JP5244627B2 - Emotion estimation method and apparatus - Google Patents

Description

  The present invention relates to an emotion estimation method and apparatus for estimating a subject's emotion based on the biological information, and in particular, allows the subject to use a communication service when the correspondence relationship between the biological information and the emotion information is created in advance as a database. The present invention relates to an emotion estimation method and apparatus for actively changing emotions by applying external stress.

Patent Document 1 discloses a technique for recognizing emotions by attaching a detachable sensor to a subject's body and acquiring the subject's biological information as a digital signal.
JP 2002-112969 A

  In the above prior art, the emotion of the subject is estimated by comparing the obtained biological information with a preset reference value. However, since the reference value is set based on information obtained from a large number of subjects, individual differences among the subjects cannot be reflected in the estimation result, and there is a possibility that an erroneous determination is made.

  Further, in a system that estimates a subject's emotion based on biological information, the correspondence between the biological information and emotion information needs to be stored in a database in advance. In creating this database, it is desirable to continuously change the subject's emotions from a calm state to an excited state and sample biological information in various emotions. However, it has been difficult to change the subject's emotion during sampling from a calm state to an excited state over a wide range without omission.

  The object of the present invention is to solve the above-mentioned problems of the prior art and to change the stress of the subject when the correspondence relationship between the biological information and the emotion information of the subject is created in advance as a database, and the individual differences of the subjects It is an object to provide an emotion estimation method and apparatus that enables accurate emotion estimation by reflecting the above in the estimation result.

  In order to achieve the above object, the present invention is characterized in that the following means is taken in an emotion estimation apparatus that estimates a subject's emotion based on biological information.

  (1) means for providing a communication service to a subject, biological measurement means for measuring biological information of a subject using the communication service, and emotion measurement means for measuring emotion information representative of the emotion of the subject using the communication service Storage means for associating and storing biological information and emotion information, and emotion estimation means for extracting emotion information associated with the biological information measured by the biological measurement means from the storage means and estimating the subject's emotion It is characterized by having.

  (2) The biometric information and emotion information are stored in association with each subject, and the emotion estimation means extracts the emotion information associated with the subject's biological information from the storage means and estimates the subject's emotion And

  (3) It is characterized by comprising communication quality designating means for designating the quality of the communication service used by the subject and communication quality control means for controlling the quality of communication used by the subject to the designated quality.

  According to the present invention, the following effects are achieved.

  (1) When a biological database and emotion information of a subject are measured and a database that links the two is constructed in advance, external stress can be applied by allowing the subject to use a communication service, so that the living body in various emotions Information can be easily obtained.

  (2) Since the correspondence between the biological information and emotion information is recorded in the database for each subject and the emotion is estimated, accurate emotion estimation reflecting individual differences of each subject can be performed.

  (3) When measuring a subject's biological information and emotion information and constructing a database that links the two in advance, the external stress applied to the subject can be reduced by forcibly degrading the quality of the communication service used by the subject. Since it can be changed quantitatively, it becomes possible to easily obtain biological information in various emotions.

  Hereinafter, the best embodiment of the present invention will be described in detail with reference to the drawings. FIG. 1 is a functional block diagram showing the configuration of the main part of the emotion estimation apparatus according to the present invention. In this embodiment, the test subject terminal 5 operated by the test subject 2 and the Web server 3 or the test subject terminal 5 to be accessed when the test subject terminal 5 functions as a client function as a call terminal such as a mobile phone or a soft phone. At this time, an access point (AP) 4 of a call terminal 6 that is a partner terminal is connected via a network (NW) 7.

  An emotion estimation apparatus 1 according to the present invention is connected between the NW 7 and the subject terminal 5. The emotion estimation device 1 is configured to apply a communication quality control unit 1a that controls communication quality between the subject terminal 5 and the Web server 3 or the call terminal 6 and to apply external stress to the subject 2 by degrading the communication quality. And the emotion estimation unit 1b for measuring the biological information and emotion information of the subject 2 in advance and constructing a database in advance, and then estimating the emotion of the subject 2 based only on the biological information of the subject 2.

  FIG. 2 is a functional block diagram showing the configuration of the main part of the emotion estimation unit 1b. Here, the configuration unnecessary for the description of the present invention is omitted.

  The emotion measuring unit 11 measures the emotion information of the subject 2. In the present embodiment, voice is measured as information representing the emotion of the subject 2. The behavior-dependent emotion estimation unit 12 estimates the emotion of the subject 2 by applying the speech signal of the subject 2 to appropriate speech psychoanalysis.

  A technique for estimating a psychological state by analyzing speech is well known. For example, Alpha Omega Soft's speech psychological analysis software “Trust Pro” (http://pr.fujitsu.com/jp/news/2001/02/ 15-3.html) and “Truster” (http: //www/syatyu.com/shop/143/top.htm) of Trust Tech are known.

  The biological measurement unit 13 measures biological information of the subject. In this embodiment, a series of pulse interval data (RR interval) is measured from the pulse wave data of the subject 2 as biological information, and a power spectrum component (0.04 to 0.15 Hz) obtained by frequency analysis of the RR interval ( The ratio between the integrated value of the low frequency component (LF) and the integrated value of the power spectrum component (high frequency component HF) of 0.15 to 0.40 Hz, that is, the LF / HF value that is the relative activity ratio of the sympathetic / parasympathetic nerve Is measured.

  The communication quality designation unit 15 outputs an instruction to degrade the communication quality to the communication quality control unit 1a in order to apply external stress to the subject 2 to forcibly change the emotion of the subject 2. In the database (DB) 14, the estimation result by the behavior-dependent emotion estimation unit 12 and the measurement result (LF / HF value) by the living body measurement unit 13 are linked and recorded. The living body dependent emotion estimation unit 16 searches the DB 14 based on the measurement result by the living body measurement unit 13, estimates the emotion level associated with the measurement result as the current emotion of the subject 2, and outputs it.

  Next, the operation of this embodiment will be described in detail with reference to a flowchart. FIG. 3 is a flowchart showing the operation of an embodiment of the present invention, in which the subject terminal 5 is a personal computer or mobile phone in which a softphone application is installed, and the subject 2 is the subject terminal 5 and the partner terminal. An example of emotion estimation when talking to 6 will be described.

  In step S1, “DB construction processing” that records the biometric information and emotion information of the subject 2 in the DB 14 is recorded, and the emotion of the subject 2 is estimated by applying the measurement result of the biometric information to the DB 14 that has already been constructed. One of the “estimation processes” is selected. In the present embodiment, since it is necessary to construct the DB 14 prior to emotion estimation, “DB construction processing” is selected here, and the process proceeds to step S2.

  In step S2, the communication quality designating unit 15 outputs a signal designating the communication quality level Q to the communication quality control unit 1a. In the present embodiment, five levels from the highest level “1” to the lowest level “5” can be set with respect to the communication quality level Q, and the highest level “1” is designated here.

  In step S3, when a communication session is established between the subject terminal 5 and the partner terminal 6 and a call is started, in step S4, emotion measurement by the emotion measurement unit 11 and emotion estimation by the behavior-dependent emotion estimation unit 12 are performed. Is started, and the current emotional level of the subject is calculated. In step S5, the biological measurement by the biological measurement unit 13 is started, and the current LF / HF value of the subject is measured.

  In step S6, the emotion estimation result and the biological information measurement result are linked to each other, and are recorded in the DB 14 with the subject identifier (ID) as an index, as shown in FIG. That is, in this embodiment, the correspondence between the emotion estimation result and the biological information measurement result is recorded for each subject. The emotion estimation, biometric measurement, and recording thereof are continued until it is determined in step S7 that the predetermined sampling period has been completed.

  When the sampling period is completed, the process proceeds to step S8, and the emotion estimation, biological measurement, and recording thereof are interrupted. In step S9, it is determined whether or not the current communication quality level is “5”. Since the initial level is other than “5”, the process proceeds to step S10. In step S10, the communication quality level is lowered by one level from the current level, and then the process returns to step S4 and the above processing is repeated. That is, the correspondence relationship between the emotion estimation result and the biological information measurement result is recorded in the DB 14 while the communication quality is deteriorated. When emotion estimation, biometric measurement, and recording are completed at all communication quality levels and the construction of the DB 14 is completed, the process ends.

  Thereafter, when “Emotion estimation” is selected in step S1, the process proceeds to step S11. If the start of communication is detected in step S11, it will progress to step S12 and the biological information of the test subject 2 will be measured. In step S13, the DB 14 is referenced using the measurement result of the subject ID and the biological information as a search key, and the emotion level associated with the measurement result of the biological information is extracted. In step S14, the current emotion estimation result of the subject 2 is output based on the extracted emotion level. In the present embodiment, for example, when there are a plurality of entries having the same measurement result (LF / HF value) of the biological information, the emotion of the subject 2 is estimated based on the average value of the emotion level of the entry. In step S15, the emotion estimation result is output.

  In the above embodiment, the subject terminal 5 is a soft phone or a mobile phone, and when the subject terminal 5 and the partner terminal 6 are talking, the emotion of the subject 2 is estimated based on the speech voice. The present invention is not limited to this.

  For example, if the subject terminal 5 estimates an emotion when accessing the Web server 3 as a client, the pressing strength and pressing speed when the subject 2 operates the keyboard of the subject terminal 5 are measured. As shown, it may be estimated that the stronger the pressing intensity or the higher the pressing speed, the higher the emotion of the subject 2 is.

  Moreover, the packet exchanged between the subject terminal 5 and the Web server 3 may be monitored, and the subject's emotion may be estimated based on the frequency with which the forced termination operation by the subject 2 is detected. That is, the frequency at which the TCP RST message is detected during communication using HTTP and the frequency at which the ABOR command is detected during communication using FTP are monitored. As shown in FIG. You may make it estimate that the test subject's 2 emotion is high.

  Furthermore, in the case of real-time communication such as VoIP, the duration from session establishment to release is measured, and as shown in an example in FIG. 7, the shorter the session duration is than the reference time tref, You may make it estimate that the test subject's 2 emotion is high. Further, the facial expression of the subject 2 may be photographed with a camera (not shown), and the emotion of the subject may be estimated by applying this to an appropriate facial expression psychoanalysis.

  Furthermore, in the above embodiment, the biological information is described as adopting the relative activity ratio (LF / HF value) of the sympathetic / parasympathetic nerve obtained from the pulse interval data (RR interval). The body temperature, blood pressure, apex heartbeat, myoelectric potential, nystagmus mode, nystagmus frequency, cardiac / intravascular pressure, respiratory flow rate, respiratory pressure, and ventilation volume may be employed.

It is the functional block diagram which showed the structure of the principal part of the emotion estimation apparatus which concerns on this invention. It is the functional block diagram which showed the structure of the principal part of the emotion estimation part 1b. It is the flowchart which showed operation | movement of one Embodiment of this invention. It is the figure which expressed the contents of the database typically. It is the figure which showed the relationship between the operation of the keyboard by a test subject, and feelings. It is the figure which showed the relationship between the occurrence frequency of a TCP RST message and ABOR command, and a test subject's emotion. It is the figure which showed the relationship between the duration of a session, and a test subject's emotion.

  DESCRIPTION OF SYMBOLS 1 ... Emotion estimation apparatus, 1a ... Communication quality control part, 1b ... Emotion estimation part, 2 ... Test subject, 3 ... Web server, 4 ... Access point (AP), 5 ... Test subject terminal, 6 ... Call terminal, 7 ... Network ( NW), 11 ... emotion measurement unit, 12 ... action-dependent emotion estimation unit, 13 ... living body measurement unit, 14 ... database (DB), 15 ... communication quality instruction unit, 16 ... biological dependency emotion estimation unit

Claims (15)

In an emotion estimation device that estimates a subject's emotion based on biological information,
Means for providing a communication service to the subject;
Communication quality specifying means for specifying the quality of the communication service;
Communication quality control means for controlling the quality of the communication service to the specified quality ;
Biological measurement means for measuring biological information of a subject using the quality-controlled communication service;
An emotion measuring means for measuring emotion information representative of the emotion of the subject using the communication service;
Storage means for storing the biological information and emotion information in association with each other;
An emotion estimation apparatus comprising: emotion estimation means for extracting emotion information associated with biological information measured by the biological measurement means from the storage means to estimate a subject's emotion. The biological information and emotion information are stored in association with each subject,
The emotion estimation apparatus according to claim 1, wherein the emotion estimation unit extracts emotion information associated with the biological information of the subject from the storage unit and estimates the subject's emotion.   The emotion estimation apparatus according to claim 1, wherein the biological measurement unit measures a relative activity ratio of sympathetic / parasympathetic nerves based on fluctuations in the heartbeat of the subject.   The emotion estimation apparatus according to any one of claims 1 to 3, wherein the emotion measurement unit measures emotion information based on a voice of a subject.   The emotion estimation apparatus according to claim 1, wherein the emotion measurement unit measures emotion information based on an input operation on a communication service being used by a subject.   The emotion estimation apparatus according to any one of claims 1 to 3, wherein the emotion measurement unit measures emotion information based on a frequency at which a communication service being used by the subject is forcibly terminated by the subject. The emotion estimation device according to claim 6 , wherein the emotion measurement unit represents the frequency at which the communication is forcibly terminated by the frequency at which the TCP RST message is transmitted and forcibly terminated during communication by HTTP. . The emotion estimation device according to claim 6 , wherein the emotion measurement unit represents the frequency at which the communication is forcibly terminated by the frequency at which the ABOR command is transmitted and forcibly terminated during FTP communication. The emotion estimation apparatus according to claim 1 , wherein the emotion measurement unit measures emotion information based on a duration of real-time communication. The emotion estimation apparatus according to claim 1 , wherein the emotion measurement unit measures emotion information based on a video signal in which a facial expression of a user is copied. In an emotion estimation device that estimates a subject's emotion based on biological information,
  Means for providing a communication service to the subject;
  A biological measuring means for measuring biological information of a subject using the communication service;
  An emotion measuring means for measuring emotion information representative of the emotion of the subject using the communication service;
  Storage means for storing the biological information and emotion information in association with each other;
  Emotion estimation means for extracting emotion information associated with the biological information measured by the biological measurement means from the storage means and estimating the emotion of the subject,
The emotion estimation device is characterized by measuring emotion information based on a frequency at which a communication service being used by a subject is forcibly terminated by the subject.   12. The emotion estimation device according to claim 11, wherein the emotion measurement unit represents the frequency at which the communication is forcibly terminated by the frequency at which the TCP RST message is transmitted and forcibly terminated during communication by HTTP. .   12. The emotion estimation apparatus according to claim 11, wherein the emotion measuring unit represents the frequency at which the communication is forcibly terminated by the frequency at which the ABOR command is transmitted and forcibly terminated during FTP communication. A communication quality specifying means for specifying the quality of the communication service used by the subject;
14. The emotion estimation apparatus according to claim 11 , further comprising communication quality control means for controlling the quality of communication used by the subject to the designated quality. In an emotion estimation method for estimating a subject's emotion based on biological information,
Providing a communication service to the subject;
Specifying the quality of the communication service;
A procedure for controlling the quality of the communication service to the designated quality ;
Measuring biological information of a subject using the quality-controlled communication service;
A procedure for measuring emotion information representative of the emotion of a subject using a communication service;
A procedure for associating and storing the biological information and emotion information;
A procedure for measuring the biological information of the subject;
A procedure for extracting emotion information associated with the measurement result of the biological information;
A method for estimating an emotion of a subject based on the extracted emotion information.

Images