JP2019174768A - Information processing apparatus - Google Patents

Abstract

To objectively present an evaluation result by evaluating a speech made by a person.SOLUTION: An information processing apparatus 1 as an image forming apparatus includes: an utterance time detecting unit 101 for detecting for each person, an utterance time period for each utterance from voice data of a plurality of persons; an interjection/opinion determination part 102 for determining that the detected utterance of each person is an interjection, or an opinion depending respectively on the utterance time period thereof being within a first time period, or a second time period, which is longer than the first time period; an HDD 114 for storing, for each person, a result that the utterance is determined to be an interjection or an opinion; an evaluation unit 103 to give, for each result of each person stored in the HDD 114, a first evaluation point to the interjection, and a second evaluation point to the opinion, wherein the second evaluation point is different from the first evaluation point; and a collection part 104 for tallying and graphing the given first and second evaluation points for each person, and determining an evaluation level of each person on the basis of the result of the tabulation. The image forming apparatus prints the graphed image data.SELECTED DRAWING: Figure 9

Description

  The present invention relates to an information processing apparatus, and more particularly to a technique for evaluating a person's speech.

  Today, a technique has been proposed in which the activation of discussion is estimated for each group and the situation is clearly indicated (see Patent Document 1 below). This technology memorizes the utterance time and utterance time length of each group member, classifies each member's utterance time and utterance time length in time series, and utterance time length of all group members The ratio of the utterance time length of each member to the above is calculated, and a graph depicting the utterance density contribution rate of each member is generated for each group.

JP 2006-162339 A

  The above technology provides each member's utterance density contribution rate to determine whether the group discussion is active, but this utterance density contribution rate is the utterance time length of each member relative to the utterance time length of all members of the group. It is only a ratio of Depending on this technology, it is not possible to evaluate the content that each member understands the speech time detection section.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to evaluate an utterance by a person and objectively present the evaluation result.

  An information processing apparatus according to one aspect of the present invention extracts and extracts each person's speech from a communication interface that performs data communication with the outside and voice data of a plurality of persons acquired through the communication interface. A speech time detection unit that individually detects the speech time for each speech, and whether the speech time of each person detected by the speech time detection unit is within a predetermined first time Determine whether or not the speech within the first time is a conflict, determine the speech made during a predetermined second time longer than the first time as an opinion, and for each person A result storage for storing the result of whether the remark is a reconciliation or an opinion determined by the reconciliation / opinion determination unit and the individual, respectively, Part and said For each of the results stored in the fruit storage unit, a predetermined first evaluation point is assigned to the conflict, and the opinion is predetermined different from the first evaluation point. The evaluation unit for assigning the second evaluation points, and the first and second evaluation points assigned by the evaluation unit are tabulated for each person, and the tabulation results for each person are graphed. A totaling unit for creating image data, and when the evaluation unit detects a pre-determined important keyword in the opinion converted into text by the conflict / opinion determination unit, for the opinion In addition, when the second evaluation point is added and given by a predetermined addition value and a predetermined unnecessary keyword is detected, the second evaluation point is predetermined for the opinion. The subtracted value is subtracted and added, and the evaluation unit adds the assigned first evaluation points to each of the persons, and the point indicated by the calculation is a predetermined first evaluation value. If it is above, the person's evaluation level is determined to be the first predetermined high level, the assigned second evaluation points are totaled for each person, and the points indicated by the totalization Is equal to or higher than a predetermined first opinion evaluation value, the person's evaluation level is determined to be a predetermined second high level.

  According to the present invention, it is possible to evaluate a person's utterance and objectively present the evaluation result.

It is a figure which shows the information processing apparatus which concerns on one Embodiment of this invention, and the subject person whom speech analysis is performed by the said information processing apparatus. It is a block diagram which shows the outline of an internal structure of information processing apparatus. It is a figure which shows an example of audio | voice data. It is a flowchart which shows 1st Embodiment of the speech analysis process by information processing apparatus. It is an image figure which shows each person's each utterance by changing the shape according to the utterance time. It is a flowchart which shows 2nd Embodiment of the speech analysis process by information processing apparatus. It is a flowchart which shows 3rd Embodiment of the speech analysis process by information processing apparatus. It is a flowchart which shows the determination process which determines different 1st time and said 2nd time for each person. It is a block diagram which shows the outline of further embodiment about the internal structure of information processing apparatus. It is a flowchart which shows 1st Embodiment of the evaluation process with respect to each person's utterance by information processing apparatus. It is a figure which shows an example of the image which graphed the total result which totaled the 1st and 2nd evaluation point for every person. It is a flowchart which shows the process which changes and assign | provides the 2nd evaluation point with respect to an opinion according to the presence or absence of the conflict with respect to the said opinion. It is a flowchart which shows the process which changes and assign | provides the 2nd evaluation point with respect to an opinion according to the keyword contained in the said opinion. It is a flowchart which shows 1st Embodiment of the evaluation level determination process which evaluates about each person. It is a flowchart which shows 2nd Embodiment of the evaluation level determination process which evaluates about each person. It is a flowchart which shows 3rd Embodiment of the evaluation level determination process which evaluates about each person. It is a figure which shows the example of the image which made the graph the total value of the said evaluation point produced by the evaluation part.

  Hereinafter, an information processing apparatus such as an image forming apparatus according to an embodiment of the present invention, an utterance analysis program, an evaluation program, an utterance analysis method, and an evaluation method will be described with reference to the drawings. FIG. 1 is a diagram illustrating an information processing apparatus according to an embodiment of the present invention and a target person whose speech analysis is performed by the information processing apparatus.

  The information processing apparatus 1 acquires, by data, voices uttered by each person in a conversation group composed of a plurality of people. For example, a conversation group consisting of a plurality of people P1, P2, and P3 (in the present embodiment, described as three people, but is not limited to this) is a meeting, class, conference, etc. (hereinafter simply referred to as a conference). It is assumed that Each person in the conversation group speaks using the headset 2 having a microphone function. That is, each headset 2 used by each person acquires the voice of the conversation of the person wearing the headset 2, converts it into an electrical signal indicating the voice, and outputs it to the information processing apparatus 1. The information processing device 1 and each headset 2 are connected by, for example, wired communication by cable connection or wireless communication such as Bluetooth (registered trademark) or wireless LAN. The information processing apparatus 1 converts the electric signal indicating the sound output from each headset 2 into sound data including a digital sound signal, and for each headset 2, that is, for the persons P1, P2, and P3. Audio data is stored in each.

  Next, the configuration of the information processing apparatus 1 will be described. FIG. 2 is a block diagram illustrating an outline of the internal configuration of the information processing apparatus 1.

  The information processing apparatus 1 is a computer, for example. The information processing apparatus 1 includes a control unit 10, a ROM 112, a RAM 113, an HDD 114, a display unit 115, a communication interface 118, and an input unit 119. These units can transmit / receive data or signals to / from each other via a CPU bus.

  The control unit 10 controls the operation of the entire information processing apparatus 1. The ROM 112 stores an operation program for basic operations of the information processing apparatus 1. The RAM 113 is used as an operation area of the control unit 10.

  The HDD 114 stores an utterance analysis program according to an embodiment of the present invention in a part of the storage area. Further, the HDD 114 stores the respective audio data for the persons P1, P2, and P3. The HDD 114 is also an example of a result storage unit in the claims.

  The display unit 115 includes an LCD (Liquid Crystal Display) or the like, and displays operation guidance for an operator who operates the information processing apparatus 1.

  The communication interface 118 has a USB interface or a wireless LAN interface. The communication interface 118 functions as an interface for performing data communication with each headset 2 described above.

  The input unit 119 includes a keyboard, a mouse, and the like, and an operation instruction is input from an operator.

  The control unit 10 includes a processor, a RAM (Random Access Memory), a ROM (Read Only Memory), and the like. The processor is a CPU (Central Processing Unit), MPU, ASIC, or the like. The control unit 10 functions as the control unit 100, the speech time detection unit 101, and the conflict / opinion determination unit 102 when the speech analysis program stored in the HDD 114 is executed by the above processor. Note that the control unit 100, the speech time detection unit 101, and the conflict / opinion determination unit 102 may be configured by a hard circuit.

  The control unit 100 has a function for controlling operation of the entire information processing apparatus 1.

  The utterance time detection unit 101 detects the utterance time for each utterance in the utterance recorded in the audio data from each of the audio data for the persons P1, P2, and P3 stored in the HDD 114. . FIG. 3 is a diagram illustrating an example of audio data. The vertical axis in FIG. 3 indicates the amplitude of sound (unit db), and the horizontal axis indicates time. The speech time detection unit 101 analyzes the voice data, and is the amplitude indicated by the voice data, which is continuously larger than a predetermined magnitude (for example, 0.25 seconds) for a predetermined time (for example, 0.25 seconds). , 20dB) is extracted as a comment. In the audio data shown in FIG. 3, the speech time detection unit 101 extracts a part, b part, and c part as speech.

  The conflict / opinion determination unit 102 determines that the speech time detected by the speech time detection unit 101 is a predetermined first time (the time from the specified time to a predetermined time longer than the specified time). For example, if it is within 0.25 seconds or more and 2.0 seconds as the specified time), the speech made at this speech time is determined as a conflict. In addition, the conflict / opinion determination unit 102 has a predetermined second time (a time exceeding the first time) that is longer than the first time, the speech time detected by the speech time detection unit 101. If there is, the speech made during this speech time is determined as an opinion. The meeting / opinion determination unit 102 causes the HDD 114 to store the result of the determination as to whether the determination is an opinion or an opinion, and the time of each comment and opinion.

  In addition, the conflict / opinion determination unit 102 has a known voice recognition function, and thereby converts the voice data into text by converting it into text. Then, the conflict / opinion determination unit 102 detects, for example, words included in all the comments made by the person P1, and counts the number of times the word is detected when the same word is detected a plurality of times. The conflict / opinion determination unit 102 extracts words that appear more than a predetermined number (for example, five times) as frequent keywords from the detected words. The conflict / opinion determination unit 102 extracts frequent keywords in the same manner for the voice data of all persons.

  Next, a first embodiment of the utterance analysis process by the information processing apparatus 1 will be described. FIG. 4 is a flowchart showing a first embodiment of the speech analysis processing by the information processing apparatus 1.

  The scene where the speech analysis is performed is a scene where a conversation group composed of a plurality of persons P1, P2, and P3 is having a meeting. Each person P1, P2, P3 wears a headset 2, and each headset 2 is communicably connected to the information processing apparatus 1 as described above. In this state, people P1, P2, and P3 speak during the conference. Voices uttered by the persons P1, P2, and P3 are collected by the headsets 2 of the persons P1, P2, and P3 and output to the information processing apparatus 1.

  The information processing apparatus 1 acquires audio data from each headset 2 via the communication interface 118 (S1). That is, the communication interface 118 receives an electrical signal indicating the sound output from each headset 2, and the speech time detection unit 101 includes the electrical signal indicating the acquired sound as a digital audio signal. The sound data is converted and stored in the HDD 114. The speech time detection unit 101 stores the audio data in the HDD 114 for each person P1, P2, P3.

  Subsequently, the utterance time detection unit 101 determines each utterance in the utterance indicated by the voice data from each of the voice data stored in the HDD 114 for each person P1, P2, P3 as described above. Extract (S2). Then, the speech time detection unit 101 detects the speech time of each extracted speech (S3).

  Further, the conflict / opinion determination unit 102 determines whether the utterance time of each utterance is within the first time or the second time (S4). If the speech time is within the first time (“first time” in S4), the conflict / opinion determination unit 102 determines that the speech made during this speech time is a conflict (S5). If the speech time is the second time (“second time” in S4), the conflict / opinion determination unit 102 determines the speech made at the speech time as an opinion (S8).

  Further, the conflict / opinion determination unit 102 converts the voice data of the persons P1, P2, and P3 into text by using the voice recognition function, and extracts frequent keywords from the voice data for each of the persons P1, P2, and P3 ( S6). The meeting / opinion determination unit 102 stores the result of the determination of the above-described determination or opinion, each speaking time of the meeting and the opinion, and the extracted frequent keywords for each person P1, P2, and P3 and stores them in the HDD 114. (S9).

  The conflict / opinion determination unit 102 creates an image in which the determined conflict or opinion result is divided into persons P1, P2, and P3 as shown in FIG. 5 (S10). The control unit 100 displays the created image on the display unit 115 (S11). Here, when the information processing apparatus 1 is an image forming apparatus, the image created in S10 and the data obtained by converting the audio data into text may be printed.

  FIG. 5 is an image diagram showing the utterances of the people P1, P2 and P3 with different shapes according to the utterance time.

  The control unit 100 causes the display unit 115 to display the results of the conflict or opinion determined by the conflict / opinion determination unit 102 for each person P1, P2, and P3 as shown in FIG.

  For example, in FIG. 5, the horizontal axis indicates the time axis, and the strip-shaped images im shown in the respective columns of the persons P1, P2, and P3 (pointing to all the strip-shaped images shown in FIG. 5) are remarks. It is an image showing. In each column of the persons P1, P2, and P3, images im for the number of times of speaking are displayed side by side at positions corresponding to the time of speaking. Further, the horizontal width of the image im is set longer according to the speech time, that is, the horizontal width of the image im is set longer as the speech time is longer. Note that the vertical shape of the image im is set to be the same regardless of the duration of the speech.

  According to the above image diagram, the user who visually recognizes this image diagram can grasp the number of utterances from the number of images im for each person, and the utterance time from the display position of the image im. Can be grasped.

  If the first time is a width w shown in FIG. 5, an image im having a width less than the width w indicates a conflict, and an image im having a width greater than or equal to the width w indicates an opinion. The opinion / opinion determination unit 102 (i) makes the color of the image im different depending on the opinion or opinion, or (ii) makes the notation indicating the difference of opinion or opinion adjacent to the image in the image im. Thus, each image im may be created.

  By visually recognizing the image diagram shown in FIG. 5, the user can further understand what remarks each person P1, P2 and P3 made at what time. Further, in FIG. 5, since a common time axis is used for the persons P1, P2, and P3, it is possible to grasp the timing of the timing of the statements made by one person and another person.

  For example, according to FIG. 5, the person P1 makes an opinion first among all the persons (image im211), and subsequently, the person P2 also gives an opinion (image im221), followed by the person P3. The user can grasp that he is also giving an opinion (image im231).

  In addition, the person P1 continuously gives an opinion (image im111) immediately after the opinion by the person P2 (image im221), and the person P3 continuously gives an opinion (image im131) immediately after the opinion by the person P1 (image im212). ) Can be grasped by the user.

  In addition, from FIG. 5, person P1 gives an opinion six times in 10 minutes, person P2 gives an opinion four times in 10 minutes, and person P3 gives an opinion twice in 10 minutes. Therefore, the user can grasp that the person P1 has the most opinions. From this result, the user can assume that the person P1 has the most active opinion.

  On the other hand, the person P1 performs 5 times of competing in 10 minutes (images im111 to im115), the person P2 performs 4 times of competing in 10 minutes (images im121 to im124), and the person P3 performs 10 minutes. In addition, it is understood that the person P3 is performing the largest amount of consideration since the comparison is performed seven times (images im131 to im137). From this result, the user can assume that the person P3 is most willing to hear the opinions of others.

  It should be noted that the conflict / opinion determination unit 102 may determine that only the speech that is continued after the speech determined to be an opinion is a conflict. In this case, for example, in FIG. 5, only the images im 111, im 112, im 131, im 133, im 135, and im 136 are determined as conflicts by the conflict / opinion determination unit 102.

  Thus, according to the first embodiment, according to the present invention, when a person speaks, it is possible to analyze up to the type of the person's utterance and provide the result.

  Next, a second embodiment of the utterance analysis process by the information processing apparatus 1 will be described. FIG. 6 is a flowchart showing a second embodiment of the utterance analysis process by the information processing apparatus 1. In the description of the second embodiment, the description of the same processing as that of the first embodiment is omitted.

  In the second embodiment, after performing the processing of S11 to S13, which is the same processing as S1 to S3 in the first embodiment, in S14, the competing / opinion determination unit 102 has the first speech time of the speech. When it is determined that it is within the time ("first time" in S14), it is further determined whether or not the comment includes a predetermined keyword (S15). For example, at this point in time, the opinion / opinion determination unit 102 converts the voice data of the people P1, P2, and P3 into text using the voice recognition function, and for each comment, for example, “Yes” “Yes” “Is it?” “I see,” “I ca n’t,” “Yes,” “No,” “Yes,” or “No,” or not. Data converted to text in this way is also used when extracting frequent keywords in S17.

  Here, if the opinion / opinion determination unit 102 determines that the predetermined keyword is included in the utterance (YES in S15), the utterance / opinion determination unit 102 determines that the utterance is an affinity (S16). That is, in the second embodiment, the conflict / opinion determination unit 102 satisfies the condition that the utterance time of the utterance is within the first time and the utterance includes a predetermined keyword. The said speech is determined to be a conflict.

  In addition, when determining that the predetermined keyword is not included in the utterance (NO in S15), the affinity / opinion determination unit 102 determines that the utterance is neither an affinity nor an opinion (S19).

  According to the second embodiment, since the determination of the conflict or opinion by the conflict / opinion determination unit 102 is performed using not only the time element but also an element including a predetermined keyword, the conflict / opinion determination unit When it is determined that 102 is a match, the certainty that each person's statement is a match is further increased.

  Next, a third embodiment of the utterance analysis process by the information processing apparatus 1 will be described. FIG. 7 is a flowchart showing a third embodiment of the utterance analysis process by the information processing apparatus 1. In the description of the third embodiment, the description of the same processing as that of the first or second embodiment is omitted.

  In the said 3rd Embodiment, after performing the process of S21-S24 which is the process similar to S11-S14 in 2nd Embodiment, in S25, the conflict / opinion judgment part 102 was previously defined by the said statement. If it is determined that a keyword is included (YES in S25), the conflict / opinion determination unit 102 further determines that a specific keyword further specified from the predetermined keywords is a processing target in S25. If it is included (YES in S26), it is determined that the comment is a conflict and a whisper that particularly indicates consent (S31). For example, the conflict / opinion determination unit 102 asks for the remarks to be processed in S25 when the specific keyword includes, for example, "I see", "I see", "Yes", or "Yes". Is determined.

  According to the third embodiment, it is possible to discriminate whether or not the person who made the agreement agrees with the opinion of the other person by further determining whether or not the consideration is made by the consideration by the consideration / opinion determination unit 102. It becomes possible.

  In the first to third embodiments, the conflict / opinion determination unit 102 uses the first time used for the determination of the conflict and the second time used for the determination of the opinion of the persons P1, P2, and P3. Although the same time is used for each person, the conflict / opinion determination unit 102 may use different times for each person as the first time and the second time. FIG. 8 is a flowchart showing a determination process for determining a different first time and the second time for each person. The determination process is performed between S3 and S4 in the first embodiment, between S13 and S14 in the second embodiment, and between S23 and S24 in the third embodiment.

  For example, after the processing for detecting the utterance time of each extracted speech is performed by the utterance time detection unit 101 (S3, S13, S23), the conflict / opinion determination unit 102 determines each person P1, P2, P3. The contents of all the utterances are converted into texts using, for example, a voice recognition function, and the respective text amounts are detected for each utterance (S51).

  Subsequently, the conflict / opinion determination unit 102 calculates, as a speech speed, a value obtained by dividing the detected text amount by the speech time of a speech whose target is the text amount. The solicitation / opinion determination unit 102 calculates the speaking rate for all the remarks made by each of the persons P1, P2, and P3, and calculates the average remarking speed for all remarks made by the person for each person. Then, the calculated average speed is set as the speaking speed for the person (S52).

  Further, the conflict / opinion determination unit 102 determines the first time and the second time for each person according to the calculated speaking speed of each person (S53). For example, the conflict / opinion determination unit 102 has a data table indicating each speaking speed and each first and second time associated with each speaking speed. For each person, the first and second times corresponding to the calculated speech time are read, and the read first and second times are determined as the first and second times used for the determination of the conflict or opinion. The data table stores first and second times that become longer as the speech rate becomes slower.

  According to this, the conflict / opinion determination unit 102 determines the first and second times used for determination of the conflict or opinion according to the speaking speed of each of the persons P1, P2, and P3. It becomes possible to determine whether each person's remark is a mutual opinion or opinion.

  Next, a further embodiment of the configuration of the information processing apparatus 1 will be described. FIG. 9 is a block diagram illustrating an outline of a further embodiment of the internal configuration of the information processing apparatus 1. In the description of the further embodiment, the description of the same configuration and function as the configuration of the information processing apparatus 1 described with reference to FIG. 2 is omitted.

  In the further embodiment, in the information processing apparatus 1, the control unit 10 includes a processor, a RAM (Random Access Memory), a ROM (Read Only Memory), and the like, similarly to the configuration illustrated in FIG. The processor is a CPU (Central Processing Unit), MPU, ASIC, or the like. When the evaluation program stored in the HDD 114 is executed by the above processor, the control unit 10 further includes the evaluation unit 103 and the aggregation in addition to the control unit 100, the speech time detection unit 101, and the conflict / opinion determination unit 102. It functions as the unit 104. Note that the control unit 100, the speech time detection unit 101, the conflict / opinion determination unit 102, the evaluation unit 103, and the totaling unit 104 may be configured by hardware circuits.

  The control unit 100, the speech time detection unit 101, and the conflict / opinion determination unit 102 have the same functions as those described above with reference to FIG. Also in this embodiment, the conflict / opinion determination unit 102 summarizes the result of the determined conflict or opinion, each speech and opinion duration, and the extracted frequent keywords for each person P1, P2, and P3. And stored in the HDD 114.

  The evaluation unit 103 assigns a predetermined first evaluation point to each other for each result stored in the HDD 114, and what is the first evaluation point for an opinion? A different predetermined second evaluation point is given. For example, the evaluation unit 103 gives the second evaluation point for the opinion as a larger value (for example, 10 points) than the first evaluation point (for example, 1 point) for the conflict.

  The totaling unit 104 totals the first and second evaluation points given by the evaluation unit 103 for each person P1, P2, P3. For example, the control unit 100 outputs the total result for each person to the display unit 115 and causes the display unit 115 to display the total result for each person. Alternatively, the control unit 100 outputs the data indicating the total result for each person to the computer connected to the information processing apparatus 1 via the communication interface 118. The computer can print or display data indicating the total result for each person.

  Further, the totaling unit 104 creates image data in which the totaling result for each person is graphed. The control unit 100 outputs and displays the image data in which the aggregation result for each person is graphed on the display unit 115.

  Next, an evaluation process for each person's remarks by the information processing apparatus 1 according to the further embodiment will be described. FIG. 10A is a flowchart illustrating a first embodiment of an evaluation process for each person's remarks by the information processing apparatus 1. FIG. 10B is a diagram illustrating an example of an image obtained by graphing the aggregation results obtained by aggregating the first and second evaluation points for each person P1, P2, and P3. In addition, description is abbreviate | omitted about the process similar to 1st Embodiment of the said speech analysis process demonstrated using FIG.

  In the first embodiment of the evaluation process, as in the first embodiment of the utterance analysis process described with reference to FIG. 4, the conflict / opinion determination unit 102 determines the conflict (S65) and the opinion (S70). ), The evaluation unit 103 performs a first evaluation point (e.g., for each of the above results of each person stored in the HDD 114, and for each of the opinions and opinions indicated by the results). 1 point) and a second evaluation point (for example, 10 points) is given to the opinion (S66). The meeting / opinion determination unit 102 assigns the first evaluation point or the second evaluation point to all statements (consideration or opinion) by each of the persons P1, P2, and P3.

  Subsequently, the tabulation unit 104 tabulates the first and second evaluation points given by the evaluation unit 103 for each person P1, P2, P3 (S67). Further, the totaling unit 104 creates an image in which the totalization result is graphed (S68). As illustrated in FIG. 10B, the control unit 100 outputs and displays an image on which the tabulation result for each person is graphed on the display unit 115 (S69).

  According to the first embodiment of such an evaluation process, the evaluation points indicated by numerals are assigned to each of the considerations and opinions determined by the opinion / opinion determination unit 102, and the first and second evaluation points given above are given. Since the two evaluation points are counted for each person P1, P2, and P3, it is possible to evaluate the remarks made by each person in the meeting and present the evaluation results objectively based on the counting results for each person. become. As a result, the user can objectively grasp the evaluation of each person's remarks at the conference.

  Here, the evaluation unit 103 may give the second evaluation point for the opinion differently depending on the presence or absence of the opinion. FIG. 11 is a flowchart showing a process of assigning the second evaluation point for an opinion differently depending on the presence or absence of the opinion. This process is performed during the point grant process of S66 shown in FIG. 10A.

  The evaluation unit 103 continues to the opinion at the timing immediately after the opinion is given by a certain person after the consideration / opinion determination unit 102 performs the determination of the conflict (S65) and the determination of the opinion (S70). It is determined whether there is a conflict made by the person (S71). When the evaluation unit 103 determines that there is a conflict performed at the timing (YES in S71), the evaluation unit 103 sets the evaluation point of the opinion to a predetermined third evaluation point that is larger than the second evaluation point ( For example, 20 points) are given to the opinion (S72).

  When the evaluation unit 103 determines in S71 that there are a plurality of conflicts made at the above timing, the evaluation unit 103 assigns the third evaluation point to the opinion as a further increased value (for example, 25 points). You may do it. The meaning of “there is a plurality of conflicts” may be either a case in which a plurality of people have a conflict or a case in which the same person has a plurality of conflicts.

  In addition, the evaluation unit 103 determines a fourth evaluation point (for example, 2 points) that is larger than the first evaluation point for the above-described interaction performed immediately after the opinion at the above timing. ) Is given (S73).

  If the evaluation unit 103 determines in S73 that there are a plurality of conflicts performed at the above timing, the evaluation unit 103 assigns the fourth evaluation point to the conflict as a value (for example, 3 points) further increased. You may do it. The meaning of “there is a plurality of conflicts” may be either a case in which a plurality of people have a conflict or a case in which the same person has a plurality of conflicts.

  On the other hand, when the evaluation unit 103 determines that there is no conflict performed at the timing (NO in S71), the evaluation unit 103 assigns the evaluation point of the opinion as the second evaluation point to the opinion (S74).

  Furthermore, the evaluation unit 103 assigns a first evaluation point to the conflict performed at a timing that is not continuous immediately after the opinion (S75). In this embodiment, the third evaluation point is set larger than the fourth evaluation point.

  According to this process, immediately after the opinion, other people have been in conflict with each other, and the opinion that is assumed to be a good opinion that has attracted the interest of other people is Since a higher evaluation point is given than other opinions that have not been discussed, it is possible to accurately give a high evaluation to an opinion that is assumed to be of good quality.

  In addition, it was assumed that the controversy that was made immediately after the opinion was an accurate consensus to express one's intention to the opinions of other people. Give a higher evaluation point than Ao. Thereby, it becomes possible to give high evaluation with respect to the consideration assumed that it was performed correctly.

  Further, the evaluation unit 103 may assign the second evaluation point for the opinion differently depending on the keyword included in the opinion. FIG. 12 is a flowchart showing a process of assigning the second evaluation point for an opinion differently depending on a keyword included in the opinion. This process is performed during the point grant process of S66 shown in FIG. 10A. Further, the second evaluation point determined by the processing is also used as the second evaluation point in the processing shown in FIG.

  The solicitation / opinion determination unit 102, after performing the determination of the companion (S65) and the determination of the opinion (S70), the contents of all the remarks (at least about the opinion) by each of the persons P1, P2, and P3 are voiced Text is prepared by the recognition function (S81). Note that the text conversion process may be performed by the evaluation unit 103.

  The evaluation unit 103 detects the words included in all opinions by each of the people P1, P2, and P3 after the text processing is performed by the conflict / opinion determination unit 102, and the same word ( When a keyword is detected a plurality of times, the number of times of detection is counted for each word. That is, the evaluation unit 103 performs the count for each opinion.

  Then, the evaluation unit 103 determines whether or not there is an opinion including the same word more than a predetermined number (for example, 5 times) (S82). When the evaluation unit 103 determines that there is an opinion including the same number of words or more (YES in S82), the evaluation unit 103 sets the second evaluation point as not to be given to the opinion ( S87). The evaluation unit 103 does not perform the setting when there is no opinion in which the same word appears more than a predetermined number (NO in S82).

  Furthermore, the evaluation unit 103 determines a predetermined unnecessary keyword (for example, a word determined not to be related to the current meeting, or “do” for all opinions by the persons P1, P2, and P3. It is determined whether or not there is an opinion including a predetermined inappropriate word such as “I can't do it” (S83). If the evaluation unit 103 determines that there is an opinion including the unnecessary keyword (YES in S83), the value of the second evaluation point to be given to the opinion including the unnecessary keyword is set to a predetermined subtraction value. A value (for example, 5 points) subtracted by (for example, 5 points) is set (S86). If there is no opinion including an unnecessary keyword, the evaluation unit 103 does not perform the subtraction (NO in S83).

  Subsequently, the evaluation unit 103 determines a predetermined important keyword (for example, a word that is desired to be spoken in the current meeting. For example, an image for all opinions by each of the persons P1, P2, and P3. If the conference is about the development of the image forming mechanism portion of the forming apparatus, it is determined whether or not there is an opinion including words such as “charging”, “development”, and “exposure” (S84).

  When the evaluation unit 103 determines that there is an opinion including the important keyword (YES in S84), the value of the second evaluation point to be given to the opinion including the important keyword is set as a predetermined added value. A value obtained by adding (for example, 5 points) (for example, 15 points) is set (S88). The evaluation unit 103 does not perform the addition when there is no opinion including the important keyword (NO in S84).

  Thereafter, the evaluation unit 103 adds the second evaluation points calculated at this time, such as the value of the second evaluation point subjected to the addition or / and subtraction, or the second evaluation point itself, Use as second evaluation point for opinion.

  According to such an evaluation process, the same is repeated many times in one opinion, and a second evaluation point is not assigned to an opinion that is assumed to be useless (S87). ) For opinions that are unneeded words and are considered to have a low contribution to the meeting, the second evaluation point is subtracted to lower the evaluation (S86), and important words are remarked. For those opinions that are assumed to have a high degree of contribution to the conference, the evaluation content will be raised according to the content of the opinion, such as increasing the evaluation by adding to the second evaluation point (S88). Can be determined.

  Next, a first embodiment of an evaluation level determination process for determining an evaluation level for each of the persons P1, P2, and P3 after assigning evaluation points as described above will be described. FIG. 13 is a flowchart showing a first embodiment of an evaluation level determination process for evaluating each of the persons P1, P2, and P3.

  As an evaluation level determination process for determining the evaluation level for each person, a method will be described in which attention is paid to the opinions made by each person. In this case, the totaling unit 104 calculates the total value by totaling the second evaluation points given for each opinion for each of the persons P1, P2, and P3 (S101).

  Then, the evaluation unit 103 determines whether or not each of the total values for each person counted by the counting unit 104 is equal to or greater than a predetermined first opinion evaluation value (S102). Here, when the evaluation unit 103 determines that there is a total value that is less than the first opinion evaluation value (NO in S102), the evaluation for the person who has the total value for the opinion is set to the second low level. Determination is made (S105). The second low level is a low evaluation level given to the person as a person who does not give an opinion or a person who approximates it at the meeting.

  Further, when the evaluation unit 103 determines that there is a total value equal to or greater than the first opinion evaluation value (YES in S102), the total value is determined in advance to be larger than the first opinion evaluation value. It is determined whether or not the second opinion evaluation value is exceeded (S103). That is, the second opinion evaluation value is a value indicating the number of opinions that are assumed to be excessive.

  Here, when the evaluation unit 103 determines that the total value is equal to or higher than the second opinion evaluation value (YES in S103), the evaluation for the person who has the total value for the opinion is determined as the first low level. (S106). This first low level is a low evaluation given to the person as a person who gives opinions at the meeting but does not contribute to the meeting because of the excessive number of opinions, or a person who approximates it. Is a level.

  On the other hand, when the evaluation unit 103 determines that the total value is less than the second opinion evaluation value (NO in S103), the evaluation for the person who has the total value for the opinion is determined as the first high level. (S104). The first high level is a high evaluation level given to the person as a person who contributes to the meeting or who approximates the opinion at an appropriate number of times in the meeting.

  According to such an evaluation level determination process, it is possible to accurately determine the degree of contribution of the person who made the opinion to the meeting according to the number of times the opinion was made.

  Next, a second embodiment of an evaluation level determination process for determining an evaluation level for each of the persons P1, P2, and P3 after giving evaluation points as described above will be described. FIG. 14 is a flowchart showing a second embodiment of an evaluation level determination process for evaluating each of the persons P1, P2, and P3.

  As an evaluation level determination process for determining the evaluation level for each person, a method will be described which pays attention to the interaction performed by each person. In this case, the totaling unit 104 calculates the total value by summing up the first evaluation points given for each of the people P1, P2, and P3 (S201).

  Then, the evaluation unit 103 determines whether or not each of the total values for each person tabulated by the tabulation unit 104 is equal to or greater than a predetermined first conflict evaluation value (S202). Here, when the evaluation unit 103 determines that there is a total value less than the first consideration evaluation value (NO in S202), the evaluation with respect to the person having the total value regarding the consideration is set to the third low level. Determination is made (S206). This third low level is a low evaluation level that is given to the person as a person who is not in a meeting (a person who is assumed not to hear other people's opinions) or a person who approximates it. is there.

  If the evaluation unit 103 determines that there is a total value equal to or greater than the first conflict evaluation value (YES in S202), the total value is determined in advance to be larger than the first conflict evaluation value. It is determined whether or not it is equal to or greater than the second evaluation value (S203).

  Here, when the evaluation unit 103 determines that the total value is less than the second consideration evaluation value (NO in S203), the evaluation unit 103 has a predetermined second evaluation for the person having the total value regarding the consideration. Is determined to be a high level (S208). The second high level is a high evaluation level given to the person as a person who has been meeting with the appropriate number of times in the meeting and who listens to the opinions of others well or who approximates them.

  On the other hand, when the evaluation unit 103 determines that the total value is equal to or greater than the second conflict evaluation value (YES in S203), the counting unit 104 further determines each opinion for each person P1, P2, P3. The total evaluation value is calculated by summing up the second evaluation points assigned to the evaluation points, and the evaluation unit 103 determines whether or not the total value for the opinion is equal to or greater than the first opinion evaluation value described above. (S204).

  Here, if the evaluation unit 103 determines that the total value for the opinion is less than the first opinion evaluation value (NO in S204), the evaluation for the person who has the total value for the opinion is the fourth evaluation. It is determined that the level is low (S205). The fourth low level is a low level given to the person as a person who has a meeting in the meeting but has too many meetings and does not have much opinions in the meeting, or a person who approximates it. Evaluation level.

  In addition, when the evaluation unit 103 determines that the total value for the opinion is equal to or greater than the first opinion evaluation value (YES in S204), the evaluation for the person who has the total value for the opinion is evaluated as the third high value. The level is determined (S207). This third high level is a high evaluation level that is given to the person as a person who contributes to the meeting because he has a lot of opinions but contributes to the meeting or who approximates it. It is.

  According to such an evaluation level determination process, it is possible to accurately determine the degree of contribution to the meeting of the person who has made the relevant consultation according to the number of times of having given the opinion and the number of times of having given an opinion.

  Next, a third embodiment of an evaluation level determination process for determining an evaluation level for each of the persons P1, P2, and P3 after assigning evaluation points as described above will be described. FIG. 15 is a flowchart showing a third embodiment of an evaluation level determination process for evaluating each of the persons P1, P2, and P3.

  Processing for performing evaluation level determination for determining the evaluation level for each person by paying attention to the total value of evaluation points given to the opinions and opinions performed by each person will be described.

  In this case, the totaling unit 104 calculates the total value of the evaluation points given to the opinions and opinions for each of the people P1, P2, and P3 (S301). Note that the totaling unit 104 can calculate the total value for any evaluation point given to the opinion and opinion in each of the above embodiments.

  Subsequently, the evaluation unit 103 determines whether or not the total value is equal to or greater than a predetermined comprehensive evaluation value (S302). This comprehensive evaluation value is a predetermined value for determining whether or not each person contributes to the conference. For example, 50 points are used as a comprehensive evaluation value in a 10-minute meeting.

  Here, when the evaluation unit 103 determines that the total value of the opinion is equal to or greater than the comprehensive evaluation value (YES in S302), the comprehensive evaluation for the person having the comprehensive evaluation value is set to a predetermined high level. The level is determined (S303). This high-level level is a high evaluation given to the person as a person who contributes to the meeting or who approximates the person who has moderately given opinions and opinions at the meeting.

  On the other hand, if the evaluation unit 103 determines that the total value for the opinion is less than the total evaluation value (NO in S302), the general evaluation for the person having the total evaluation value is set to a predetermined general level. Determination is made (S304). The general level is an evaluation given to the person as a person who does not contribute to the meeting as much as the person at the high level or a person who approximates the person.

  Subsequently, the evaluation unit 103 creates an image in which the total value of the evaluation points calculated for each of the persons P1, P2, and P3 is graphed (S305). The control unit 100 displays the created image on the display unit 115 (S306).

  FIG. 16 is a diagram illustrating an example of an image created by the evaluation unit 103 and graphed with the total value of the evaluation points. In the image shown in FIG. 16, the horizontal axis represents time (elapsed time), and the vertical axis represents each person's point ratio (the width indicating each person's point in the vertical direction is a width corresponding to the point ratio). It is created by. The said point ratio is a ratio of the total value of each person with respect to the total value which totaled the total value of the said evaluation point calculated for every person.

  According to the image shown in FIG. 16, the user can visually grasp the contribution degree of each person P1, P2, P3 to the meeting and the time-series change in the contribution ratio over time. Is possible. For example, at the time when 10 minutes have passed since the start of the conference, the contribution rate is higher in the order of person P1, person P2, and person P3, but the contribution rate of person P2 is the highest since about 15 minutes, and 60 minutes at the end of the meeting. It can be understood that the contributions of the person P1, the person P2, and the person P3 are substantially the same at the time.

  Moreover, in the said embodiment, the structure and process which were shown by the said embodiment using FIG. 1 thru | or FIG. 16 are only one Embodiment of this invention, and are not the meaning which limits this invention to the said structure and process. For example, as the information processing apparatus 1, an information processing apparatus such as a personal computer or a communication terminal apparatus having the configuration shown in FIG. 2 as well as an image forming apparatus having the same configuration as that shown in FIG. 2 is applied to the present invention. can do.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 2 Headset 10 Control unit 100 Control part 101 Speech time detection part 102 Conflict and opinion determination part 103 Evaluation part 104 Total part 115 Display part

Claims (5)

A communication interface for data communication with the outside,
A speech time detection unit that extracts each speech of a person from voice data of a plurality of persons acquired via the communication interface, and separately detects the speech time for each extracted speech;
Determining whether the speech time of each person detected by the speech time detection unit is within a predetermined first time, determining a speech within the first time as a conflict, An opinion / determination determination unit that determines a comment made at a predetermined second time longer than the first time as an opinion, and converts the content of the opinion into text for each person,
A result storage unit that stores the result of whether the remark is either a reconciliation or an opinion determined by the reconciliation / opinion determination unit;
For each person's result stored in the result storage unit, a predetermined first evaluation point is given to the conflict, and the opinion is different from the first evaluation point in advance. An evaluation unit for granting a predetermined second evaluation point;
A totaling unit that totalizes each of the first and second evaluation points given by the evaluation unit and creates image data in which the totaling result of each person is graphed, and
When the evaluation unit detects a predetermined important keyword in the opinion converted into text by the conflict / opinion determination unit, the second evaluation point is determined in advance for the opinion When adding and adding only the addition value and detecting a predetermined unnecessary keyword, the second evaluation point is subtracted by a predetermined subtraction value and given to the opinion,
The evaluation unit totals the assigned first evaluation points for each person, and when the point indicated by the total is equal to or higher than a predetermined first conflict evaluation value, the evaluation of the person is performed. The level is determined to be a predetermined first high level, the assigned second evaluation points are totaled for each of the persons, and the point indicated by the total is a predetermined first opinion evaluation value When it is above, the information processing apparatus which determines the said person's evaluation level as the 2nd predetermined high level.   Based on the result of each person stored in the result storage unit, the evaluation unit calculates the second evaluation point for the opinion based on the result of each person based on the first opinion evaluation value. Is greater than a predetermined second opinion evaluation value, and if it is determined that the condition is satisfied, the evaluation level of the person is set to a predetermined first level. The information processing apparatus according to claim 1, wherein the information processing apparatus determines that the level is low.   The evaluation unit, for each person, based on the result of each person stored in the result storage unit, the total of the first evaluation points for the interaction is greater than the first interaction evaluation value. The information processing apparatus according to claim 1 or 2, wherein when the determination is small, the person's evaluation level is determined to be a predetermined third low level.   The evaluation unit, for each person, based on the result of each person stored in the result storage unit, the total of the first evaluation points for the conflict is greater than the first value. When it is equal to or higher than a predetermined second opinion evaluation value, and it is determined whether or not the aggregation of the second evaluation point for the opinion satisfies the condition that the first opinion evaluation value or more satisfies the condition The information processing apparatus according to claim 1, wherein the person's evaluation level is determined to be a predetermined third high level.   When the evaluation unit detects a predetermined number or more of the same keywords in the opinion converted into text by the conflict / opinion determination unit, the evaluation unit gives the second evaluation point to the opinion The information processing apparatus according to any one of claims 1 to 4.