sound is compressed using one of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation. c) One or both sounds captured with or without separation in several frequencies are transmitted to a remote computer where they will be compared. d) The frequency spectra of the sound emitted by each electronic medium at a time and the sound received by each person that is part of the statistical sample at the same moment of capture are compared, making continuous and successive comparisons. e) If the comparison is positive, the moment of capture of that positive comparison, the electronic media code and the code of the test subject are inserted into a database system. f) A statistical projection of the audience is calculated using positive comparisons.

This method can be developed in an App for smartphones, tablets, and any other mobile device and a smart mobile device could also use the described method. The sounds to be compared must be digitized to start the comparison process. Each sound must be separated into several frequencies and then the data obtained is compressed, then the data is transmitted to the processing center and compared to obtain the final result. For digitization, the sound is entered by means of a microphone and is digitized by means of analog to digital converters.

With respect to the frequency spectrum, the digitized sound is broken down into multiple frequencies using some analysis method such as the fast Fourier transform. The sample size for this analysis is variable. The result is the sound power data for each frequency.

For data compression, the frequency spectrum data is compressed to be sent to the equipment responsible for making the comparison. Finally, the spectral data is transmitted by some usual means to the computers that store it and then make the comparison. The comparison indicates how similar it is to the sound received by each person and the sound emitted by each electronic medium, classical comparison methods such as neural networks, and fuzzy logic can be used. By means of already existing statistical analyzes, the audience, error percentages, projections, categories, market segments and all the usual information for this type of service are calculated.

This method generates reports based on the information obtained, which can be in HTML, WORD, EXCEL, etc. Other custom programs can also access the data to prepare their own reports.

As we have already indicated, the method for measuring audiences consists of three fundamental parts: The capture of the sound of the environment of the test subject. The capture of the sound of radio, television and internet stations. And the comparison of the two sounds using special algorithms.

Sound capture of the test subject's environment

This sound is captured by a microphone or by intercepting the sound in the audio player itself that the test subject is using. Or by means of a wired connection with some sound recording device. You can also receive the sound through a wireless device such as a hands-free, bluetooth or a device with WiFi. As an additional method to the microphone can be an accelerometer or an optical interferometer which can listen to the sound from a distance.

This sound is digitized and separated into several frequency ranges, this is done with separation techniques similar to that of equalizers (digital filters and Fourier transform), separation The sound must be compressed to be sent quickly even when the available bandwidth is limited. Compression can be in MP3 format or any other standard format.

It is then transmitted along with the exact time of capture to a central computer for storage and processing. The transmitted sound must be separated into frequencies but the original sound can also be sent.

The device that performs this sound capture can be anyone capable of acquiring, digitizing, storing in a buffer and transmitting the sound, so it can be from a computer, cell phone, tablet or a device designed for this purpose, the device can be portable so that the test subject carries it at all times or it can be stationary and placed in a place close to the test subject.

It can even be a group of people exposed to the same sound.

In case there is no continuous communication channel, the sound must be stored in a memory which is collected and taken manually to a place where it will be transmitted to the central computer. Sound capture emitted by electronic means

The sound emitted by the source can be captured at the same source or by means of radio and television receivers or sets of multiple tuners. If a standard receiver is used then a multi-channel sound digitizer card must be used for example.

Transmissions through the internet can be captured directly on a computer.

The sound must be digitized and sent to a central computer where it is separated into several frequency ranges and given the same treatment as the sound coming from the test subject. The transmission of the information can be direct to the comparison computer or indirectly through various communication nodes.

It is essential that every medium that you want to include in the audience measurement study must be captured. Sound comparison

Pre-captured sounds of both test subjects and electronic media must reach the same point and the time they were captured must be known.

That point can be a central computer or the same sound capture equipment. There the sounds must be compared using existing algorithms for this purpose, this comparison must be synchronous, that means that the hours of capture of both sounds must coincide exactly. To prevent time differences between the two capture devices, a comparison scan must be made from several seconds before to several seconds after the moment they are supposed to be in sync. If both sounds match, the data must be stored in a database system for further processing.

The stored data are: time of capture of the compared segment, name or code of the electronic medium and name or code of the test subject.

For example, if the test subject has a cell phone, while not using it, the phone can capture the sound, compress it and send it. Or if the phone has a radio or television tuner then it can capture the ambient sound and the sound received by the tuner, the tuning can be varied from time to time and in each change it is compared with the ambient sound. If there is a match then the data is sent to the central computer.

Another example would be this: a set of tuners of FM, AM or television are kept in one place which are continuously recording the sound of multiple stations of interest. This data is sent to the central computer. In addition, many test subjects are sending the collected sound with microphones to the central computer. Some comparison program compares each sound of each test subject with each sound of each recorded electronic medium. The comparison hours must be the same, to prevent synchronization errors can be compared from one minute before to one minute after the set time. If there is any coincidence it means that the test subject is exposed to the sound of that particular station. There is a statistical model that takes the set of data obtained and generates the information of audience levels by sex, income, ages, schooling, and other parameters. useful. These models are already existing or you can make special models to work in particular locations. The data generated by the statistical model will be delivered to the customer through a website or by sending emails.

Claims

1. A method to statistically measure the audience of electronic media, where:

to. Sound is collected by mobile, portable, electronic, cloud or fixed devices.

b. Sound is used to obtain the frequency spectrum that is compressed and transmitted to a process center consisting of multiple computers, c. The frequency spectrum of the sound emitted by multiple radio and television stations is collected,

d. Both spectra are compared to determine the similarity between the sound heard by people and the sound emitted by electronic means e. When there is a match, a report is generated,

F. The set of matches is used to calculate the audience of each electronic medium.

2. A method of claim 1, wherein the electronic means in which the audience is measured comprises: AM and FM analog and digital radio, satellite radio, aerial television, cable television, satellite television, terrestrial television, signals transmitted by Internet, telephone, wireless radio frequency transmissions, transmissions through electrical conductors, light transmissions with and without fiber optics, transmissions by material means using acoustic vibrations, signals recorded in mechanical, magnetic, electrostatic, semiconductor, chemical, biological and graphic media printed.

3. A method of claim 1, comprising the following steps: a. Processing of the signal captured in devices at a short distance from the person who is part of the statistical sample comprising the following steps: a.1. Compress the captured sound using one of the following comprehension algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman encoding, Psychoacoustic method with loss, Vorbis algorithm, MP3 algorithm, Adaptive Differential pulse-code modulation, using a device microphone portable or stationary electronic located a short distance from the person who is part of a statistical sample,

a.2 Separate each of the sounds from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes of One to 300 seconds. a.3. Repeat the separation of sounds in succession to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds, leaving a record of the exact time of the start of capture. a.4. Compress the captured and separated sound / by any of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation. b. Processing of the signal emitted by electronic means comprising the following steps:

b.1. Digitize the sound emitted by electronic media.

b.2. Separate each of the sounds from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast Fourier transforms to obtain a representation of the power spectrum of the frequencies, using variable sample sizes, from one to 300 seconds, always using the same sound separation parameters used for short distance devices of the person who is part of the statistical sample.

b.3 Repeat the separation of sounds in frequency ranges successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds with record of the exact time of the start of capture.

b.4. Compress the captured and separated sound, using a list of algorithms that are selected from the following by any of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation.

4. Transmit one or both sounds captured with or without separation at various frequencies to a remote computer where they will be compared.

5. Compare the frequency spectra of the sound emitted by each electronic medium at a time and the sound received by each person who is part of the statistical sample at the same time of capture, making continuous and successive comparisons.

6. If the comparison is positive, the moment of capture of that positive comparison, the electronic media code and the code of the test subject are inserted into a database system.

7. Calculate a statistical projection of the audience using positive comparisons.

A method of claim 1, wherein a wired or wireless connection is used with the device that is reproducing the sound heard by the person who is part of the statistical sample to capture it.

A method of claim 1, wherein the sound of the electronic medium and the sound reproduced by the person who is part of the statistical sample can be received by means of a microphone, an accelerometer or directly from other equipment by means of a wired or wireless connection and can be digitized with equipment located in the facilities of the stations or in remote locations.

10. A method of claim 1, which captures, digitizes and processes the sound emitted by electronic means such as computers, cell phones, tablets, sound reproduction equipment, video reproduction equipment, sound recording equipment, audio equipment video recording, microprocessor equipment, microcontroller equipment, video game equipment, electronic agendas, microprocessors, microcontrollers, digital signal processors.

11. A method of claim 1, which captures, digitizes and processes the sound heard by the person who is part of the statistical sample, by means of computers, cell phones, tablets, sound reproduction equipment, video reproduction equipment , sound recording equipment, video recording equipment, equipment with microprocessors, equipment with microcontrollers, video game equipment, electronic agendas, microprocessors, microcontrollers, digital signal processors, programmable logic devices or other device where the sounds emitted are digitized by those means.

12. A method of claim 1, wherein the comparison is performed on the same device that collects the sound heard by the person, on the same device that collects the sound emitted by electronic means, or on remote computer equipment. .

13. A method of claim 1, wherein the capture of the sound heard by the person and the capture of the sound emitted by electronic means is performed on the same equipment.

14. A method of claim 1, which alternates over time and in fixed or variable periods the digitization of the sound emitted by the electronic means, and alternates the digitization and processing of the sound received by the people who are part of the sample statistics, with the normal use of the equipment used for digitization.

15. A method of claim 1, which uses wired, wireless, fiber optic, Internet, magnetic media, mechanical media, chemical media, biological media, electrostatic media, semiconductor media and printed graphics systems to transmit the collected information. .

16. A method of claim 1, wherein the sounds captured from both the person who is part of the statistical sample and that transmitted by the electronic medium are transmitted to an intermediate node and retransmitted to various nodes until they reach the remote computer where they will be compared .

17. A method of claim 1, wherein there are multiple test subjects exposed to the same sound.

18. A technological platform for measurement of spectra of sound frequencies emitted by electronic and communication means, where: o The sound captured by a microphone located in a portable or stationary electronic device of short range is digitized, this sound is processed by separating it into several frequencies and transmitting it to a computer system.

o The sound of the electronic media is digitized, this sound is processed by separating it into several frequencies and transmitting it to a computer system.

o Both data captured at the same time are compared to determine the sound matches and determine the time, means and person where the measurement coincides.

o Statistical projections of the measurements that are sent electronically to a database are generated.

19. A platform of claim 14, wherein the electronic and communication media from which the audience is measured comprises: AM and FM digital and analog radio, satellite radio, air television, cable television, terrestrial television, satellite television, signals transmitted over the Internet,

20. telephone, wireless radio frequency transmissions, transmissions through electrical conductors, light transmissions with and without fiber optics, transmissions by material means using acoustic vibrations, as well as signals recorded in mechanical, magnetic, electrostatic, semiconductor, chemical, biological media and printed graphics.

21. A platform of claim 14, wherein the sounds are separated from 2 to one thousand twenty-four frequency ranges, successively, using digital filters, analog filters, Fourier transforms or Fast Fourier transforms to obtain a representation of the power spectrum of the frequencies and obtain spectra for adjustable periods of 1 to 1000 milliseconds, leaving exact records of the capture start times.

22. A platform of claim 14, wherein sounds captured with or without separation at various frequencies are transmitted to a remote computer for comparison purposes.

23. A platform of claim 14, wherein the moment data, electronic code and code of the test subject of the positive comparisons are stored in database systems.

24. A platform of claim 14, wherein a statistical projection of the audience measurement of electronic media is performed.

25. A platform of claim 14, by means of which automated reports are generated that are sent by the technological means.

26. A platform of claim 14, wherein sound can be received by means of a wired or wireless connection microphone or accelerometer.

27. A platform of claim 14, wherein the emitted sound can be digitized with equipment located in the facilities where the sounds are emitted or in remote locations.

28. A platform of claim 14, wherein the digitization and processing of the sound received by people is alternated with the normal use of the equipment used for digitization.

29. A platform of claim 14, wherein the capture and listening of sounds emitted by electronic means is performed on the same equipment from time to time, selecting tuning moments of various electronic means in the time slots programmed in the system.

30. A platform of claim 14, wherein the sounds captured from both the test subject and that transmitted by the electronic medium are transmitted to an intermediate point and retransmitted to various nodes until they reach the remote computer where they will be compared.

31. A process of measuring audiences by electronic means comprising the following steps: a. A signal captured on devices within a short distance of the person who is part of the statistical sample is processed by means of the following steps: a.1. Using a microphone or accelerometer of a portable or stationary electronic device located a short distance from the person who is part of a statistical sample, the sound is captured and compressed using one of the following compression algorithms: Lempel-Ziv, Lempel- Ziv-Renau, Huffman coding, Psychoacoustic method with loss, Vorbis algorithm, MP3 algorithm, Adaptive Differential pulse-code modulation. a.2 Each of the sounds is separated from two to one thousand twenty-four frequency ranges using digital filters, analog filters, Fourier transforms or fast transforms Fourier to obtain a representation of the power spectrum of the frequencies, using variable sample sizes from one to 300 seconds. a.3. The separation of sounds is repeated successively to obtain the spectrum every certain period, adjustable from one to one thousand milliseconds, leaving a record of the exact time of the start of capture. a.4. The captured and separated sound is compressed using one of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation. b. The processing of the signal emitted by electronic means is carried out by the following steps: b.1. Sound emitted by electronic media is digitized.

b.4. The captured and separated sound is compressed using one of the following compression algorithms: Lempel-Ziv, Lempel-Ziv-Renau, Huffman Coding, Psychoacoustic Method with Loss, Vorbis Algorithm, MP3 Algorithm, Adaptive Differential pulse-code modulation.

C. One or both sounds captured with or without separation at various frequencies are transmitted to a remote computer where they will be compared. d. The frequency spectra of the sound emitted by each electronic medium at a time and the sound received by each person that is part of the statistical sample at the same moment of capture are compared, making continuous and successive comparisons. and. If the comparison is positive, the moment of capture of that positive comparison, the electronic media code and the code of the test subject are inserted into a database system. F. A statistical projection of the audience is calculated using positive comparisons.

32. A method of claim 1, which can be developed in an App for smartphones, tablets, and any other mobile device.

33. An intelligent mobile device with a platform of claim 18 using the method of claim 1.