MXPA00004781A - Voice recognition unit for audience measurement system - Google Patents

Abstract

A member of an audience of a program to which a receiver is tuned is identified by a system which includes first and second microphones, a noise cancellation unit, and a voice recognition unit. One of the first and second microphones picks up relatively more of a voice signal from the member of the audience, and the other of the first and second microphones picks up relatively more of a sound signal from the receiver. The noise cancellation unit substantially eliminates the sound signal from the voice signal. The voice recognition unit identifies the member from the member's voice signal. The first and second microphones, the noise cancellation unit, and the voice recognition unit may be housed in a hand held housing or in a table top housing. The system may also include a passive people meter to passively identify the member, and a program to identify the program or channel to which the receiver is tuned.

Description

VOICE RECOGNITION UNIT FOR HEARING MEASUREMENT SYSTEM Technical Field of the Invention The present invention generally relates to a system of audience measurement that identifies members of a program audience to which a receiver, such as a television, radio or computer, is tuned and / or identifies a program to which a receiver, such as a television, radio or television computer is tuned. BACKGROUND OF THE INVENTION The measurement of program audiences has long been of interest to program providers and advertisers because the audience measurements provide the data from which the effectiveness of the programs and programs can be evaluated. ads. A variety of well-known methods have been expanded to provide an estimate of the total audience of a program, part of a program, and / or a commercial. In addition to the total audience, these methods provide additional estimates of demographically significant audience segments (for example, the number of men and / or women between certain ages who saw a selected portion of a given program). These methods include measuring in a manual and / or automatic way the listening or viewing habits of audience members of statistically selected households and then estimating the total audience. _: > based on these measurements. Members of the audience of statistically selected households are generally called panelists. The measurement of observing and / or listening habits of an audience generally requires three separate determinations: (1) a determination of the channels or stations to which the monitored recipients in the statistically selected households are tuned; (2) a determination of the programs that were available on the channels tuned in during the hours during which the receivers were tuned to these channels or stations; and, (3) a determination of the members and visitors of the household who were present in the area of the receivers during the time the recipients were tuned to the selected programs. The three determinations can affect the accuracy of audience measurements. Of the three aforementioned determinations, the determination that possibly affects the accuracy of audience measurements is (3) because members of a receiver's audience are required to cooperate in the identification process. When members are required to participate in the identification process, they often make mistakes, or simply forget to identify themselves or the other members of the audience. For example, a known system that allows members of the audience to identify themselves to identify their presence is a "People Meter". The "People Meter" can be a remote control and / or a plurality of push buttons activated by the members of the audience when entering or leaving the monitored area. However, members of the audience often forget to press any button or sometimes make mistakes to press the correct button. In an attempt to solve this problem, several measurement systems have been developed. For example, US Pat. No. 3,056,135, Currey and co-inventors discloses a system that uses switches strategically placed to count the number of people entering, leaving, and remaining in a monitored area, and a photographic recorder to periodically record the composition of the audience. This approach requires that the photographic record be observed by an operator, which invades the privacy of the members of the audience and imposes an unacceptable cost on the measurement operation. These problems led to a variety of suggestions for non-obstructive passive methods of counting (but not identifying) observers and tracking their movement in the monitored area. For example, Kiewit and co-inventors, in the U.S. Patent of North America Serial Number 4, 644, 509, discloses an ultrasonic sonar system that counts and tracks the members of the audience. However, the system described does not identify the members of the audience and, therefore, requires that the members of the audience identify themselves manually.

Other systems, such as that described by Kiewit in U.S. Patent Serial No. 4, 930.01 1, are based on electronically active labels that are used or carried by members of a hearing and that uniquely identify the corresponding members of these audience members. These systems respond to labels when the members of the audience that use them are within the monitored area of a receiver. However, members of the audience often forget to use their labels, in which case those members of the audience are not identified. Lu in his United States of America Patent Serial No. 4,858,000, and in the United States Patent Serial Number 5,031, 228, discloses an image recognition system for identifying members of a hearing of a receiver. The system uses a video camera to acquire images of the faces of the members of the audience, and a computer subsystem to recognize the faces in these images by comparing these faces with reference faces stored in a database. The system also locates and tracks observers, and includes infrared illuminators that provide a controllable level of imperceptible illumination of the monitored area for the video camera. However, the system is generally expensive and difficult to install. The system disclosed by Lu and co-inventors, in the United States Patent of North America Serial Number 5, 550,928, also uses image recognition to identify the members of the audience. This system incorporates algorithms to find bodies and then heads of members of an audience. Once the head is found, an algorithm that finds faces extracts representations of facial images that are provided to two face recognition routines, a face recognition routine, and a template comparison routine. This system can identify audience members who are moving in the monitored area and / or who are not viewing the cameras. However, it is still complex and expensive. The above-mentioned determinations (1) and (2) can also affect the accuracy of the audience measurements because one of these determinations alone is usually not sufficient to identify the programs to which the receivers are tuned. That is, a determination of the channels to which a receiver is tuned in a statistically selected home does not lead to a determination of the tuned programs without also knowing what programs were available on the tuned channels. Likewise, a determination of the programs that were available on the tuned channels does not lead to a determination of the programs tuned without knowing the channels to which the receiver is tuned in a statistically selected home. Therefore, both determinations are commonly required to determine the programs tuned by the members of the audience. The acquisition of information about the available programs and the insertion of this information into a database for a later comparison with the tuned channels leads to ad q uisition and data entry errors that adversely affect the accuracy of the data. audience measurements. Thomas and co-inventors in U.S. Patent Serial Number 5,481, 294, addresses these problems with an audience measurement system that detects and records auxiliary codes that are inserted and identify the programs to which the receivers are tuned. Therefore, you only have to make a determination if an auxiliary code is present in the observed programs. If the auxiliary codes are not available in the tuned programs, the program signatures are extracted from the programs and compared with previously extracted reference signatures. The measuring apparatus disclosed in this patent may include non-intrusive detectors, such as microphones placed at the receiver, which detect the auxiliary codes on the audio output of the monitored receivers. If microphones are used to detect auxiliary codes, microphones can be provided to detect background noise. The outputs of the detector microphones of the auxiliary code and of the noise-detecting microphones are processed to subtract the background noise from the signals provided by the detector microphones of the code at uxil ia r. This audience measurement system may also include a "People Meter" as described above for additional support. Turner and co-inventors, in U.S. Patent No. 4,907,079, discloses a system for monitoring receiver audiences. Members of the audience provide control commands (such as channel selections) to a microprocessor through an infrared unit that includes a keyboard. The information collected by the microprocessor can be saved and transmitted to a central computer through conventional telephone lines. A motion detector coupled to the microprocessor detects the moment when a member of the audience has entered or left the monitored area. A character generator coupled to the microprocessor allows communication with the members of the audience via alphanumeric messages displayed, for example, on a television screen. Also, a speech output unit coupled to the microprocessor generates speech signals to communicate with the members of the audience. The Turner and co-inventors system also includes a speech input unit coupled to the microprocessor to allow the members of the audience to identify themselves audibly. A microphone detects the audible speech of the members of the audience, amplifies it, digitizes it and provides it to the microprocessor. The microprocessor performs a dense spectrum analysis of energy in the signal to derive voice signatures from the members of the particular audience. These signatures are compared with reference signatures, which were previously saved during the initialization of the system to identify the observers. The microprocessor can also normalize the digital voice data and remove the small amplitude samples in order to minimize the effects of background noise. When the system disclosed in this Turner patent and co-inventors is activated, an initial deployment message is placed on the viewing screen which requests the audience to state their first and last names in order to provide the reference signatures. However, speech recognition itself does not allow for a viable audience measurement system. Additionally, the disclosed signal processing that minimizes the effects of background noise may not be suitable under various conditions. Kimura, in the Patent of the United States of America Serial Number 5,267,323, discloses a voice operated remote control having a transmitter 10A. The transmitter 10A includes first and second microphones M1 and M2. The first microphone M1 serves to insert a voice command and the second microphone M2 serves to capture the noise or amt sound. A noise remover 104 removes amt noise by producing a differential signal indicative of the difference between the signals of the first and second M 1 and M 2 microphones. The present invention relates to a system that identifies the members of an audience of a receiver and that solves one or more of the above mentioned problems. BRIEF DESCRIPTION OF THE INVENTION In accordance with a first aspect of the present invention, a system for identifying a member of an audience of a receiver comprises first and second microphones, a noise cancellation unit, and a speech recognition unit. The first and second microphones are positioned so that one of the first and second microphones captures more than one voice signal from the member of the audience and the other from the first and second microphones captures relatively more than one sound signal from the receiver. The noise canceling unit is coupled to the first and second microphones, and the noise cancellation unit is positioned to substantially eliminate the sound signal from the speech signal. The speech recognition unit is coupled to the noise cancellation unit, and the speech recognition unit is arranged to identify the voice signal of the audience member. In accordance with another aspect of the present invention, a system comprises a microphone, a speech recognition unit, and a program sensor. The system identifies a member of an audience of a program to which the receiver is tuned. The microphone is located to receive a voice signal from the member. The speech recognition unit is coupled to the microphone and is arranged to identify the member of the speech signal. The program sensor is configured to identify the program. In accordance with another additional aspect of the present invention, a system comprises a passive meter, a microphone, a voice recognition unit, and a controller. The system identifies a member of an audience of a program to which a receiver is tuned. The passive measurer is configured to provide a first identification of the member by passively identifying the member. The microphone is located to receive a voice signal from the member. The speech recognition unit is coupled to the microphone and is configured to provide a second identification of the member by identifying the member of the speech signal. The controller is configured to select at least one of the first and second identifications to identify the member. In accordance with yet another aspect of the present invention, an audience recognition unit comprises a housing, a microphone supported by the housing, and a voice recognition unit supported by the housing and coupled to the microphone. The voice recognition unit is set up to recognize a member of an audience of an identifier issued orally by the member. As a result, the audience recognition unit identifies the audience member of a program to which it is installed or a recipient. BRIEF DESCRIPTION OF THE DRAWINGS These and other aspects and advantages of the present invention will be more apparent from a detailed consideration of the invention taken in conjunction with the drawings in which: Figure 1 illustrates the system of the present invention for its use in identifying members of an audience of a program to which a television receiver is tuned; Figure 2 is a block diagram of a system embodiment which is in accordance with the present invention and includes a hearing recognition unit (ARU); Figure 3 is a more detailed diagram of the audience recognition unit shown in Figure 2; Figure 4 is a diagram of a noise cancellation configuration for use in the audience recognition unit shown in Figure 2; Figure 5 illustrates a portable version of the audience recognition unit shown in Figure 3; Figure 6 illustrates a table version of the audience recognition unit shown in Figure 3; Figure 7 is a flow diagram showing a program that can be implemented by the audience recognition unit shown in Figure 3, during a training period; and, Figure 8 is a flow chart showing a program that can be implemented by the audience recognition unit shown in Figure 3 during voice recognition of audience members.

Detailed Description of the Invention A hearing measurement system 10, which determines the members of an audience to a program to which a monitored receiver is tuned, is shown in Figure 1. The audience measurement system 10 includes an audience recognition unit 12, which is shown in Figure 1 and has a portable unit but alternatively can be any other type of unit, such as a table unit. The audience recognition unit 12 allows the members of the audience to be identified either passively or actively, and may or may not interact with a recipient 14 during the identification process. A program sensor 16 cooperates with the receiver 14 to determine the programs (and / or channels) to which the receiver 14 is tuned. The program sensor 16 can be any known sensor that determines channels (and / or programs). For example, the program sensor 16 can detect auxiliary codes that are inserted into the signals of the program to which the receiver 14 is tuned and which specifically identify the tuned programs. In addition, or alternatively, the program sensor 16 may include an identification extractor that extracts characteristic identifications of the program signals to which the receiver 14 is tuned so that the identifications are compared with reference identifications in order to identify the programs (and / or channels) to which the receiver 14 is tuned. In addition, or alternatively, the program sender 16 can be an apparatus that determines the channels to which the receiver 14 is tuned by monitoring the tuner of the receiver. receiver 14. The audience measurement unit 12 and the program sensor 16 communicate with a base unit 18 which may be a unit on site or a unit at home. As is known in the art, a site unit may be provided in each audience area in a home, and all site units within the home may be communicated to a home unit through the electric power lines in the home. walls of the home. Then the home unit can communicate through the public telephone system with a central unit that collects hearing and tuning information from other households so that the information is combined into useful reports for program providers, advertisers and the like. In homes where there is only one observation area, the base unit 18 may be configured to receive data from the audience measurement unit 12 and the program sensor 16, and may communicate this data via the public telephone system to the central unit. Therefore, as shown in Figure 2, the audience recognition unit 12 provides its recognition data to the base unit 18 via a communication link 20 which, for example, can be a radio frequency communication link or infrared. The program sensor 16 provides this program identification data to the base unit 1 8 via a communications link 21, which for example, can be a hard cable, but which can alternatively be any other in the communications line such as an infrared or radiofrequency communication link. The base unit 1 8 collects the audience recognition data from the audience recognition unit 1 2 and the program sensor program identification data 16, and can communicate this data through a public telephone system 22 to a site central 24 that accumulates similar data from other households. As indicated in Figure 2, the audience measurement system 10 may also include a passive-person meter 26. For example, the passive-person meter 26 may be of the type that passively identifies the audience members in the audience. an area monitored in accordance with the teachings of the United States Patent Serial Number 5, 550,928 mentioned above. The passive meter 26 provides its people identification data to the base unit 18 via a communication link 28 which, for example, can be a hard wire, but which can alternatively be any other wire in the communications line, such as a radiofrequency or infrared communication link. Therefore, if the audience measurement system 10 includes the audience recognition unit 12 and the passive person meter 26, the passive person meter 26 can be used to passively identify the m e mers. of the hearing in case the hearing recognition unit 12 can not do so. Alternatively, the audience recognition unit 12 can be used to actively or passively identify the members of the audience in the event that the passive meter 26 can not passively identify the members of the audience. As a further alternative, the outputs of the audience recognition unit 12 and the passive person meter 26 can be applied simultaneously to a soft decision logic that identifies the members of the audience based on both outputs. In any of these alternatives, it increases the confidence that the members of the audience can be recognized accurately. The audience recognition unit 12 is shown in greater detail in Figure 3. The audience recognition unit 12 includes first and second microphones 30 and 32. The first and second microphones 30 and 32 are placed in the audience recognition unit. 12 so that one of the first and second microphones 30 and 32 tends to capture relatively more of the voice signals of the members of the audience, and the other of the first and second microphones 30 and 32 tends to capture relatively more of the output from audio of the monitored receiver. For example, the first and second microphones 30 and 32 may be separated so that in most orientations of the audience recognition unit 12, one of the first and second microphones 30 and 32 is closer to the members of the audience and the other of the first and second microphones 30 and 32 is closest to the monitored receiver. In a preferred separation between the first and second microphones 30 and 32 is 17.06 cm, although any other suitable separation can be used. In this embodiment of the present invention, the first and second microphones 30 and 32 may be pointed in opposite directions. Alternatively, one or both of the first and second microphones 30 and 32 may be remotely located from the audience recognition unit 12 so that one of the first and second microphones 30 and 32 is closer to the audience members. and the other of the first and second microphones 30 and 32 is closer to the monitored receiver. The analog output of the first microphone 30 is amplified by an amplifier 34, and this amplified analog output is converted to a digital signal by an analog to digital converter 36. Likewise, the analog output of the second microphone 32 is amplified by an amplifier 38, and this amplified analog output is converted to a digital signal by an analog-to-digital converter 40. Although the first and second microphones 30 and 32 may be positioned or directed so that one or both of the first and second microphones 30 and 32 tend to capture relatively more of the voice signals of the members of the audience and the other of the first and according to microphones 30 and 32 tend to capture relatively more of the monaural receiver's output. The first and second microphones 30 and 32 pick up the voice signals of the members of the audience and the audio output of the receiver 14. Also, the first and second microphones phones 30 and 32 capture background noise rather than the audio output of the receiver 14. To reduce or eliminate the effects of background noise and the audio output of the receiver 14, the outputs of the analog-to-digital converters 36 and 40 are provided to a noise cancellation configuration 42. The noise cancellation configuration 42 processes the outputs of the analog-to-digital converters 36 and 40 to cancel the audio output of the receiver 14 as well as the background noise of the speech signals provided by the members of the audience, so that these voice signals can be more easily identified. The noise cancellation configuration 42 may use signal subtraction to subtract the background noise and audio output from the receiver 14 from the speech signals provided by the audience members. For example, the first microphone 30 is directed towards the members of the audience, and if the second microphone 32 is directed towards the receiver 14, the signal of the first microphone 30 can be called Sp (p means people), and the signal of the second microphone 32 can be referred to as Sr (r meaning reference). In general the signal Sp can be given according to the following equation: Sp = aVp + bVr (1) where Sp is the signal of the first microphone 30, Vp is the contribution to the signal of the first microphone 30 due to the signals of voice of the members of the audience, Vr is the contribution to the signal of the first microphone 30 due to the audio output of the receiver 14 and another background noise, a and b are - parameters described below, is already smaller than b. Similarly, the signal Sr can be given according to the following equation: Sr = cVp + dVr (2) where Sr is the signal of the second microphone 32, c and d are parameters that are described below, and c is less than d . Equations (1) and (2) can be subtracted with a weighting parameter e according to the following equation: D = eSp - Sr = (ae - c) Vp + (be - d) Vr (3) where D is the weighted difference between equations (1) and (2), and e is a weighting parameter described below. If a, b, c, and d, are fixed parameters determined by the environment and the characteristics of the first and second microphones 30 and 32, the weighting parameter e can be selected so that (be -d) = 0, or so that e = d / b >; 1 . Therefore, if (be - d) = 0, the difference D will consist solely of the signal Vp. In this case, the signal Vp of equation (3) is given by the following equation: Vp = D / (ae -c). (4) Since a, e, and c are constants in a given household, these constants can be treated as a single constant k in accordance with the following equation: k = 1 / (ae - c) (5) so that the signal Vp is given by the following equation: Vp = kD. (6) The weighting parameter e can be determined adaptively during periods in which the receiver 14 is turned on but no member of the audience is speaking. During this period, the audience recognition unit 12 can automatically find the weighting parameter e such that D = 0. This weighting parameter e is then inserted into equations (5) and (6) to determine Vp, and Vp is processed by a speech recognition unit 44 as described below. This noise cancellation approach assumes that the first microphone 30 is aimed at, or is closer to, the audience than the second microphone 32 is, and that the second microphone 32 is aimed at, or is closer to, the receiver 14 than what is the first microphone 30. For example, if the audience recognition unit 12 is a portable unit, the first microphone 30 and the second microphone 32 can be mounted on the audience recognition unit 12 so that they are separate and for the user to be instructed to orient the audience recognition unit 12 with the first microphone 30 directed towards the audience and the second microphone 32 directed towards the receiver 14. Likewise, if the audience recognition unit 12 is a table unit, the first microphone 30 and the second microphone 32 again can be mounted on the audience recognition unit 12 so that they are separated and so that they are instigated. The user is directed to orient the audience recognition unit 12 with the first microphone 30 directed towards the audience and the second microphone 32 directed towards the receiver 14. Alternatively, if the audience recognition unit 12 is a portable unit or a unit of At the table, the first and second microphones 30 and 32 can be connected to the audience recognition unit 12 by lines or other communication means that allow the first microphone 30 to be mounted in close proximity to the audience and the second microphone 32 this mounted in close proximity to the receiver 14. Instead of canceling the noise, as mentioned above, the noise cancellation setting can use phase cancellation. It should be noted that the sound travels approximately 341.37 meters per second in the air. Sounds that do not originate equidistantly from the first and second microphones 30 and 32, which are preferably 17.06 centimeters apart, arrive at these microphones at different times and are canceled or at least partially canceled. Based on the actual frequencies of the sounds, and on the separation between the first and second microphones 30 and 32, different phase angles are produced. The noise canceling configuration 42 implementing phase cancellation uses these different phase angles to cancel the audio output of the receiver 14 and other background noise of the speech signals provided by the audience members. In this embodiment of the present invention, the first and second microphones 30 and 32 are preferably broadcast microphones mounted on the same side of the audience recognition unit 12. The audience recognition unit 1 2 may have any orientation with with respect to the audience and the receiver 14 while the first and second microphones 30 and 32 are not signaled directly to the receiver 14. In this embodiment of the present invention, the noise cancellation configuration 42 may be of the form shown in the Figure 4. In accordance with this modality, the noise cancellation configuration 42 includes an amplifier 60, which amplifies the signal of the first microphone 30, and an amplifier 62, which amplifies the signal of the second microphone 32. As the amplifiers 60 and 62 can be essentially the same, only one of the amplifiers is shown in detail in Figure 4. The outputs of the amplifiers 60 and 62 are summed by an adder 64 so that the background noise is substantially eliminated. The output of the adder 64 is detected and converted to a direct current level by a detector 66 and is provided to a timer 68. If the output of the adder 64 is above a predetermined level ^ the timer 68 is activated and provides an output to a switch r 70 for connecting the output of the adder 64 to the voice recognition unit 44. The voice recognition unit 44 provides a setting signal through a terminal 72 in order to set the ringer tempo 68 in its activated state until a predetermined time after the voice recognition unit 44 recognizes a voice for the last time. The speech recognition unit 44 can be an RSC164 speech recognition processor provided by Sensory I nc. The speech recognition unit 12 compares the output of the noise cancellation configuration 42 to previously stored reference speech signals (identifiers) stored in a memory. These previously stored reference speech signals (identifiers) can be spoken identification words or voice characteristics of a frequency spectrum of the person speaking. When similarities are found, the members of the audience are identified, and the voice recognition unit 44 provides an appropriate identification record to a controller 46 for each similarity. The controller 46 stores the identification register in a memory 48 for a subsequent transmission to the base unit 18 via a communications configuration 50. The communications configuration 50 may use the communication link 20 as described above. The controller 46 may also be coupled to a clock 52 so that times that the members of the audience are identified may also be stored in the memory 48. An indication unit 54 may also be coupled to the controller 46. The controller 46 controls the indicating unit 54 for indicating to the audience m em bers that provide their voice signals corresponding to the first and second microphones 30 and 32, or using a keyboard 56 to manually identify themselves. The indication unit 54 may be a speech synthesizer and / or a graphics or video display (such as an liquid crystal display or a light emitting diode) of the hearing recognition unit 12 arranged to provide audible indications or visible to the members of the audience. Alternatively, the indication unit 54 may be the horns and / or the display of the receiver 14 configured to provide audible and / or visible indications to the audience members in response to the interaction with the audience recognition unit 12. The unit of audience recognition 12 can be implemented as a portable unit 90 that is illustrated in Figure 5. The portable unit 90 includes a housing 92 that houses the elements of the audience recognition unit 12 shown in Figure 3. The housing 92 supports a light emitting diode array 94 that can perform the functions of the indication unit 54. However, as mentioned above, the indication unit 54 can alternatively be a speech synthesizer in which case the housing 92 will incorporate an exit horn to project audible voice prompts from the speech synthesizer to audience members. The display of light emitting diode 94 (and / or the speech synthesizer) can perform other functions besides indication. The housing 92 also houses an infrared transmitter 96 to establish the communications link 20 shown in Figure 2. Alternatively, the communication link 20 may be established by an ultrasonic or RF link, a hard cable, or the like. The housing 92 additionally supports a keyboard 98, which may be the keyboard 56 shown in Figure 3, to allow audience members to insert data. For example, some of the default keys on the keyboard 98 may be assigned to members and visitors of the audience so that they can identify themselves manually if they can not be recognized by their voice signals. The housing 92 may contain a rechargeable battery that is configured to provide power to the audience recognition unit 1 2 and that can be charged overnight or when the audience recognition unit 12 is not in use. The portable unit 90 also has a power on / off button 95 to turn the portable unit 90 on and off, and the light emitting diode 97 to indicate that the portable unit 90 is ready to identify the members and visitors of the portable unit 90. the audience. Alternatively, the audience recognition unit 1 2 can be implemented as a table unit 1 00 which is illustrated in Figure 6. The table unit 1 00 includes a housing 1 02 which gives the elements of the unity Audience Recognition 1 2 shown in Fig. 3. The housing 1 02 can support an LED em- uor diode display (not shown) to implement the functions of the indication unit 54. However, as mentioned above, the indication unit 54 can alternatively be a speech synthesizer, in which case the housing 102 will incorporate an output horn to project audible voice prompts from the speech synthesizer to the audience members. The display of light-emitting diode or the speech synthesizer can perform other functions besides indication. The housing 102 may also house an infrared or RF transmitter (not shown) to establish the communications link 20 shown in Figure 2. Alternatively, the communication link 20 may be established by an ultrasonic link, a hard cable, or Similary. The housing 102 supports a keyboard 104, which may be the keyboard 56 shown in Figure 3, to allow the members of the audience to insert data. For example, some predetermined keys on the keyboard 104 may be assigned to the members of the audience so that they can identify themselves manually if they can not be recognized by their voice signals. The table unit 100 may contain a rechargeable battery that is configured to provide power to the audience recognition unit 12 and that can be charged overnight or when not in use. Alternatively, the table unit 100 may have a power cable to provide power to the audience recognition unit 12 from an electrical connection. Also, the table unit 100 may have an energy indicator light 106 and a detected voice indicator light 108. During a training mode, the controller 46 of the audience recognition unit 12 may implement a program 120 shown in FIG. Figure 1 in the form of a flow diagram. The program 120 is executed during the training mode to store identification words or voice characteristics associated with the members of the audience. During training, a member of the audience must assume a position at a specific distance from the hearing recognition unit 12. This hearing member initiates the training by pressing a button or appropriate key on the keyboard 56 of the hearing recognition unit. audience 12. When the training mode for executing the program 120 is started, the audience recognition unit 12, in a block 122, determines whether a predetermined amount of time has elapsed since the last activity of the audience member. If the hearing recognition unit 12 determines that there has been no activity of the hearing member during the predetermined amount of time, program 120 assumes that the training mode has been terminated and program 120 ends. If the predetermined amount of time has not passed since the last activity of the audience member, the audience recognition unit 12, in a block 124, continues to wait for an activation command to be said. The activation command must be a single word or a combination of words not normally present in the audio of the receiver 14. The audience member must pronounce the activation command in a normal voice to the first and second microphones 30 and 32. When the audience recognition unit 12 detects the activation command, the audience recognition unit 12, in a block 126, instructs the audience member to say the identification word that the viewer has selected to identify himself. This word of identification can be the name of the viewer. The audience recognition unit 12 then determines in a block 128 whether the audience member has pronounced the identifier. If not, the audience recognition unit 12, in block 126, again instructs the member of the audience to say the word of identification. If the audience recognition unit 12 determines in block 128 that the hearing member has uttered the identification word, the audience recognition unit 12, in a block 130, appropriately processes the identification word and saves it in the memory 48. The flow returns to block 122 where the training mode is terminated or another member of the audience inserts a corresponding identification word. During an identification mode, the controller 46 of the audience recognition unit 12 can implement a program 140 shown in Figure 8 by a flow chart. Program 140 runs during the identification mode to identify the members of the audience. The identification mode can be initiated by pressing an appropriate button or key on the keypad 46 of the audience recognition unit 12. When the identification mode for executing the program 140, the audience recognition unit 12, is initiated in a block 142, determines whether a member of the audience wishes to terminate the identification mode by pressing a button or appropriate key on the keyboard 56 of the audience recognition unit 12. If the audience recognition unit 12 determines that the audience member you want to end the identification mode, program 140 ends. If the audience recognition unit 12 determines that the member of the audience does not wish to terminate the identification mode, the audience recognition unit 12, in a block 144, waits for an activation command to be issued. The member of the audience should pronounce the activation command in a normal voice to the first and second microphones 30 and 32. When the audience recognition unit 12 detects the activation command, the audience recognition unit 12, in a biochem 146 , tells the member of the audience to say his word of identification. The audience recognition unit 12 then determines in a block 148 whether the audience member has uttered the word of identification. If not, the audience recognition unit 12, in block 146, continues to direct the member of the audience to pronounce the word of identification. If the audience recognition unit 12 determines in block 148 that the hearing member has uttered the word of identification, the hearing recognition unit 12, in a block 150, instructs the speech recognition unit 44 to compare the identification word with the vocabulary of the identification words stored in the memory 48. If the audience recognition unit 12 determines in a block 152 that the speech recognition unit 44 detects a similarity between the pronounced identification word and a of the identification words stored in the memory 48, the audience recognition unit 12, in a block 154, stores the pronounced identification word and the similarity time as an identification record in the memory 48 for immediate or subsequent communication to the base unit 18. (Alternatively, the audience recognition unit 12, in block 154, can communicate the pronounced identification word and the time directly to the base unit 18). If the audience recognition unit 12 determines in block 152 that the speech recognition unit can not find a similarity of the identification word pronounced with one of the identification words stored in memory 48, the audience recognition unit 12, in a block 156, instructs the member of the audience to try again or use another form of identification. For example, indicator 1 40 may instruct the member of the audience to pronounce his or her identification word a predetermined number of times before telling the member of the audience to use another form of identification. (Such as inserting the identification word by using the keyboard 56). Alternatively, or additionally, if the controller 46 determines that the speech recognition unit 44 can not detect a similarity in the block 152, the controller 46 in the block 156 may send a message to the base unit 18 indicating that the identification should be made from the passive-person meter 26. After the audience recognition unit 12, in block 1 54, stores the pronounced identification word and the similarity time as an identification record in memory 48, or after the audience recognition unit 12, in block 1 56, instructs the audience member to try again, or use another form of identification, or send an appropriate message to the base unit 1 8 to search for a meter identification. passive people 26, or similar, the flow returns to block 1 42. The advantage of using an activation command during normal speech recognition is that the comparison process is simplified ifica. That is, during normal hearing recognition, each word of identification pronounced is compared to each reference identification word stored in the memory 48. However, the activation command only has to be compared with a reference in the memory 48. Block 126 of Figure 7 and block 146 of Figure 8 can be configured to also instruct each member of the audience to pronounce a track in addition to the member identification word. The advantage of coupling a hint to a word of identification is to minimize the possibility that the audience recognition unit 12 may incorrectly identify a person as a member of the audience when the identifying word of that person is contained in the output of audio from receiver 14, but that person is not in the audience. For example, the word of identification may be a name, and the word may be a common word for all members of the audience, such as the word "return," or the word "exit," or the like. Therefore, a member of the audience when he enters (or leaves) the monitored area, pronounces his name, as the identification word coupled with the track. For example, the member may say "John return" or "Mary return". Alternatively, the track may be unique to each member of the audience. Any configuration reduces the possibility that a word emitted audibly by the receiver 14 is mistakenly considered as an identification word uttered by a member of the audience. Certain modifications of the present invention have been described above. Other modifications will occur to those who practice the technique of the present invention. For example, the invention has been described above in relation to measuring television program audiences. Nevertheless, the present invention is also useful in measuring the audiences of radio programs and other media programs. Also, the program sensor 16 is described above by detecting the program and / or channel to which the receiver 14 is tuned. Instead, the audience recognition unit 12 may be configured to accept manual entry of program identifications by members of the audience through the use of the keyboard 56. Additionally, the functions of a remote control can be incorporated into the audience recognition unit 12. Additionally, instead of relying on two microphones, such as the first and second microphones 30 and 32, the audience recognition unit 12 can be based on a single near-range directional microphone. If so, noise cancellation may not be necessary. In addition, the use of an activation command can be eliminated. Also, as described above, the passive meter 26 may be of the type that passively identifies the members of the audience in a monitored area in accordance with the teachings of the United States Patent Serial Number 5,550,928 , described above. However, the passive measurer 26 can be any type of passive device that is able to passively identify the members of an audience. For example, the passive meter 26 can use an ultrasonic and / or infrared transducer or a configuration of these transducers to capture an image of the audience area and can use processing algorithms to recognize the members in the audience. Additionally, the passive measurer 26, as mentioned above, passively identifies the members of the audience in a monitored area. If the passive measurer 26 can not passively identify the members in the hearing, it may be configured to count the members of the audience and to indicate a manual identification of the members until the number of manually identified members is equal to the account. Alternatively, if the passive measurer 26 can not passively identify each member at the hearing, it may be configured to count the members of the audience and to indicate a manual identification of these members that they can not passively identify. As a further alternative, a counter may be used in place of the passive meter 26. The counter may comprise an ultrasonic and / or infrared transducer or a configuration of these transducers or other device or devices that can detect people in the audience and which can count the number of people detected. Then, the audience measurement system 10 could be configured to indicate a manual identification until the number of people identified manually is equal to the account. Additionally, the audience measurement unit 12, the program sensor 16, and the passive person meter 26 are shown in Figure 1 as separate elements. Instead, the audience measurement unit 12 may include the program sensor 16 and / or the passive meter 26. Accordingly, the description of the present invention should be construed as illustrative only and is for the purpose of teach those skilled in the art the best way to carry out the invention. The details may be varied substantially without departing from the spirit of the invention, and exclusive use of all modifications within the scope of the appended claims is reserved.

Claims (68)

1. CLAIMS 1. A system (10) for identifying a member of the audience of a receiver (14) having a first microphone (30) and a voice recognition unit (44), wherein the first microphone (30) picks up a speech signal of the audience member, and wherein the speech recognition unit (44) is configured to identify the member of the voice signal audience, the system is characterized in that: the first microphone (30) is configured to capture relatively more than the voice signal of the audience member; a second microphone (32) is configured to capture relatively more than one sound signal from the receiver (14); and a noise canceling unit (42) is coupled to the first and second microphones (30 and 32), wherein the noise cancellation unit (42) is configured to substantially eliminate the sound signal of the speech signal with base in a phase relationship between the sound signal and the voice signal to identify the audience member of the voice signal. The system of claim 1, further comprising a program sensor (16) configured to identify a program received by the receiver (14). The system of claim 2, wherein the program sensor (16) comprises a channel sensor. The system of claim 2, wherein the program sensor (16) comprises a code reader configured to detect a program identification code in a program to which the receiver is tuned. The system of claim 2m wherein the program sensor (1 6) comprises an identification extractor configured to extract unique identifiable identifications from a program to which the receiver is tuned. The system of claim 2, wherein the program sensor (16) comprises a keyboard configured to allow the audience member to insert the identity of a program or channel. 7. The system of claim 1, further comprising a clock (52) and a memory (48), wherein the memory (48) stores an identity of the audience member based on the voice signal of the member, and wherein the memory (48) maintains a clock time (52) when the member is identified. The system of claim 1, wherein the receiver (14) is a television set. The system of claim 1, wherein the receiver (14) is a radius. The system of claim 1, wherein the receiver (14) is a multimedia reception device. eleven . The system of claim 1, further comprising a keypad (56) configured to accept an identification input from the audience member. The system of claim 1, further comprising an indicating device (54) configured to instruct the audience member to pronounce the voice signal. The system of claim 12, further comprising a sensor (16) for identifying a program or channel received by the receiver. The system of claim 12, further comprising a passive meter (26) configured to passively identify the audience member. The system of claim 14, further comprising a sensor (16) for identifying a program or channel received by the receiver. The system of claim 1, wherein the voice signal comprises an identifier of the audience member. The system of claim 1, wherein the voice signal comprises an identifier and a track of the audience member. The system of claim 1, further comprising a passive pass meter (26) configured to passively identify the audience member. 19. The system of claim 1, further comprising a communication unit (50) configured to communicate the identity of the member and / or identity of the program to a remote point. 20. The system of claim 19, wherein the communications unit (50) is configured to communicate the identity of the member to the remote point by infrared signals. The system of claim 1, wherein the first and second microphones (30 and 32), the noise cancellation unit (42) and the speech recognition unit (44) are housed in a portable housing (92) . The system of claim 1, wherein the first and second microphones (30 and 32), the noise canceling unit (42) and the speech recognition unit (44) are housed in a tabletop housing (100). ). The system of claim 1, wherein the first and second microphones (30 and 32) are directionally oriented in substantially opposite directions of acquisition. The system of claim 1, wherein the first microphone (30) is mounted closer to the audience than the second microphone (32) is, and wherein the second microphone (32) is mounted closer to the receiver than what is the first microphone (30). The system of claim 1, further comprising a counter, a display, and a manual recording device, wherein the speech recognition unit is configured to provide indications to the display to indicate manual identification of the members of the hearing through the use of the manual recording device until the identified members are equal to the count provided by the meter. 26. The system of claim 1, wherein the phase relationship is based on the speed of sound. The system of claim 26, wherein the first and second microphones are separated by a distance of substantially 17.06 cm. 28. A system (10) for identifying a member of the audience of a program to which a receiver (14) is tuned has a microphone (30) and a speech recognition unit (44), the microphone (30) is placed to receive a voice signal from the member, and the speech recognition unit (44) is configured to identify the member of the voice signal, the system is characterized in that: the speech recognition unit (44) is configured to identifying the member of the voice signal that includes an identifier and a track for the member. The system of claim 28, further comprising a memory (48) configured to store reference identification information corresponding to the member, and wherein the speech recognition unit (44) is configured to identify the member based on a comparison of the speech signal and the reference identification information. 30. The system of claim 29, wherein the function is an identification code inserted in the program. 31 The system of claim 29, wherein the function is a unique identifiable identification extracted from the program. 32. The system of claim 28, wherein the receiver (14) is a television. 33. The system of claim 28, wherein the receiver (14) is a radius. 34. The system of claim 28, wherein the receiver (14) is a multimedia reception device. 35. The system of claim 28, further comprising a keypad (56) configured to accept a member identification entry, 36. The system of claim 35, wherein the identification entry is an identification of the member 37. The system of claim 35, wherein the identification entry is a channel or program identification. 38. The system of claim 28 further comprising an indicating device (54) configured to indicate to the member to pronounce the voice signal. 39. The system of claim 38, further comprising a passive meter (26) configured to passively identify the member. 40. The system of claim 28, wherein the voice signal comprises a member identifier. 41. The system of claim 28, further comprising a passive meter (26) configured to passively identify the member. 42. The system of claim 28, further comprising a communications unit (50) configured to communicate the identity of the member to a remote point. 43. The system of claim 42, wherein the communication unit (50) is configured to communicate the identity of the member to the remote point by infrared signals. 44. The system of claim 28, wherein the microphone (30) and the speech recognition unit (44) are housed in a portable housing (92). 45. The system of claim 28, wherein the microphone and speech recognition unit are housed in a table housing (102). 46. The system of claim 28, further comprising a counter, a display, and a manual recording device, wherein the voice recognition unit is configured to provide indications to the display to indicate the manual identification of the members of the hearing through the use of the manual recording device until the identified members are equal to the count provided by the meter. 47. A system (10) for identifying a member of the audience of a program to which a receiver (14) having a microphone (30) is positioned to receive a voice signal from the member and a unit of speech is tuned. voice recognition (44) configured to provide a first identification of the member identifying the member of the voice signal, the system is characterized in that: a passive person meter (26) is configured to provide a second identification of the member identifying passively to the broader; and, a controller (46) is configured to select the first identification or the second identification to identify the member. 48. The system of claim 47, wherein the controller (46) includes an indication unit (54) configured to indicate to the member speaking the voice signal. 49. The system of claim 48, further comprising a sensor (16) configured to identify the program or channel. 50. The system of claim 47, wherein the controller (46) includes an indication unit (54) configured to instruct the member to pronounce the voice signal if the passive meter (26) is not able to provide the first identification. 51 The system of claim 47, wherein the controller (46) is configured to select the first identification if the speech recognition unit (44) is not able to provide the second identification. 52. The system of claim 47, wherein the voice signal has an identifier. 53. The rejection system 47, wherein the voice signal comprises an identifier and a track. 54. The system of claim 47, further comprising a sensor (16) configured to identify the program or channel. 55. The system of claim 47, further comprising a communications unit (50) configured to communicate the identity of the member to a remote point. 56. The system of claim 55, the communication unit (50) is configured to communicate the identity of the member to the remote point by infrared signals. 57. The system of claim 47, wherein the microphone (30) and the speech recognition unit (44) are housed in a portable housing (92). 58. The system of claim 47, wherein the microphone and speech recognition unit are housed in a table housing (102). 59. The system of claim 47, further comprising a counter, a display, and a manual recording device, wherein the controller is configured to provide indications to the display to indicate the manual identification of the members of the audience through the use of the manual recording device until the identified members are equal to the count provided by the meter. 60. An audience recognition unit (90 or 100) that identifies a member of the audience of a program to which a receiver having a microphone (30) and a voice recognition unit (44) is tuned. given to recognize the member of an identifier pronounced by the member, the audience recognition unit is characterized by: a housing (92 or 102) supports the microphone (30) and the speech recognition unit (44); and, a communications unit (50) is supported by the housing (92 or 102), wherein the communications unit (50) is configured to communicate data related to the recognition of the member via air to a local base unit (18) 61 The audience recognition unit of claim 60, further comprising an indication unit (54) supported by the housing (92 or 102), wherein the indication unit (54) is configured to instruct the member to perform an action . 62. The audience recognition unit of claim 61, further comprising a keyboard (56) supported by the housing (92 or 102), wherein the keyboard (56) is configured to allow the member to insert data. 63. The audience recognition unit of claim 60, further comprising a keyboard (56) supported by the housing (92 or 102), wherein the keyboard (56) is configured to allow the member to insert data. 64. The hearing recognition unit of claim 60, further comprising a clock (52) and a memory (48) supported by the housing (92 or 102), wherein the memory (48) stores an identity of the member based on the pronoun identifier, and eh the memory (48) stores one hour of the clock (52) when the member is identified. 65. The audience recognition unit of claim 64, wherein the identifier includes a word and a track. 66. The audience recognition unit of claim 60, wherein the identifier includes a word and a track. 67. The audience recognition unit of claim 60, wherein the audience recognition unit is a portable unit (90). 68. The audience recognition unit of claim 60, wherein the audience recognition unit is a table unit (100). RESU MEN A member of an audience of a program to which a receiver is tuned is identified by a system that includes first and second microphones, a noise cancellation unit, and a voice recognition unit. one of the first and second microphones captures relatively more than one voice signal from the member of the audience and the other from the first and second microphones captures relatively more than one sound signal from the receiver. The noise canceling unit substantially eliminates the sound signal from the speech signal. The voice recognition unit identifies the member of the voice signal of the member. The first and second microphones, the noise cancellation unit, and the speech recognition unit may be housed in a portable housing or in a table housing. The system may also include a passive person meter to passively identify the member, and a program to identify the program or channel to which the receiver is tuned.