WO2019130818A1 - Information processing device and group reconstruction method - Google Patents

Information processing device and group reconstruction method Download PDF

Info

Publication number
WO2019130818A1
WO2019130818A1 PCT/JP2018/040838 JP2018040838W WO2019130818A1 WO 2019130818 A1 WO2019130818 A1 WO 2019130818A1 JP 2018040838 W JP2018040838 W JP 2018040838W WO 2019130818 A1 WO2019130818 A1 WO 2019130818A1
Authority
WO
WIPO (PCT)
Prior art keywords
unit
group
person
time
utterance
Prior art date
Application number
PCT/JP2018/040838
Other languages
French (fr)
Japanese (ja)
Inventor
祐毅 小林
菜美 西村
知子 真野
Original Assignee
京セラドキュメントソリューションズ株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 京セラドキュメントソリューションズ株式会社 filed Critical 京セラドキュメントソリューションズ株式会社
Publication of WO2019130818A1 publication Critical patent/WO2019130818A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/10—Speech classification or search using distance or distortion measures between unknown speech and reference templates

Definitions

  • the present invention relates to an information processing apparatus and a group reconstruction method, and more particularly to a technology for analyzing contents uttered by a person.
  • This technology stores the speech time and speech time length uttered by each member of the group, divides the speech time and speech time length of each member into time series for each group, and speaks the speech time length of all members of the group The ratio of the speaking time length of each member to is calculated, and a graph in which the speaking density contribution rate of each member is drawn is generated for each group.
  • the speech density contribution rate corresponds to the speech duration of all members in the group. It is only a ratio of the speaking time length of each member to which it belongs. That is, the above technology detects the degree of contribution of each member in a group that has already been configured, and it is expected that more effective discussions will be conducted based on the degree of contribution of discussions held in the past, etc. It does not reorganize new groups that can.
  • the present invention has been made in view of the above-mentioned circumstances, and is to reconstruct a new group which can be expected to have more effective debate based on the degree of contribution of the debate conducted in the past, etc. With the goal.
  • the information processing apparatus is a speech time for each utterance in the utterance from voice data in which the utterances of each person in a plurality of predetermined number of people are recorded.
  • the first detection unit that detects the first message
  • the first counting unit that counts the speech time for each message detected by the first detection unit, and the speaking time of each person counted by the first counting unit
  • a reconfiguring unit configured to reconstruct a group based on the information.
  • a group reconfiguring method in which speech is generated from speech data in which speech of each person in a plurality of predetermined groups is recorded.
  • a speech time detection step for detecting a speech time for speech, a speech time counting step for each person for counting the speech time for each speech detected in the speech time detection step, and a speech time count for each person
  • D a group reconstruction step of reconstructing a group based on the speaking time of each person tabulated in the step.
  • an audio input unit to which an electrical signal indicating audio is input and an electrical signal each time an audio signal is input to the audio input unit is based on the input electrical signal.
  • the processor includes a storage unit that stores voice data for each person who has produced voice, and a processor, and the processor executes a group configuration program to extract a portion corresponding to a speech from the voice data, and the speech continues Based on a first detection unit that detects time as a speaking time, a first counting unit that counts at least one speaking time for each person and calculates a speaking time of each person, and based on the speaking time of each person, A configuration unit that configures a group to which each person belongs, and a control unit that functions as a unit.
  • FIG. 1 is a diagram showing an information processing apparatus according to a first embodiment of the present invention, and a target person who is evaluated by the information processing apparatus. It is a block diagram which shows the outline of an internal structure of the information processing apparatus which concerns on 1st Embodiment. It is a figure which shows an example of audio
  • FIG. 1 is a diagram showing an information processing apparatus according to a first embodiment of the present invention, and a target person who is evaluated by the information processing apparatus.
  • the information processing apparatus 1 acquires, as voice data, voices uttered by each person belonging to a plurality of predetermined conversation groups G1 to G3. For example, a plurality of persons P11, P12, P13 belonging to the group G1, persons P21, P22, P23 belonging to the group G2, persons P31, P32, P33 belonging to the group G3 (in this embodiment, each person belonging to three groups Although it is explained as a total of nine people consisting of, but not limited to), it is holding a meeting, a discussion, a class, a meeting, etc. (hereinafter referred to simply as "meeting") on a group basis. It shall be.
  • 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 the voice into an electrical signal indicating the voice, and outputs the electric signal to the information processing device 1.
  • the information processing apparatus 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 electrical signal indicating the voice output from each headset 2 into voice data consisting of a digital voice signal, and for each headset 2, that is, nine people P11 to P33.
  • the voice data about is stored in each.
  • FIG. 2 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1 according to the first embodiment.
  • the information processing apparatus 1 is, for example, a computer.
  • the information processing apparatus 1 includes a control unit 10, a read only memory (ROM) 112, a random access memory (RAM) 113, a hard disk drive (HDD) 114, a display unit 115, a communication interface 118, and an instruction input. And a unit 119. These units are capable of transmitting and receiving data or signals to each other via a central processing unit (CPU) bus.
  • CPU central processing unit
  • the control unit 10 controls the operation of the entire information processing apparatus 1.
  • the ROM 112 stores an operation program for the basic operation of the information processing device 1.
  • the RAM 113 is used as an operation area or the like of the control unit 10.
  • the HDD 114 stores an evaluation program according to the first embodiment of the present invention in part of its storage area. Further, the HDD 114 stores the above-mentioned audio data of nine persons P11 to P33.
  • the HDD 114 is also an example of a storage unit in the claims. However, a non-volatile ROM (for example, built in the control unit 10) included in the information processing apparatus 1 may function as the storage unit.
  • Identification information for specifying the headset 2 is attached in advance to the headset 2 connected to the information processing apparatus 1.
  • the identification information is not particularly limited as long as it is information that can identify the headset 2. For example, there is an identification number.
  • the HDD 114 stores the identification information for each headset 2 in advance.
  • the HDD 114 also stores each of the identification information in association with group information for specifying a group in accordance with an instruction input by the user via the instruction input unit 119.
  • the HDD 114 stores identification information of the headset 2 used by each of the persons P21, P22, and P23 in association with group information specifying the group G1.
  • the HDD 114 also stores identification information of the headset 2 used by each of the persons P21, P22, and P23 in association with group information specifying the group G2.
  • the HDD 114 further stores identification information of the headset 2 used by each of the persons P31, P32, and P33 in association with group information specifying the group G3.
  • the display unit 115 is formed of an LCD (Liquid Crystal Display) or the like, and displays operation guidance or the like for the operator who operates the information processing apparatus 1.
  • LCD Liquid Crystal Display
  • 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 of the headsets 2.
  • the communication interface 118 is an example of the voice input unit in the claims.
  • the instruction input unit 119 includes a keyboard, a mouse, and the like, and the operator inputs an operation instruction.
  • the control unit 10 is configured of a processor, a RAM, a ROM, and the like.
  • the processor is a CPU, a micro processing unit (MPU), an application specific integrated circuit (ASIC), or the like.
  • the evaluation program stored in the HDD 114 is executed by the processor, whereby the control unit 100, the first detection unit 101, the first aggregation unit 102, the first calculation unit 103, and the second calculation are performed.
  • the control unit 100 has a function of controlling the operation control of the entire information processing apparatus 1.
  • the first detection unit (speech time detection unit) 101 receives each speech in speech recorded in speech data from each of speech data of nine persons P11 to P33 stored in the HDD 114. Detect the speaking time about.
  • FIG. 3 is a diagram showing an example of audio data. The vertical axis in FIG. 3 represents the amplitude of sound (in dB), and the horizontal axis represents time.
  • the first detection unit 101 analyzes the audio data, and among the amplitudes indicated by the audio data, the amplitude (for example, a predetermined amplitude or more) continuously for a predetermined time (for example, 0.25 seconds) or more. Extract the part where 20 dB) continues as an utterance.
  • the first detection unit 101 detects the time during which each of the extracted utterances is continuing as the utterance time and stores it in the HDD 114. In the voice data shown in FIG. 3, the first detection unit 101 extracts a part, b part, and c part as speech.
  • a first counting unit (speech-based speech time counting unit) 102 counts, for each person, the speech time for each of the utterances detected by the first detection unit 101. In this case, the speech time for each of the above-mentioned speeches is summed up for each identification information stored in the HDD 114.
  • the first calculating unit (total utterance time calculating unit) 103 adds up the utterance times of all the utterances detected by the first detection unit 101, and calculates the total time of the utterance times of all persons in all the groups. .
  • the second calculating unit (the individual-by-person ratio calculating unit) 104 sets the ratio of the speaking time of each person counted by the first counting unit 102 to the total time calculated by the first calculating unit 103 as the first Calculated as a percentage.
  • the first granting unit (first evaluation point granting unit) 105 sets evaluation points higher for people who have a larger ratio (first ratio) of the speaking time of each person calculated by the second calculating unit 104, A first evaluation point corresponding to the first ratio is given to each person.
  • the third calculation unit (group-by-group utterance time calculation unit) 106 counts the utterance time for each utterance detected by the first detection unit 101 for each group, and the total utterance time of each person belonging to the group is grouped Calculate every time. In this case, the speech time for each of the above-mentioned speeches is counted for each group information stored in the HDD 114.
  • the fourth calculating unit (in-group ratio calculating unit) 107 calculates the speaking time of each person detected by the first counting unit 102 based on the total speaking time for each group calculated by the third calculating unit 106.
  • the ratio in the group which is the ratio to the above-mentioned total utterance time of the group to which each person concerned belongs is calculated as the second ratio.
  • the second granting unit (second evaluation point granting unit) 108 sets evaluation points higher for people with a larger ratio (second ratio) of the above-described persons calculated by the fourth calculating unit 107, and Each person is further awarded with a second evaluation point according to the in-group ratio.
  • the determination unit (completion / opposition determination unit) 109 determines that the utterance time detected by the first detection unit 101 is a predetermined first time (predetermined time longer than the prescribed time from the prescribed time). If the time to time is within, for example, 0.25 seconds or more and 2.0 seconds as the above-mentioned prescribed time), the utterance made at this utterance time is determined to be a compliment (b in the example of FIG. 3) . In addition, if the utterance time detected by the first detection unit 101 is the second predetermined time (time exceeding the first time), the determination unit 109 determines that the utterance time detected by the first detection unit 101 is longer than the first time.
  • the utterance made at this utterance time is judged as an opinion (a, c in the example of FIG. 3).
  • the determination unit 109 causes the HDD 114 to store the result as to whether or not the determined combination or opinion, and each utterance time of the combination and opinion.
  • the storage control unit (result storage unit) 110 causes the HDD 114 to store the result of the determination made by the determination unit 109, that is, the result of whether the remark is a compliment or an opinion separately for each of the persons P11 to P33.
  • the third granting unit (third evaluation point granting unit) 121 is based on the result of each of the persons P11 to P33 stored by the storage control unit 110, at another timing immediately after the opinion is given by the other person. If it is determined that the person has made the above-mentioned reciprocation, the third evaluation point is further given to the person who made the opinion.
  • the text conversion unit 122 has a known voice recognition function, and thereby converts the contents of the speech of each person included in the voice data into characters.
  • the addition giving unit (additional point giving unit) 123 determines, based on the text data made into text by the textification unit 122, whether or not the remarks by each of the persons P11 to P33 include a predetermined keyword, An additional point is given to an utterance judged to contain the keyword.
  • the fourth giving unit (fourth evaluation point giving unit) 124 adds up the addition points for the message given by the addition giving unit 123 for each person, and sets the total value of the addition points as a fourth evaluation point, Further grant to those who are targeted for the calculation.
  • the fifth calculation unit (total value calculation unit) 125 calculates the total value of the addition points for all persons belonging to the plurality of groups G1 to G3 using the addition points given by the addition giving unit 123. .
  • the fourth provision unit 124 calculates the ratio (third ratio) of the fourth evaluation points to the total value calculated by the fifth calculation unit 125 for each of the persons P11 to P33, and the ratio is The higher it is, the 4th evaluation point is increased.
  • the sixth calculation unit (group-specific point calculation unit) 126 calculates, for each of the groups G1 to G3, the sum value of each person belonging to the group, for the addition points given by the addition application unit 123.
  • the fourth assignment unit 124 sets the ratio (fourth ratio) of the fourth evaluation point of each person belonging to the group to the total value of the group calculated by the sixth calculation unit 126, for each of the above persons. The fourth evaluation point is increased as the ratio is higher.
  • FIG. 4 is a flowchart showing an evaluation process of a conference participant by the information processing apparatus 1.
  • the scene in which the evaluation is performed is a scene in which each person belonging to the conversation groups G1 to G3 is holding a meeting for each group.
  • Each of the persons P11 to P33 wears the headset 2, and the headsets 2 are communicably connected to the information processing apparatus 1 as described above.
  • the people P11 to P33 speak during the meeting in the respective groups to which they belong.
  • the voice uttered by the people P11 to P33 is collected by the headset 2 of each of the people P11 to P33, and is output to the information processing device 1.
  • the information processing apparatus 1 acquires audio data from each headset 2 via the communication interface 118 (step S1). That is, when the communication interface 118 receives an electrical signal indicating the voice output from each headset 2, the first detection unit 101 includes the electrical signal indicating the acquired voice from a digital voice signal. It is converted into voice data and stored in the HDD 114. The first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
  • the first detection unit 101 extracts, as described above, the utterances in the speech indicated by the voice data from the voice data stored in the HDD 114 for each of the persons P11 to P33. (Step S2). Then, the first detection unit 101 detects the utterance time of each of the extracted utterances (step S3).
  • the first counting unit 102 counts, for each of the persons P11 to P33, the utterance times of the respective utterances detected by the first detection unit 101 individually for each of the persons P11 to P33 (step S4).
  • the first calculation unit 103 sums up the utterance times for all the utterances detected for each person by the first detection unit 101, and calculates the total time of the utterance times of all the persons described above ( Step S5).
  • the second calculating unit 104 calculates, as a first ratio, the ratio of the speaking time of each person, which is counted by the first counting unit 102, to the total time calculated by the first calculating unit 103 (step S6). . That is, the second calculation unit 104 individually calculates the ratio to the person P33, such as the ratio of the speech time of the person P11 to the total time, the ratio of the speech time of the person P12 to the total time, and so on.
  • the first assignment unit 105 sets the evaluation point higher for the person with the larger proportion of the speaking time of each person calculated by the second calculation unit 104, and the first grant unit 105 sets the evaluation point for each person P11 to P33.
  • a first evaluation point is given according to the first ratio (step S7).
  • the first giving unit 105 has 2 points when the first ratio is 0 to less than 20%, 4 points for 20% or more and less than 40%, 6 points for 40% or more and less than 60%, In the case of 60% or more and less than 80%, 8 points are awarded, and in the case of 80% or more and 100%, 10 points are awarded as first evaluation points.
  • the third calculation unit 106 counts the utterance time for each utterance detected by the first detection unit 101 in step S3 for each of the groups G1 to G3, that is, for each group information stored in the HDD 114.
  • the total speech time of each person in group G1, the total speech time of each person in group G2, and the total speech time of each person in group G3 are calculated (step S8).
  • the fourth calculating unit 107 calculates the utterances of the persons P11 to P33 detected by the first counting unit 102 in step S4 with respect to the total utterance time for each group calculated by the third calculating unit 106 in step S8.
  • the intra-group ratio of time is calculated as the second ratio (step S9). That is, the fourth calculating unit 107 calculates, for each of the persons P11 to P33, the ratio of the group to which each person belongs to the total speech time as the ratio within the group.
  • the second giving unit 108 sets evaluation points to be higher as the ratio of the in-group ratio of each person calculated by the fourth calculating unit 107 in step S9 increases, and the group for each of the persons P11 to P33 is A second evaluation point according to the inside ratio is further awarded (step S10).
  • the second granting unit 108 is 4 points when the second ratio is 0 to less than 20%, 8 points when 20% or more and less than 40%, 12 points when 40% or more and less than 60%. In the case of 60% or more and less than 80%, 16 points are awarded, and in the case of 80% or more and 100%, 20 points are awarded as the second evaluation points.
  • the second provision unit 108 sets the second evaluation point higher than the first evaluation point provided by the first application unit 105 (in the present embodiment, the second evaluation point is twice the first evaluation point). It is preferable to give it as an evaluation point. If the ratio of the speech time by each person to the total speech time in each group is higher than the ratio to the total time, which is the sum of the speech times of all the persons P11 to P33, the contribution in the meeting is high. It is because it seems.
  • step S7 the first granting unit 105 performs a first evaluation according to the ratio of the speaking time of each person to the total speaking time of all the persons P11 to P33 who have passed through all the groups G1 to G3. While giving points to each person, in step S10, the second giving unit 108 sets the second evaluation point according to the ratio within the group which is the ratio of the speaking time of each person to the total speaking time for each group. Further, it is possible to make a comprehensive evaluation taking into consideration both the contribution of each person in the group and the overall contribution of all of the persons P11 to P33 through all the groups G1 to G3.
  • control unit 100 displays, on the display unit 115, information indicating the first evaluation point and the second evaluation point given to each person in accordance with the instruction input by the user via the instruction input unit 119. It may be configured as possible.
  • FIG. 5 is a flowchart showing a first modified example of the evaluation processing of the conference participant by the information processing device 1.
  • the description of the first modification the description of the processing similar to that of the first embodiment is omitted.
  • step S11 determines whether the speaking time of each statement is within the first time after the determination unit 109 is further performed. It is determined whether it is the second time (step S11). If the speech time is within the first time ("first time” in step S11), the determination unit 109 determines that the speech made in this speech time is a reunion (step S12). In addition, when the utterance time is the second time (“second time” in step S11), the determination unit 109 determines that the utterance made in the utterance time is an opinion (step S16).
  • the storage control unit 110 determines the result of the determination made by the determination unit 109 in step S12 and step S16, that is, the result of whether the statement is a summary or an opinion, the time at which the statement indicating the summary or opinion is made And each of the persons P11 to P33 is stored in the HDD 114 (step S13).
  • the third assignment unit 121 based on the result of each of the persons P11 to P33 stored by the storage control unit 110, for each of the groups G1 to G3, the opinion given by a person in the group is At the timing immediately after being done, it is determined whether the above-mentioned sumo wrestling is being performed by another person (step S14). When it is determined that there is such an opinion (YES in step S14), the third giving unit 121 further gives the third evaluation point to the person who made the opinion (step S15). For example, the third applying unit 121 applies 10 points as the third evaluation point. When it is determined that there is no such opinion (NO in step S14), the third giving unit 121 does not give the third evaluation point.
  • the opinion, etc. assumed to be a good opinion which is being followed up by another person immediately after the opinion and which has attracted the interest of the other person Since such an opinion will be given a higher evaluation than other opinions that have not been countered, it is possible to appropriately give a high evaluation to an opinion that is supposed to be good.
  • FIG. 6 is a flow chart showing a second modified example of the evaluation processing of the conference participant by the information processing device 1.
  • the description of the second modification the description of the same processes as those of the first embodiment and the first modification will be omitted.
  • the second modification is performed after steps S1 to S10 in the first embodiment or after steps S11 to S16 in the first modification.
  • the text conversion unit 122 converts the contents of the speech of each person included in the voice data into characters and converts them into text (step S20).
  • the addition application unit 123 determines whether the utterance by each of the persons P11 to P33 included in the text data includes a predetermined keyword (refer to FIG. Step S21).
  • the addition giving unit 123 determines that the utterance by each of the persons P11 to P33 included in the text data contains a predetermined keyword (YES in step S21), the addition giving unit 123 includes the keyword. An addition point is added to the determined speech (step S22). Although the text data includes a plurality of utterances by each person, the addition imparting unit 123 determines whether all the utterances include the keyword, and the addition imparting unit 123 Add points to all messages judged to contain a keyword.
  • the addition giving unit 123 determines that the remarks by each of the persons P11 to P33 do not include a predetermined keyword (NO in step S21), the addition giving unit 123 does not give the addition points.
  • the fourth adding unit 124 adds up the added points for each of the above-mentioned statements added by the adding up unit 123 in step S22 for each of the persons P11 to P33 (step S23), and adds up the added points.
  • a value is further assigned as a fourth evaluation point to each person targeted for the aggregation (step S24).
  • the addition giving unit 123 gives the above-mentioned addition points (for example, 1 point) to the respective utterances each time the above-mentioned keyword included in the utterance appears, and adds up each of the persons P11 to P33. Then, the addition giving unit 123 gives the total value of each person as the fourth evaluation point to the person who made the statement including the keyword.
  • the fourth evaluation point is added to the above first to third evaluation points for the person who has made the keyword for which the organizer of the meeting or the like assumed that the degree of contribution is high. Therefore, based on the total value of the points, it is possible to accurately determine the persons P11 to P33 who are truly contributing to the conference.
  • FIG. 7 is a flow chart showing a third modification of the evaluation processing of the conference participant by the information processing apparatus 1.
  • the description of the third modification the description of the same processes as those of the first embodiment, the first modification, and the second modification will be omitted.
  • the fourth adding unit 124 adds up the addition points for each of the utterances added by the addition adding unit 123 in step S22 for each of the persons P11 to P33.
  • the fifth calculation unit 125 further uses the addition points for each person given by the addition giving unit 123 to calculate the total value of the addition points for all persons belonging to a plurality of groups. It calculates (step S31).
  • the fourth adding unit 124 calculates the ratio of the total value of added points to each person to the total value calculated by the fifth calculating unit 125 (step S32), and the higher the calculated ratio, the more The fourth evaluation point is set high and given to the target person (step S33).
  • the fourth giving unit 124 adds 20% of the total value when the ratio is less than 0 to 20%, adds 40% of the total value when 20% or more and less than 40%, 40% or more and 60% In the case of less than 60% of the aggregate value is added, in the case of 60% or more and less than 80%, 80% of the aggregate value is added, and in the case of 80% or more to 100%, 100% of the aggregate value is added.
  • the aggregate value of is given as the fourth evaluation point.
  • the third modification it is possible to evaluate the person who has made a comment on the keyword after adding an objective element in consideration of the state of the user who made a statement in the group and the degree of speech of the person who made the above keyword become.
  • FIG. 8 is a flow chart showing a fourth modification of the evaluation processing of the conference participant by the information processing apparatus 1.
  • the description of the same processes as those of the first embodiment and the first to third modifications will be omitted.
  • the fourth modification is performed after the third modification.
  • the sixth calculator 126 adds the fourth evaluation point to each of the groups G1 to G3 as a group.
  • the total points are calculated by totaling addition points of each person who belongs (step S34).
  • the fourth assignment unit 124 calculates the ratio of the total value of the added points to each person to the total value calculated by the sixth calculation unit 126 (step S35), and the person with the calculated ratio is higher
  • the fourth evaluation point is set high, and is further given to the target person (step S36).
  • the fourth giving unit 124 adds 10% of the total value when the ratio is less than 0 to 20%, adds 20% of the total value when 20% or more and less than 40%, 40% or more and 60% In the case of less than 30% of the aggregate value is added, in the case of 60% or more and less than 80%, 40% of the aggregate value is added, and in the case of 80% or more to 100%, 50% of the aggregate value is added.
  • the aggregate value of is given as the fourth evaluation point.
  • the fourth assignment unit 124 sets the fourth evaluation point for the ratio (third ratio) to the total value of the entire group higher than the fourth evaluation point for the ratio (fourth ratio) to the total value in group units. It is preferable to give a fourth evaluation point.
  • the ratio of the number of occurrences of the above keyword by each person is the one in which the ratio of the number of occurrences of all the persons P11 to P33 in all groups is higher than the ratio to the total of the number of appearances in the individual group It is because it seems that the degree of contribution is high.
  • the degree of speech of the person who has said the keyword is more objective in consideration of the state of speech of the keyword by all members of the whole group. Above, it will be possible to evaluate the person who said the keyword.
  • the information processing apparatus 1A acquires voices uttered by persons belonging to the conversation groups G1 to G3 as voice data.
  • FIG. 9 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1A according to the second embodiment. Descriptions of processes similar to those of the information processing apparatus 1 according to the first embodiment will be omitted.
  • the control unit 10 of the information processing apparatus 1A executes the group reconfiguration program stored in the HDD 114 by the above processor, whereby the control unit 100, the first detection unit 101, and the first aggregation are performed. It functions as the unit 102, the reconstruction unit 130, the textification unit 122, the second detection unit 131, and the second aggregation unit 132.
  • the control unit 100, the first detection unit 101, the first aggregation unit 102, the reconstruction unit 130, the textification unit 122, the second detection unit 131, and the second aggregation unit 132 may be configured by a hardware circuit. .
  • the reconfiguration unit 130 reconfigures each group member based on the speech time of each person counted by the first counting unit 102.
  • the reconfiguration unit 130 reconfigures the members of the groups G1 to G3 based on the speech times of the persons P11 to P33 counted by the first counting unit 102.
  • the text conversion unit 122 converts the contents of each person's utterance included in the voice data into text.
  • the second detection unit (keyword detection unit) 131 determines, based on the text data converted into text by the text conversion unit 122, whether the utterance by each person includes a predetermined keyword.
  • the second tabulating unit (tabulation unit for each group) 132 tabulates the speech time of each of the persons P11 to P33 tabulated by the first tabulating unit 102 for each group reconstructed by the reconstructing unit 130.
  • FIG. 10 is a flowchart showing group reconfiguration processing by the information processing apparatus 1A.
  • the scene in which the evaluation is performed is the same as in the case of the information processing apparatus 1 according to the first embodiment, and as described with reference to FIG. In a meeting.
  • Each of the persons P11 to P33 wears the headset 2, and each headset 2 is communicably connected to the information processing apparatus 1A as described above.
  • the people P11 to P33 speak during the meeting in the respective groups to which they belong.
  • the voice uttered by the people P11 to P33 is collected by the headset 2 of each of the people P11 to P33, and is output to the information processing apparatus 1A.
  • the information processing apparatus 1A acquires audio data from each headset 2 via the communication interface 118 (step S101).
  • the first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
  • the first detection unit 101 extracts, as described above, each utterance in the speech indicated by the voice data from each of the voice data stored in the HDD 114 for each of the persons P11 to P33. (Step S102). Then, the first detection unit 101 detects the utterance time of each of the extracted utterances (step S103).
  • the first aggregation unit 102 individually aggregates, for each of the persons P11 to P33, the speech time for each of the utterances detected by the first detection unit 101 (step S104).
  • the reconfiguration unit 130 performs ranking in order from the person with the longest utterance time counted by the first counting unit 102 (step S105). Then, the reconstruction unit 130 performs grouping into groups of predetermined numbers from the top ranks to reconstruct a group (step S106).
  • the reordering unit 130 ranks the speaking time lengths of the people P11 to P33 from P31, P21, P11, P22, P23, P12, P12, P13, P33, and P32 from the top.
  • the reconstruction unit 130 sets the predetermined number of people as, for example, three, sets members of the group G1 as P31, P21, and P11, and sets members of the group G2 as P22, P23, and P12. Each group is reconfigured with the members as P13, P33, and P32.
  • the control unit 100 updates the identification information and the group information stored in association with each other in the HDD 114 so that the result of the reconfiguration by the reconfiguration unit 130 is reflected.
  • the HDD 114 stores identification information of the headset 2 used by each of P31, P21, and P11 in association with group information specifying the group G1.
  • the HDD 114 also stores identification information of the headset 2 used by each of P22, P23, and P12 in association with group information specifying the group G2.
  • the HDD 114 further stores identification information of the headset 2 used by each of P13, P33, and P32 in association with group information specifying the group G3.
  • the second totaling unit 132 totals the speech time of each person counted by the first totaling unit 102 for each of the reconfigured groups G1 to G3 (step S107).
  • control unit 100 Based on the identification information and the group information stored in HDD 114 and the counting result by second counting unit 132, control unit 100 counts the rearranged groups G1 to G3 and members of each group in S107. The display time is displayed on the display unit 115 for each of the groups G1 to G3 (step S108).
  • step S106 the following processing may be performed in step S106.
  • grouping is performed for each predetermined number of people from the top ranks
  • a group is reconfigured, but at this time, a group is reconfigured without making people whose speaking time is longer than a predetermined time set in advance into the same group.
  • the reconstruction unit 130 performs grouping into groups according to the predetermined number of people from the top of the ranking and reconstructs a group, there are a plurality of persons whose speaking time is longer than the specified time in the same group.
  • the person concerned is replaced with a person who belongs to a group constituted by persons of lower ranks and whose speaking time is not longer than the specified time. This prevents people with long utterances from being in the same group, which can increase the possibility of making a meeting meaningful.
  • FIG. 11 is a flowchart showing a first modified example of the group reconfiguration processing by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
  • the reconfiguring unit 130 divides the utterance into predetermined groupings.
  • each of the persons P11 to P33 is combined to create all the groups that can be created (step S205).
  • a predetermined grouping rule is to create three groups with three members in each group.
  • the reconstruction unit 130 divides the people P11 to P33 into three groups G1 to G3 by three people, covers all combinations that can be distributed in this way, and groups G1 to G3 for all combinations that can be distributed. Create
  • the reconstructing unit 130 counts the utterance time of each member who belongs to all the groups created as described above (step S206).
  • the reconstruction unit 130 calculates the difference in group utterance time between the groups, and extracts the group G1 to G3 in which the difference is the smallest. (Step S207).
  • the control unit 100 causes the display unit 115 to display a tabulation of the utterance time of each of the groups G1 to G3 extracted in step S207 and the extracted groups G1 to G3 (step S208).
  • the first modification in order to adjust the total speaking time in each group to be equal according to the speaking time length (contribution) of each person in the conference held in the past, a certain group There is no bias in the group structure, such as only people with many utterances, and other groups with only few people.
  • FIG. 12 is a flow chart showing a second modification of the group reconstruction processing by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
  • the reconstruction unit 130 sequentially ranks in order from the person with the longest utterance time counted by the first aggregation unit 102 (step S305), and then reconstructs it.
  • Section 130 satisfies the condition that the people from the lowest rank to the predetermined rank and the people from the highest rank to the predetermined rank are in the same group, and for each predetermined number of people.
  • the group is reconfigured (step S306).
  • the reordering unit 130 ranks the speaking time lengths of the people P11 to P33 from P31, P21, P11, P22, P23, P12, P12, P13, P33, and P32 from the top.
  • the predetermined number of people is, for example, three.
  • the reconstruction unit 130 sets only one person with the highest rank to the above “from the highest rank to a predetermined rank”, and “up to two persons from the lowest rank” to the “predetermined rank from the lowest rank Reconfigure each group as "up".
  • the reconfiguration unit 130 first sets P31 of the highest rank and P33 and P32 of the lowest rank to two members of the group G1.
  • the reconfiguration unit 130 sets P21 having the highest rank and two P12 and P13 from the lowest rank as members of the group G2. Similarly, the reconstruction unit 130 excludes the three persons (B21, P12, and P13), and sets P11 having the highest rank and two P22 and P23 from the lowest rank as members of the group G3.
  • the second totaling unit 132 totals the speech time of each person counted by the first totaling unit 102 for each of the reconstructed groups G1 to G3 (step S307).
  • the control unit 100 causes the display unit 115 to display the reconstructed groups G1 to G3 and the sum total of the utterance times of the groups G1 to G3 (step S308).
  • each group is relatively actively written with a person who speaks relatively positively according to the speech time length (contribution) of each person in a conference held in the past.
  • Groups can be configured by combining with people who make a comment, so that there is no bias in group configuration, such that only one group is composed of only people who speak a lot, and another group is composed only of people who have a few voices. .
  • FIG. 13 is a flowchart showing a third modification of the group reconfiguration process by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
  • the third modification is processing executed after group reconfiguration is performed in the second embodiment and the first and second modifications of the second embodiment.
  • the text conversion unit 122 converts the contents of the utterances of the above persons included in the audio data into characters and converts them into text (step S401).
  • the second detection unit 131 specifies an utterance including a predetermined keyword and a person who made the utterance based on the text data (step S402).
  • the reconfiguration unit 130 determines whether or not a plurality of persons who have made a statement including the keyword are members of the same group (step S403). ).
  • the reconfiguration unit 130 determines that a plurality of persons who have made a remark including the above-mentioned keyword are not members of the same group in the groups G1 to G3 after the above-mentioned reconfiguration (Step The process ends while maintaining the groups G1 to G3 configured at this point in time).
  • the reconfiguration unit 130 has a plurality of persons who have made a statement including the keyword become members of the same group (step S403 YES), it is determined whether or not there is a group consisting of only those who do not speak including the keyword among the groups G1 to G3 (step S404).
  • the reconfiguration unit 130 determines that there is a group consisting of only those who do not speak including the keyword, among the groups G1 to G3 (YES in step S404), the reconfiguration unit 130 is one of a group consisting of only one person among those who are both members (people who have made a statement including the above keyword) in the same group but who are not making a statement including that keyword And the group is reconstructed (step S405).
  • the reconfiguration unit 130 rearranges the group. The process is terminated without maintaining the groups G1 to G3 configured at this time without performing the process.
  • the third modified example it is possible to avoid grouping situations in which persons who perform similar utterance content belong to the same group, and to realize grouping into which utterances of non-biased contents are made.
  • FIG. 14 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1B according to the third embodiment. Descriptions of processes similar to those of the information processing apparatus 1 according to the first embodiment or the information processing apparatus 1A according to the second embodiment will be omitted.
  • the evaluation program stored in the HDD 114 is executed by the processor, whereby the control unit 100, the third detection unit 135, the creation unit 136, the evaluation It functions as the unit 137, the text conversion unit 122, and the second detection unit 131.
  • the control unit 100, the third detection unit 135, the creation unit 136, the evaluation unit 137, the text conversion unit 122, and the second detection unit 131 may be configured by a hardware circuit.
  • the third detection unit (speech detection unit) 135 made each of the speeches in the speech from each of the speech data of nine persons P11 to P33 stored in the HDD 114. Detect with the person and the time each utterance was made.
  • the creation unit (speech distribution creation unit) 136 creates a statement distribution map showing a change in the amount of statement according to the passage of time, using each statement detected by the third detection unit 135 and its time.
  • the evaluation unit 137 sets the evaluation level of the person who made the first utterance in the time zone in which the amount of utterance increases in the utterance distribution map created by the creator 136 to a predetermined first high level. Perform processing to determine that
  • the text conversion unit 122 converts the contents of each person's utterance included in the voice data into text.
  • the second detection unit 131 determines, based on the text data converted into text by the text conversion unit 122, whether the utterance by each person includes a predetermined keyword.
  • FIG. 15 is a flowchart showing evaluation processing by the information processing apparatus 1B.
  • the scene where the evaluation process is performed is the same as in the case of the information processing apparatus 1 according to the first embodiment, and each person belonging to the conversation groups G1 to G3 is a group as described with reference to FIG. It is a scene where we have a meeting every time.
  • the information processing apparatus 1B acquires audio data from each headset 2 via the communication interface 118 (step S501).
  • the first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
  • the third detection unit 135 determines, from the voice data stored in the HDD 114 for each of the persons P11 to P33, each utterance being performed in the utterance indicated by the voice data, together with the time of each utterance. Extraction is performed as described above (step S502).
  • time is an elapsed time from the start of audio data.
  • the creating unit 136 creates a statement distribution map showing a change in the amount of messages according to the passage of time, using the utterance and the time of each utterance detected by the third detection unit 135 (step S503). For example, as shown in FIG. 16, in accordance with the passage of time from the start of the conference, a line graph indicating the number of utterances per one minute, for example, is created for each of the groups G1 to G3.
  • the evaluation unit 137 extracts a time zone in which the amount of speech has a tendency to increase in the speech distribution map generated by the generation unit 136 (step S504). For example, the evaluation unit 137 increases, for example, in a predetermined time (for example, one minute) whether the inclination indicated by the broken line in the above-mentioned statement distribution map is equal to or more than a predetermined angle. A time zone in which the rate (differential value) is equal to or more than a predetermined value (50% increase rate in the number of messages) is extracted. For example, in the case of the utterance distribution map shown in FIG. 16, the evaluation unit 137 determines that the slope (increase rate, differential value) from point P1 to point P2 one minute after is 50% or more of the predetermined value. Therefore, the relevant time zone is extracted.
  • a predetermined time for example, one minute
  • the evaluation unit 137 determines the first utterance and the person who made the utterance (step S505). For example, after the time point of point P1 (when 11 minutes has elapsed from the start), the evaluation unit 137 specifies the first speech and speaker.
  • the evaluation unit 137 determines that the evaluation level of the speaker of the first-to-be-sent utterance is the first high level determined in advance (step S506).
  • the first high level is a high evaluation level given to the person as a person who contributes to or is similar to the person at the meeting after making a statement and increasing other remarks. is there.
  • the person who has caused another utterance to increase according to his / her own utterance is judged to be the first high level and evaluated, whereby the influence of the utterance of a certain member on the other utterance is made. Can be given to the members concerned.
  • FIG. 17 is a flowchart showing a first modification of the evaluation process by the information processing apparatus 1B. The same processes as in the third embodiment will not be described.
  • the textification unit 122 converts the contents of the utterance of each person included in the voice data into a character Convert and convert into text (step S 607).
  • the evaluation unit 137 extracts, from the text data converted into text by the text conversion unit 122, each word included in the first utterance made in the time zone in which the extracted utterance amount shows an increasing tendency. (Step S608).
  • the control unit 100 causes the display unit 115 to display the words extracted by the evaluation unit 137 (step S609).
  • FIG. 18 is a flowchart showing a second modified example of the evaluation process by the information processing apparatus 1B. The same processes as those of the third embodiment or the first modification of the third embodiment will not be described.
  • the evaluation unit 137 determines each of the words included in the first utterance. It is determined whether the word is included in the other utterances performed after the first utterance during the extracted time zone to increase the appearance number (step S710).
  • the evaluation unit 137 may be configured such that each word included in the first utterance is the other utterance (all utterances) performed after the first utterance during the extracted time zone. It is determined whether the data appears in the text data converted into text data. When the respective words appear, the evaluation unit 137 detects the number of appearances of each of the words, and determines whether the number of appearances of each of the words is greater than the number of appearances in the first utterance. to decide.
  • the evaluation unit 137 indicates that the word included in the first utterance is included in the other utterances performed after the utterance during the extracted time zone, and the appearance number is increased. If it is determined (YES in step S710), the evaluation level of the person who made the first utterance is determined to be a second evaluation level higher than the first high level (step S711). For example, if the number of occurrences of even one of the above-described words increases, the evaluation unit 137 determines that the number of occurrences is “increased” in step S710. This second high level means that people who contribute to the meeting or have an approximation of the increase in the amount of occurrence of the word after the utterance of a certain word increases in the meeting. As a person, it is a high evaluation level given to the person.
  • the person's words are later frequently used by others in the discussion, and those who give the words that cause the activation of the conference It will be possible to accurately find out and evaluate.
  • Control unit 100 may be configured to be able to display the evaluation result by evaluation unit 137 on display unit 115 in accordance with an instruction input by the user via instruction input unit 119.
  • the evaluation result is not particularly limited as long as it is information indicating an evaluation result, but indicates, for example, whether or not at least one of the first evaluation level and the second evaluation level is given to each person. There is information.
  • the identification of the person who made the utterance was performed based on the identification information attached to the headset 2.
  • the present invention relates to such an embodiment. It is not limited to.
  • a general speaker identification technique may be used to identify the person who made the utterance.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Physics & Mathematics (AREA)
  • Economics (AREA)
  • Acoustics & Sound (AREA)
  • Marketing (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • General Physics & Mathematics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Multimedia (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Data Mining & Analysis (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

An information processing device (1A) that comprises: a first detection part (101) that, from voice data in which is recorded the speech of each person in a plurality of groups that comprise a predetermined plural number of people, detects an utterance duration for each utterance included in the speech; a first totaling part (102) that, by person, totals the utterance durations of the utterances detected by the first detection part (101); and a reconstruction part (130) that reconstructs the groups on the basis of the utterance durations totaled for the people by the first totaling part (102).

Description

情報処理装置及びグループ再構成方法Information processing apparatus and group reconfiguration method
　本発明は、情報処理装置及びグループ再構成方法に関し、特に、人により発話された内容を解析する技術に関する。 The present invention relates to an information processing apparatus and a group reconstruction method, and more particularly to a technology for analyzing contents uttered by a person.
　今日では、グループ毎に討論の活性化を推定してその状況を明示する技術が提案されている（下記特許文献１参照）。この技術は、グループの各メンバーが発話した発話時刻及び発話時間長を記憶しておき、グループ毎に、各メンバーの発話時刻及び発話時間長を時系列に区分し、グループ全メンバーの発話時間長に対する各メンバーの発話時間長の比を算出して、グループ毎に、各メンバーの発話密度寄与率を描画したグラフを生成するというものである。 Nowadays, there is proposed a technique for estimating activation of discussion for each group and clarifying the situation (see Patent Document 1 below). This technology stores the speech time and speech time length uttered by each member of the group, divides the speech time and speech time length of each member into time series for each group, and speaks the speech time length of all members of the group The ratio of the speaking time length of each member to is calculated, and a graph in which the speaking density contribution rate of each member is drawn is generated for each group.
特開２０１６－１６２３３９号公報JP, 2016-162339, A
　上記技術は、各メンバーの発話密度寄与率をグループの討論が活性化しているかの判断のために提供するが、この発話密度寄与率は、グループ内の全メンバーの発話時間長に対する、当該グループに属する各メンバーの発話時間長の比に過ぎない。すなわち、上記技術は、既に構成されているグループにおける各メンバーの貢献度を検出するものであり、過去に行われた討論の貢献度等に基づいて、より効果的な討論が行われることが期待できる新たなグループを再構成するものではない。 Although the above technology provides the speech density contribution rate of each member for determining whether the group discussion is activated, the speech density contribution rate corresponds to the speech duration of all members in the group. It is only a ratio of the speaking time length of each member to which it belongs. That is, the above technology detects the degree of contribution of each member in a group that has already been configured, and it is expected that more effective discussions will be conducted based on the degree of contribution of discussions held in the past, etc. It does not reorganize new groups that can.
　本発明は、上記の事情に鑑みてなされたものであり、過去に行われた討論の貢献度等に基づいて、より効果的な討論が行われることが期待できる新たなグループを再構成することを目的とする。 The present invention has been made in view of the above-mentioned circumstances, and is to reconstruct a new group which can be expected to have more effective debate based on the degree of contribution of the debate conducted in the past, etc. With the goal.
　本発明の一局面に係る情報処理装置は、予め定められた複数人数からなる複数のグループにおける各人の発話が記録された音声データから、当該発話の中でされている各発言についての発言時間を検出する第１検出部と、第１検出部によって検出された各発言についての発言時間を各人についてそれぞれ集計する第１集計部と、第１集計部によって集計された各人の発言時間に基づいて、グループを再構成する再構成部と、を備えるものである。 The information processing apparatus according to one aspect of the present invention is a speech time for each utterance in the utterance from voice data in which the utterances of each person in a plurality of predetermined number of people are recorded. In the first detection unit that detects the first message, the first counting unit that counts the speech time for each message detected by the first detection unit, and the speaking time of each person counted by the first counting unit And a reconfiguring unit configured to reconstruct a group based on the information.
　また、本発明の他の一局面に係るグループ再構成方法は、予め定められた複数人数からなる複数のグループにおける各人の発話が記録された音声データから、当該発話の中でされている各発言についての発言時間を検出する発言時間検出ステップと、発言時間検出ステップで検出された各発言についての発言時間を各人についてそれぞれ集計する各人別発言時間集計ステップと、各人別発言時間集計ステップで集計された各人の発言時間に基づいて、グループを再構成するグループ再構成ステップと、を備えるものである。 Further, according to another aspect of the present invention, there is provided a group reconfiguring method, in which speech is generated from speech data in which speech of each person in a plurality of predetermined groups is recorded. A speech time detection step for detecting a speech time for speech, a speech time counting step for each person for counting the speech time for each speech detected in the speech time detection step, and a speech time count for each person And D. a group reconstruction step of reconstructing a group based on the speaking time of each person tabulated in the step.
　また、本発明の他の一局面に係る情報処理装置は、音声を示す電気信号が入力される音声入力部と、音声入力部に電気信号が入力される度に、入力された電気信号に基づく音声データを、音声を発した人ごとに記憶する記憶部と、プロセッサーを含み、当該プロセッサーがグループ構成プログラムを実行することにより、音声データから発言に対応する部分を抽出し、発言が続いている時間を発言時間として検出する第１検出部と、少なくとも１つの発言時間を、人ごとに集計して、各人の発言時間を算出する第１集計部と、各人の発言時間に基づいて、各人が所属するグループを構成する構成部と、して機能する制御ユニットと、を備える。 Further, in the information processing apparatus according to another aspect of the present invention, an audio input unit to which an electrical signal indicating audio is input and an electrical signal each time an audio signal is input to the audio input unit is based on the input electrical signal. The processor includes a storage unit that stores voice data for each person who has produced voice, and a processor, and the processor executes a group configuration program to extract a portion corresponding to a speech from the voice data, and the speech continues Based on a first detection unit that detects time as a speaking time, a first counting unit that counts at least one speaking time for each person and calculates a speaking time of each person, and based on the speaking time of each person, A configuration unit that configures a group to which each person belongs, and a control unit that functions as a unit.
　本発明によれば、人が発話した際に当該人の発言の種類までを解析してその結果を提供することができる。 According to the present invention, when a person utters, it is possible to analyze up to the type of the person's remark and provide the result.
図１は、本発明の第１実施形態に係る情報処理装置、及び当該情報処理装置により評価が行われる対象者を示す図である。FIG. 1 is a diagram showing an information processing apparatus according to a first embodiment of the present invention, and a target person who is evaluated by the information processing apparatus. 第１実施形態に係る情報処理装置の内部構成の概略を示すブロック図である。It is a block diagram which shows the outline of an internal structure of the information processing apparatus which concerns on 1st Embodiment. 音声データの一例を示す図である。It is a figure which shows an example of audio | voice data. 第１実施形態に係る情報処理装置による会議参加者の評価処理を示すフローチャートである。It is a flowchart which shows the evaluation process of the meeting participant by the information processing apparatus which concerns on 1st Embodiment. 第１実施形態に係る情報処理装置による会議参加者の評価処理の第１変形例を示すフローチャートである。It is a flowchart which shows the 1st modification of the evaluation processing of the meeting participant by the information processing apparatus which concerns on 1st Embodiment. 第１実施形態に係る情報処理装置による会議参加者の評価処理の第２変形例を示すフローチャートである。It is a flowchart which shows the 2nd modification of the evaluation processing of the meeting participant by the information processing apparatus which concerns on 1st Embodiment. 第１実施形態に係る情報処理装置による会議参加者の評価処理の第３変形例を示すフローチャートである。It is a flowchart which shows the 3rd modification of the evaluation processing of the meeting participant by the information processing apparatus which concerns on 1st Embodiment. 第１実施形態に係る情報処理装置による会議参加者の評価処理の第４変形例を示すフローチャートである。It is a flowchart which shows the 4th modification of the evaluation processing of the meeting participant by the information processing apparatus which concerns on 1st Embodiment. 第２実施形態に係る情報処理装置の内部構成の概略を示すブロック図である。It is a block diagram showing an outline of an internal configuration of an information processor concerning a 2nd embodiment. 第２実施形態に係る情報処理装置によるグループ再構成処理を示すフローチャートである。It is a flowchart which shows the group reconfiguration | reconstruction process by the information processing apparatus which concerns on 2nd Embodiment. 第２実施形態に係る情報処理装置によるグループ再構成処理の第１変形例を示すフローチャートである。It is a flowchart which shows the 1st modification of the group reconfiguration | reconstruction processing by the information processing apparatus which concerns on 2nd Embodiment. 第２実施形態に係る情報処理装置によるグループ再構成処理の第２変形例を示すフローチャートである。It is a flow chart which shows the 2nd modification of group reconstruction processing by an information processor concerning a 2nd embodiment. 第２実施形態に係る情報処理装置によるグループ再構成処理の第３変形例を示すフローチャートである。It is a flow chart which shows the 3rd modification of group reconstruction processing by an information processor concerning a 2nd embodiment. 第３実施形態に係る情報処理装置の内部構成の概略を示すブロック図である。It is a block diagram showing an outline of an internal configuration of an information processor concerning a 3rd embodiment. 第３実施形態に係る情報処理装置による評価処理を示すフローチャートである。It is a flowchart which shows the evaluation process by the information processing apparatus which concerns on 3rd Embodiment. 発言部分布図の位置系を示す図である。It is a figure which shows the position system of a speaking part distribution map. 第３実施形態に係る情報処理装置による評価処理の第１変形例を示すフローチャートである。It is a flowchart which shows the 1st modification of the evaluation process by the information processing apparatus which concerns on 3rd Embodiment. 第３実施形態に係る情報処理装置による評価処理の第２変形例を示すフローチャートである。It is a flow chart which shows the 2nd modification of the evaluation processing by the information processor concerning a 3rd embodiment.
　以下、本発明の一実施形態に係る情報処理装置、評価方法、及びグループ再構成方法について図面を参照して説明する。図１は、本発明の第１実施形態に係る情報処理装置、及び当該情報処理装置により評価が行われる対象者を示す図である。 Hereinafter, an information processing apparatus, an evaluation method, and a group reconfiguration method according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an information processing apparatus according to a first embodiment of the present invention, and a target person who is evaluated by the information processing apparatus.
　情報処理装置１は、予め定められた複数人からなる複数の会話グループＧ１～Ｇ３に属する各人が発した音声を音声データとして取得する。例えば、グループＧ１に属する人Ｐ１１，Ｐ１２，Ｐ１３、グループＧ２に属する人Ｐ２１，Ｐ２２，Ｐ２３、グループＧ３に属する人Ｐ３１，Ｐ３２，Ｐ３３からなる複数人（本実施形態では３つのグループに属する各人からなる合計９人として説明するが、これに限定されるものではない）が、グループ単位でミーティング、討論、授業、又は会議など（以下、総称して単に「会議」という。）を行っているものとする。 The information processing apparatus 1 acquires, as voice data, voices uttered by each person belonging to a plurality of predetermined conversation groups G1 to G3. For example, a plurality of persons P11, P12, P13 belonging to the group G1, persons P21, P22, P23 belonging to the group G2, persons P31, P32, P33 belonging to the group G3 (in this embodiment, each person belonging to three groups Although it is explained as a total of nine people consisting of, but not limited to), it is holding a meeting, a discussion, a class, a meeting, etc. (hereinafter referred to simply as "meeting") on a group basis. It shall be.
　当該会話グループの各人は、マイクロフォン機能を備えたヘッドセット２を使用した状態で発声する。すなわち、当該各人が用いるそれぞれのヘッドセット２は、ヘッドセット２を装着した人の会話の音声を取得し、当該音声を示す電気信号に変換して、情報処理装置１に出力する。情報処理装置１と、各ヘッドセット２とは、例えば、ケーブル接続による有線通信、又は、Bluetooth（登録商標）若しくは無線ＬＡＮ等の無線通信により接続されている。情報処理装置１は、各ヘッドセット２から出力されてくる上記音声を示す電気信号を、ディジタルの音声信号からなる音声データに変換し、ヘッドセット２毎、すなわち、上記人Ｐ１１～Ｐ３３の９人についての音声データをそれぞれに蓄積する。 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 the voice into an electrical signal indicating the voice, and outputs the electric signal to the information processing device 1. The information processing apparatus 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 electrical signal indicating the voice output from each headset 2 into voice data consisting of a digital voice signal, and for each headset 2, that is, nine people P11 to P33. The voice data about is stored in each.
　次に、第１実施形態に係る情報処理装置１の構成について説明する。図２は、第１実施形態に係る情報処理装置１の内部構成の概略を示すブロック図である。 Next, the configuration of the information processing apparatus 1 according to the first embodiment will be described. FIG. 2 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1 according to the first embodiment.
　情報処理装置１は、例えばコンピューターである。情報処理装置１は、制御ユニット１０と、ＲＯＭ（Read Only Memory）１１２と、ＲＡＭ（Random Access Memory）１１３と、ＨＤＤ（Hard Disk Drive）１１４と、表示部１１５と、通信インターフェイス１１８と、指示入力部１１９とを備える。これら各部は、互いにＣＰＵ（Central Processing Unit）バスによりデータ又は信号の送受信が可能とされている。 The information processing apparatus 1 is, for example, a computer. The information processing apparatus 1 includes a control unit 10, a read only memory (ROM) 112, a random access memory (RAM) 113, a hard disk drive (HDD) 114, a display unit 115, a communication interface 118, and an instruction input. And a unit 119. These units are capable of transmitting and receiving data or signals to each other via a central processing unit (CPU) bus.
　制御ユニット１０は、情報処理装置１全体の動作制御を司る。ＲＯＭ１１２は、情報処理装置１の基本動作についての動作プログラムを記憶する。ＲＡＭ１１３は、制御ユニット１０の動作領域等として使用される。 The control unit 10 controls the operation of the entire information processing apparatus 1. The ROM 112 stores an operation program for the basic operation of the information processing device 1. The RAM 113 is used as an operation area or the like of the control unit 10.
　ＨＤＤ１１４は、その記憶領域の一部に、本発明の第１実施形態に係る評価プログラムを記憶している。また、ＨＤＤ１１４は、上記人Ｐ１１～Ｐ３３の９人についてのそれぞれの上記音声データを記憶する。なお、ＨＤＤ１１４は、特許請求の範囲における記憶部の一例でもある。但し、情報処理装置１が備える不揮発性ＲＯＭ（例えば、制御ユニット１０が内蔵）が当該記憶部として機能してもよい。 The HDD 114 stores an evaluation program according to the first embodiment of the present invention in part of its storage area. Further, the HDD 114 stores the above-mentioned audio data of nine persons P11 to P33. The HDD 114 is also an example of a storage unit in the claims. However, a non-volatile ROM (for example, built in the control unit 10) included in the information processing apparatus 1 may function as the storage unit.
　情報処理装置１と接続されているヘッドセット２には、当該ヘッドセット２を特定するための識別情報が予め付されている。識別情報としては、ヘッドセット２を特定可能な情報であれば特に限定されないが、例えば、識別番号がある。ＨＤＤ１１４は、ヘッドセット２ごとに当該識別情報を予め記憶している。ＨＤＤ１１４はまた、指示入力部１１９を介してユーザーによって入力される指示に応じて、識別情報のそれぞれを、グループを特定するためのグループ情報と対応付けて記憶している。 Identification information for specifying the headset 2 is attached in advance to the headset 2 connected to the information processing apparatus 1. The identification information is not particularly limited as long as it is information that can identify the headset 2. For example, there is an identification number. The HDD 114 stores the identification information for each headset 2 in advance. The HDD 114 also stores each of the identification information in association with group information for specifying a group in accordance with an instruction input by the user via the instruction input unit 119.
　本実施の形態では、ＨＤＤ１１４は、人Ｐ２１，Ｐ２２，Ｐ２３がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ１を特定するグループ情報と対応付けて記憶している。ＨＤＤ１１４はまた、人Ｐ２１，Ｐ２２，Ｐ２３がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ２を特定するグループ情報と対応付けて記憶している。ＨＤＤ１１４はさらに、人Ｐ３１，Ｐ３２，Ｐ３３がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ３を特定するグループ情報と対応付けて記憶している。 In the present embodiment, the HDD 114 stores identification information of the headset 2 used by each of the persons P21, P22, and P23 in association with group information specifying the group G1. The HDD 114 also stores identification information of the headset 2 used by each of the persons P21, P22, and P23 in association with group information specifying the group G2. The HDD 114 further stores identification information of the headset 2 used by each of the persons P31, P32, and P33 in association with group information specifying the group G3.
　表示部１１５は、ＬＣＤ（Liquid Crystal Display）等からなり、情報処理装置１を操作する操作者に対する操作案内等が表示される。 The display unit 115 is formed of an LCD (Liquid Crystal Display) or the like, and displays operation guidance or the like for the operator who operates the information processing apparatus 1.
　通信インターフェイス１１８は、ＵＳＢインターフェイス又は無線ＬＡＮインターフェイス等を有する。通信インターフェイス１１８は、上記各ヘッドセット２とのデータ通信を行うためのインターフェイスとして機能する。通信インターフェイス１１８は、特許請求の範囲における音声入力部の一例である。 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 of the headsets 2. The communication interface 118 is an example of the voice input unit in the claims.
　指示入力部１１９は、キーボード又はマウス等から構成され、操作指示が操作者によって入力される。 The instruction input unit 119 includes a keyboard, a mouse, and the like, and the operator inputs an operation instruction.
　制御ユニット１０は、プロセッサー、ＲＡＭ、及びＲＯＭなどから構成される。プロセッサーは、ＣＰＵ、ＭＰＵ（Micro Processing Unit）、又はＡＳＩＣ（Application Specific Integrated Circuit）等である。この制御ユニット１０は、ＨＤＤ１１４に記憶されている評価プログラムが上記のプロセッサーで実行されることにより、制御部１００、第１検出部１０１、第１集計部１０２、第１算出部１０３、第２算出部１０４、第１付与部１０５、第３算出部１０６、第４算出部１０７、第２付与部１０８、判定部１０９、記憶制御部１１０、第３付与部１２１、テキスト化部１２２，加算付与部１２３、第４付与部１２４、第５算出部１２５、及び第６算出部１２６として機能する。なお、制御部１００、第１検出部１０１、第１集計部１０２、第１算出部１０３、第２算出部１０４、第１付与部１０５、第３算出部１０６、第４算出部１０７、第２付与部１０８、判定部１０９、記憶制御部１１０、第３付与部１２１、テキスト化部１２２，加算付与部１２３、第４付与部１２４、第５算出部１２５、及び第６算出部１２６は、それぞれハード回路により構成されてもよい。 The control unit 10 is configured of a processor, a RAM, a ROM, and the like. The processor is a CPU, a micro processing unit (MPU), an application specific integrated circuit (ASIC), or the like. In the control unit 10, the evaluation program stored in the HDD 114 is executed by the processor, whereby the control unit 100, the first detection unit 101, the first aggregation unit 102, the first calculation unit 103, and the second calculation are performed. Unit 104, first application unit 105, third calculation unit 106, fourth calculation unit 107, second application unit 108, determination unit 109, storage control unit 110, third application unit 121, textification unit 122, addition application unit It functions as 123, the fourth assignment unit 124, the fifth calculation unit 125, and the sixth calculation unit 126. The control unit 100, the first detection unit 101, the first counting unit 102, the first calculation unit 103, the second calculation unit 104, the first assignment unit 105, the third calculation unit 106, the fourth calculation unit 107, and the second The assignment unit 108, the determination unit 109, the storage control unit 110, the third assignment unit 121, the textification unit 122, the addition assignment unit 123, the fourth assignment unit 124, the fifth calculation unit 125, and the sixth calculation unit 126 respectively It may be configured by a hard circuit.
　制御部１００は、情報処理装置１全体の動作制御を司る機能を有する。 The control unit 100 has a function of controlling the operation control of the entire information processing apparatus 1.
　第１検出部（発言時間検出部）１０１は、ＨＤＤ１１４に記憶されている人Ｐ１１～Ｐ３３の９人についての音声データのそれぞれから、音声データに記録されている発話の中でされている各発言についての発言時間を検出する。図３は、音声データの一例を示す図である。図３の縦軸は音の振幅（単位dB）、横軸は時刻を示す。第１検出部１０１は、音声データを解析し、音声データが示す振幅のうち、予め定められた規定時間（例えば、0.25秒）以上連続して、予め定められた大きさ以上の振幅（例えば、20dB）が続く部分を発言として抽出する。第１検出部１０１は、当該抽出した各発言が続いている時間を発言時間として検出してＨＤＤ１１４に記憶しておく。図３に示す音声データでは、第１検出部１０１は、ａ部分、ｂ部分、及びｃ部分を発言として抽出する。 The first detection unit (speech time detection unit) 101 receives each speech in speech recorded in speech data from each of speech data of nine persons P11 to P33 stored in the HDD 114. Detect the speaking time about. FIG. 3 is a diagram showing an example of audio data. The vertical axis in FIG. 3 represents the amplitude of sound (in dB), and the horizontal axis represents time. The first detection unit 101 analyzes the audio data, and among the amplitudes indicated by the audio data, the amplitude (for example, a predetermined amplitude or more) continuously for a predetermined time (for example, 0.25 seconds) or more. Extract the part where 20 dB) continues as an utterance. The first detection unit 101 detects the time during which each of the extracted utterances is continuing as the utterance time and stores it in the HDD 114. In the voice data shown in FIG. 3, the first detection unit 101 extracts a part, b part, and c part as speech.
　第１集計部（各人別発言時間集計部）１０２は、第１検出部１０１によって検出された上記各発言についての発言時間を上記各人についてそれぞれ集計する。この場合、上記各発言についての発言時間は、ＨＤＤ１１４に記憶されている識別情報ごとに集計される。 A first counting unit (speech-based speech time counting unit) 102 counts, for each person, the speech time for each of the utterances detected by the first detection unit 101. In this case, the speech time for each of the above-mentioned speeches is summed up for each identification information stored in the HDD 114.
　第１算出部（合計発言時間算出部）１０３は、第１検出部１０１によって検出された全発言についての発言時間を合計して、上記全グループの各人全員の発言時間の合計時間を算出する。 The first calculating unit (total utterance time calculating unit) 103 adds up the utterance times of all the utterances detected by the first detection unit 101, and calculates the total time of the utterance times of all persons in all the groups. .
　第２算出部（各人別割合算出部）１０４は、第１算出部１０３によって算出された上記合計時間に対する、第１集計部１０２によって集計された上記各人の発言時間の割合を第１の割合として算出する。 The second calculating unit (the individual-by-person ratio calculating unit) 104 sets the ratio of the speaking time of each person counted by the first counting unit 102 to the total time calculated by the first calculating unit 103 as the first Calculated as a percentage.
　第１付与部（第１評価ポイント付与部）１０５は、第２算出部１０４によって算出された上記各人の発言時間の割合（第１の割合）の大きい人ほど評価ポイントを高く設定して、上記各人について当該第１の割合に応じた第１評価ポイントを付与する。 The first granting unit (first evaluation point granting unit) 105 sets evaluation points higher for people who have a larger ratio (first ratio) of the speaking time of each person calculated by the second calculating unit 104, A first evaluation point corresponding to the first ratio is given to each person.
　第３算出部（グループ別発言時間算出部）１０６は、第１検出部１０１によって検出された各発言についての発言時間を上記グループ毎に集計して、グループに属する各人の総発言時間をグループ毎に算出する。この場合、上記各発言についての発言時間は、ＨＤＤ１１４に記憶されているグループ情報ごとに集計される。 The third calculation unit (group-by-group utterance time calculation unit) 106 counts the utterance time for each utterance detected by the first detection unit 101 for each group, and the total utterance time of each person belonging to the group is grouped Calculate every time. In this case, the speech time for each of the above-mentioned speeches is counted for each group information stored in the HDD 114.
　第４算出部（グループ内割合算出部）１０７は、第３算出部１０６によって算出された上記グループ毎の上記総発言時間に基づいて、第１集計部１０２によって検出された各人の発言時間についての、当該各人が所属するグループの上記総発言時間に対する割合であるグループ内割合を第２の割合として算出する。 The fourth calculating unit (in-group ratio calculating unit) 107 calculates the speaking time of each person detected by the first counting unit 102 based on the total speaking time for each group calculated by the third calculating unit 106. The ratio in the group which is the ratio to the above-mentioned total utterance time of the group to which each person concerned belongs is calculated as the second ratio.
　第２付与部（第２評価ポイント付与部）１０８は、第４算出部１０７によって算出された上記各人のグループ内割合（第２の割合）の大きい人ほど評価ポイントを高く設定して、上記各人についてグループ内割合に応じた第２評価ポイントを更に付与する。 The second granting unit (second evaluation point granting unit) 108 sets evaluation points higher for people with a larger ratio (second ratio) of the above-described persons calculated by the fourth calculating unit 107, and Each person is further awarded with a second evaluation point according to the in-group ratio.
　判定部（相槌・意見判定部）１０９は、第１検出部１０１によって検出された上記発言時間が、予め定められた第１の時間（上記規定時間から、当該規定時間よりも長い予め定められた時間までの時間。例えば、上記規定時間としての0.25秒以上であって2.0秒までの時間）内であれば、この発言時間でされた発言を、相槌と判定する（図３の例ではｂ）。また、判定部１０９は、第１検出部１０１によって検出された上記発言時間が、上記第１の時間よりも長い予め定められた第２の時間（上記第１の時間を超える時間）であれば、この発言時間でされた発言を、意見と判定する（図３の例ではａ，ｃ）。判定部１０９は、当該判定した相槌か意見かの結果と、当該相槌及び意見の各発言時間とをＨＤＤ１１４に記憶させる。 The determination unit (completion / opposition determination unit) 109 determines that the utterance time detected by the first detection unit 101 is a predetermined first time (predetermined time longer than the prescribed time from the prescribed time). If the time to time is within, for example, 0.25 seconds or more and 2.0 seconds as the above-mentioned prescribed time), the utterance made at this utterance time is determined to be a compliment (b in the example of FIG. 3) . In addition, if the utterance time detected by the first detection unit 101 is the second predetermined time (time exceeding the first time), the determination unit 109 determines that the utterance time detected by the first detection unit 101 is longer than the first time. The utterance made at this utterance time is judged as an opinion (a, c in the example of FIG. 3). The determination unit 109 causes the HDD 114 to store the result as to whether or not the determined combination or opinion, and each utterance time of the combination and opinion.
　記憶制御部（結果記憶部）１１０は、判定部１０９によって判定された結果、すなわち、発言が相槌又は意見のいずれであるかの結果を人Ｐ１１～Ｐ３３の各別にＨＤＤ１１４に記憶させる。 The storage control unit (result storage unit) 110 causes the HDD 114 to store the result of the determination made by the determination unit 109, that is, the result of whether the remark is a compliment or an opinion separately for each of the persons P11 to P33.
　第３付与部（第３評価ポイント付与部）１２１は、記憶制御部１１０によって記憶されている人Ｐ１１～Ｐ３３のそれぞれの結果に基づいて、ある人による上記意見がされた直後のタイミングで、他の人により上記相槌がされていると判断したときは、第３評価ポイントを、当該意見をした人に更に付与する。 The third granting unit (third evaluation point granting unit) 121 is based on the result of each of the persons P11 to P33 stored by the storage control unit 110, at another timing immediately after the opinion is given by the other person. If it is determined that the person has made the above-mentioned reciprocation, the third evaluation point is further given to the person who made the opinion.
　テキスト化部１２２は、既知の音声認識機能を備え、これにより、上記音声データに含まれる上記各人の発言の内容をキャラクター変換してテキスト化する。 The text conversion unit 122 has a known voice recognition function, and thereby converts the contents of the speech of each person included in the voice data into characters.
　加算付与部（加算ポイント付与部）１２３は、テキスト化部１２２によってテキスト化されたテキストデータに基づいて、上記各人Ｐ１１～Ｐ３３による発言が予め定められたキーワードを含むか否かを判断し、当該キーワードを含むと判断した発言に加算ポイントを付与する。 The addition giving unit (additional point giving unit) 123 determines, based on the text data made into text by the textification unit 122, whether or not the remarks by each of the persons P11 to P33 include a predetermined keyword, An additional point is given to an utterance judged to contain the keyword.
　第４付与部（第４評価ポイント付与部）１２４は、加算付与部１２３によって付与された上記発言に対する加算ポイントを上記各人についてそれぞれ集計し、当該加算ポイントの集計値を第４評価ポイントとして、当該集計の対象とされている人に対して更に付与する。 The fourth giving unit (fourth evaluation point giving unit) 124 adds up the addition points for the message given by the addition giving unit 123 for each person, and sets the total value of the addition points as a fourth evaluation point, Further grant to those who are targeted for the calculation.
　第５算出部（合計値算出部）１２５は、加算付与部１２３によって付与された上記加算ポイントを用いて、上記複数のグループＧ１～Ｇ３に属する各人全員分の加算ポイントの合計値を算出する。なお、第４付与部１２４は、第５算出部１２５によって算出された合計値に対する、第４評価ポイントの割合（第３の割合）を上記各人Ｐ１１～Ｐ３３のそれぞれについて算出し、当該割合が高いほど第４評価ポイントを増加させる。 The fifth calculation unit (total value calculation unit) 125 calculates the total value of the addition points for all persons belonging to the plurality of groups G1 to G3 using the addition points given by the addition giving unit 123. . In addition, the fourth provision unit 124 calculates the ratio (third ratio) of the fourth evaluation points to the total value calculated by the fifth calculation unit 125 for each of the persons P11 to P33, and the ratio is The higher it is, the 4th evaluation point is increased.
　第６算出部（グループ別ポイント算出部）１２６は、加算付与部１２３によって付与された上記加算ポイントを、上記グループＧ１～Ｇ３のそれぞれ毎に、グループに属する各人の合計値を算出する。なお、第４付与部１２４は、第６算出部１２６によって算出されたグループの合計値に対する、当該グループに属する上記各人の第４評価ポイントの割合（第４の割合）を上記各人のそれぞれについて算出し、当該割合が高いほど第４評価ポイントを増加させる。 The sixth calculation unit (group-specific point calculation unit) 126 calculates, for each of the groups G1 to G3, the sum value of each person belonging to the group, for the addition points given by the addition application unit 123. In addition, the fourth assignment unit 124 sets the ratio (fourth ratio) of the fourth evaluation point of each person belonging to the group to the total value of the group calculated by the sixth calculation unit 126, for each of the above persons. The fourth evaluation point is increased as the ratio is higher.
　次に、第１実施形態に係る情報処理装置１による会議参加者の評価処理について説明する。図４は、情報処理装置１による会議参加者の評価処理を示すフローチャートである。 Next, an evaluation process of a conference participant by the information processing apparatus 1 according to the first embodiment will be described. FIG. 4 is a flowchart showing an evaluation process of a conference participant by the information processing apparatus 1.
　当該評価が行われる場面は、会話グループＧ１～Ｇ３に属する各人が、それぞれのグループ毎に会議を行っている場面である。人Ｐ１１～Ｐ３３は、各自がヘッドセット２を装着し、各ヘッドセット２は上述したように情報処理装置１に対して通信可能に接続されている。この状態で、人Ｐ１１～Ｐ３３は、各自が属するそれぞれのグループにおいて会議中に発話を行う。人Ｐ１１～Ｐ３３によって発話された音声は、人Ｐ１１～Ｐ３３のそれぞれのヘッドセット２により集音され、情報処理装置１に出力される。 The scene in which the evaluation is performed is a scene in which each person belonging to the conversation groups G1 to G3 is holding a meeting for each group. Each of the persons P11 to P33 wears the headset 2, and the headsets 2 are communicably connected to the information processing apparatus 1 as described above. In this state, the people P11 to P33 speak during the meeting in the respective groups to which they belong. The voice uttered by the people P11 to P33 is collected by the headset 2 of each of the people P11 to P33, and is output to the information processing device 1.
　情報処理装置１は、通信インターフェイス１１８を介して、各ヘッドセット２から音声データを取得する（ステップＳ１）。すなわち、通信インターフェイス１１８が、各ヘッドセット２から出力されてくる上記音声を示す電気信号を受信すると、第１検出部１０１は、当該取得された音声を示す電気信号を、ディジタルの音声信号からなる音声データに変換してＨＤＤ１１４に記憶させる。第１検出部１０１は、人Ｐ１１～Ｐ３３毎に、すなわち、ＨＤＤ１１４に記憶されている識別情報に対応付けて、当該音声データをＨＤＤ１１４に記憶させる。 The information processing apparatus 1 acquires audio data from each headset 2 via the communication interface 118 (step S1). That is, when the communication interface 118 receives an electrical signal indicating the voice output from each headset 2, the first detection unit 101 includes the electrical signal indicating the acquired voice from a digital voice signal. It is converted into voice data and stored in the HDD 114. The first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
　続いて、第１検出部１０１は、人Ｐ１１～Ｐ３３毎にＨＤＤ１１４に記憶されている音声データのそれぞれから、当該音声データが示す発話の中でされている各発言を、上述したようにして抽出する（ステップＳ２）。そして、第１検出部１０１は、当該抽出した各発言の発言時間を検出する（ステップＳ３）。 Subsequently, the first detection unit 101 extracts, as described above, the utterances in the speech indicated by the voice data from the voice data stored in the HDD 114 for each of the persons P11 to P33. (Step S2). Then, the first detection unit 101 detects the utterance time of each of the extracted utterances (step S3).
　更に、第１集計部１０２は、第１検出部１０１によって検出された上記各発言についての発言時間を上記人Ｐ１１～Ｐ３３のそれぞれについて個別に人毎に集計する（ステップＳ４）。 Furthermore, the first counting unit 102 counts, for each of the persons P11 to P33, the utterance times of the respective utterances detected by the first detection unit 101 individually for each of the persons P11 to P33 (step S4).
　続いて、第１算出部１０３は、例えば、第１検出部１０１によって人ごとに検出された発言の全てについての発言時間を合計して、上記各人全員の発言時間の合計時間を算出する（ステップＳ５）。 Subsequently, for example, the first calculation unit 103 sums up the utterance times for all the utterances detected for each person by the first detection unit 101, and calculates the total time of the utterance times of all the persons described above ( Step S5).
　第２算出部１０４は、第１算出部１０３によって算出された上記合計時間に対する、第１集計部１０２によって集計された上記各人の発言時間の割合を第１の割合として算出する（ステップＳ６）。すなわち、第２算出部１０４は、上記合計時間に対する人Ｐ１１の発言時間の割合、上記合計時間に対する人Ｐ１２の発言時間の割合…、のように、人Ｐ３３までの割合を個別に算出する。 The second calculating unit 104 calculates, as a first ratio, the ratio of the speaking time of each person, which is counted by the first counting unit 102, to the total time calculated by the first calculating unit 103 (step S6). . That is, the second calculation unit 104 individually calculates the ratio to the person P33, such as the ratio of the speech time of the person P11 to the total time, the ratio of the speech time of the person P12 to the total time, and so on.
　続いて、第１付与部１０５は、第２算出部１０４によって算出された上記各人の発言時間の割合の大きい人ほど評価ポイントを高く設定して、上記人Ｐ１１～Ｐ３３の各人について、当該第１の割合に応じた第１評価ポイントを付与する（ステップＳ７）。例えば、第１付与部１０５は、当該第１の割合が０～２０％未満の場合は２ポイント、２０％以上４０％未満の場合は４ポイント、４０％以上６０％未満の場合は６ポイント、６０％以上８０％未満の場合は８ポイント、８０％以上１００％までの場合は１０ポイントを第１評価ポイントとして付与する。 Subsequently, the first assignment unit 105 sets the evaluation point higher for the person with the larger proportion of the speaking time of each person calculated by the second calculation unit 104, and the first grant unit 105 sets the evaluation point for each person P11 to P33. A first evaluation point is given according to the first ratio (step S7). For example, the first giving unit 105 has 2 points when the first ratio is 0 to less than 20%, 4 points for 20% or more and less than 40%, 6 points for 40% or more and less than 60%, In the case of 60% or more and less than 80%, 8 points are awarded, and in the case of 80% or more and 100%, 10 points are awarded as first evaluation points.
　更に、第３算出部１０６は、ステップＳ３で第１検出部１０１によって検出された各発言についての発言時間を、グループＧ１～Ｇ３毎、すなわち、ＨＤＤ１１４に記憶されているグループ情報ごとに集計して、グループＧ１における各人の総発言時間と、グループＧ２における各人の総発言時間と、グループＧ３における各人の総発言時間とを算出する（ステップＳ８）。 Furthermore, the third calculation unit 106 counts the utterance time for each utterance detected by the first detection unit 101 in step S3 for each of the groups G1 to G3, that is, for each group information stored in the HDD 114. The total speech time of each person in group G1, the total speech time of each person in group G2, and the total speech time of each person in group G3 are calculated (step S8).
　そして、第４算出部１０７は、ステップＳ８で第３算出部１０６によって算出された上記グループ毎の総発言時間に対する、ステップＳ４で第１集計部１０２によって検出された人Ｐ１１～Ｐ３３のそれぞれの発言時間のグループ内割合を第２の割合として算出する（ステップＳ９）。すなわち、第４算出部１０７は、人Ｐ１１～Ｐ３３のそれぞれについて、各人が所属するグループの上記総発言時間に対する割合を、グループ内割合として算出する。 Then, the fourth calculating unit 107 calculates the utterances of the persons P11 to P33 detected by the first counting unit 102 in step S4 with respect to the total utterance time for each group calculated by the third calculating unit 106 in step S8. The intra-group ratio of time is calculated as the second ratio (step S9). That is, the fourth calculating unit 107 calculates, for each of the persons P11 to P33, the ratio of the group to which each person belongs to the total speech time as the ratio within the group.
　続いて、第２付与部１０８は、ステップＳ９で第４算出部１０７によって算出された上記各人のグループ内割合の大きい人ほど評価ポイントを高く設定して、人Ｐ１１～Ｐ３３のそれぞれについてのグループ内割合に応じた第２評価ポイントを更に付与する（ステップＳ１０）。 Subsequently, the second giving unit 108 sets evaluation points to be higher as the ratio of the in-group ratio of each person calculated by the fourth calculating unit 107 in step S9 increases, and the group for each of the persons P11 to P33 is A second evaluation point according to the inside ratio is further awarded (step S10).
　例えば、第２付与部１０８は、当該第２の割合が０～２０％未満の場合は４ポイント、２０％以上４０％未満の場合は８ポイント、４０％以上６０％未満の場合は１２ポイント、６０％以上８０％未満の場合は１６ポイント、８０％以上１００％までの場合は２０ポイントを第２評価ポイントとして付与する。 For example, the second granting unit 108 is 4 points when the second ratio is 0 to less than 20%, 8 points when 20% or more and less than 40%, 12 points when 40% or more and less than 60%. In the case of 60% or more and less than 80%, 16 points are awarded, and in the case of 80% or more and 100%, 20 points are awarded as the second evaluation points.
　第２付与部１０８は、このように、第１付与部１０５が付与する第１評価ポイントと比較して高いポイント（本実施形態では第２評価ポイントは第１評価ポイントの２倍）を第２評価ポイントとして付与することが好ましい。各人による発言時間は、個別のグループ内の総発言時間に対する割合が、人Ｐ１１～Ｐ３３の全員の発言時間の合計である合計時間に対する割合よりも高い方が、当該会議における貢献度が高いと思われるためである。 Thus, the second provision unit 108 sets the second evaluation point higher than the first evaluation point provided by the first application unit 105 (in the present embodiment, the second evaluation point is twice the first evaluation point). It is preferable to give it as an evaluation point. If the ratio of the speech time by each person to the total speech time in each group is higher than the ratio to the total time, which is the sum of the speech times of all the persons P11 to P33, the contribution in the meeting is high. It is because it seems.
　複数人数からなる複数のグループＧ１～Ｇ３により、各グループ別に会議を行う場合、各グループに属する人それぞれの個性によって、グループ内での討論が活発になることもあれば、活発にならないこともある。このため、あるグループ内での発言時間が長い（貢献度が高い）人が、他のグループに属した場合にも、当該他のグループで同様に多く発言するか（高い貢献度を示すか）は不明である。 When a conference is held separately for each group by a plurality of groups G1 to G3 consisting of a plurality of people, the discussion within the group may or may not be activated depending on the individuality of each person belonging to each group . Therefore, if a person who has a long talk time (high contribution) in one group belongs to another group, may he / she speak a lot in the other group (does it indicate a high contribution)? Is unknown.
　このため、グループ内における各人の発話時間の合計に対する、当該グループ内の各メンバーの発話時間の割合に基づくのみで、会議に対する貢献度を評価すると、グループＧ１～Ｇ３の全てを通じた人Ｐ１１～Ｐ３３全員における全体的な貢献度が不明となる。 Therefore, when the degree of contribution to the conference is evaluated based only on the ratio of the speech time of each member in the group to the total of the speech time of each person in the group, people P11 to P11 through all the groups G1 to G3 P33 The overall contribution of all members is unknown.
　本実施形態では、ステップＳ７において、第１付与部１０５が、グループＧ１～Ｇ３の全てを通じた人Ｐ１１～Ｐ３３全員による発言時間の合計時間に対する、各人の発言時間の割合に応じた第１評価ポイントを各人に対して付与する一方、ステップＳ１０においては、第２付与部１０８が、グループ毎の総発言時間に対する各人の発言時間の割合であるグループ内割合に応じた第２評価ポイントを更に付与するため、グループ内における各人の貢献度と、グループＧ１～Ｇ３の全てを通じた人Ｐ１１～Ｐ３３全員における全体的な貢献度との両方を加味した総合的な評価が可能になる。 In the present embodiment, in step S7, the first granting unit 105 performs a first evaluation according to the ratio of the speaking time of each person to the total speaking time of all the persons P11 to P33 who have passed through all the groups G1 to G3. While giving points to each person, in step S10, the second giving unit 108 sets the second evaluation point according to the ratio within the group which is the ratio of the speaking time of each person to the total speaking time for each group. Further, it is possible to make a comprehensive evaluation taking into consideration both the contribution of each person in the group and the overall contribution of all of the persons P11 to P33 through all the groups G1 to G3.
　なお、制御部１００は、指示入力部１１９を介してユーザーによって入力される指示に応じて、各人について付与されている第１評価ポイント及び第２評価ポイントを示す情報を、表示部１１５に表示可能に構成されていてもよい。 Note that the control unit 100 displays, on the display unit 115, information indicating the first evaluation point and the second evaluation point given to each person in accordance with the instruction input by the user via the instruction input unit 119. It may be configured as possible.
　次に、情報処理装置１による会議参加者の評価処理の第１変形例について説明する。図５は、情報処理装置１による会議参加者の評価処理の第１変形例を示すフローチャートである。なお、当該第１変形例の説明では、第１実施形態と同様の処理については説明を省略する。 Next, a first modified example of the evaluation process of the conference participant by the information processing device 1 will be described. FIG. 5 is a flowchart showing a first modified example of the evaluation processing of the conference participant by the information processing device 1. In the description of the first modification, the description of the processing similar to that of the first embodiment is omitted.
　第１変形例では、第１実施形態と同様に、ステップＳ１～ステップＳ１０までの処理を行った後、更に、判定部１０９が、上記各発言の発言時間が上記第１の時間内であるか第２の時間であるかを判断する（ステップＳ１１）。判定部１０９は、発言時間が第１の時間内であれば（ステップＳ１１で「第１の時間」）、この発言時間でされた発言を相槌と判定する（ステップＳ１２）。また、判定部１０９は、上記発言時間が上記第２の時間であれば（ステップＳ１１で「第２の時間」）、この発言時間でされた発言を意見と判定する（ステップＳ１６）。 In the first modification, as in the first embodiment, after performing the processing from step S1 to step S10, whether the speaking time of each statement is within the first time after the determination unit 109 is further performed. It is determined whether it is the second time (step S11). If the speech time is within the first time ("first time" in step S11), the determination unit 109 determines that the speech made in this speech time is a reunion (step S12). In addition, when the utterance time is the second time (“second time” in step S11), the determination unit 109 determines that the utterance made in the utterance time is an opinion (step S16).
　更に、記憶制御部１１０は、ステップＳ１２及びステップＳ１６で判定部１０９によって判定された結果、すなわち、発言が相槌又は意見のいずれであるかの結果を、当該相槌又は意見を示す発言がされた時刻と共に、人Ｐ１１～Ｐ３３の各別にＨＤＤ１１４に記憶させる（ステップＳ１３）。 Furthermore, the storage control unit 110 determines the result of the determination made by the determination unit 109 in step S12 and step S16, that is, the result of whether the statement is a summary or an opinion, the time at which the statement indicating the summary or opinion is made And each of the persons P11 to P33 is stored in the HDD 114 (step S13).
　続いて、第３付与部１２１は、記憶制御部１１０によって記憶されている人Ｐ１１～Ｐ３３のそれぞれの結果に基づいて、グループＧ１～Ｇ３のグループ毎に、グループ内において或る人による上記意見がされた直後のタイミングで、他の人により上記相槌がされているかを判断する（ステップＳ１４）。第３付与部１２１は、このような意見があると判断したときは（ステップＳ１４でＹＥＳ）、第３評価ポイントを、当該意見をした人に更に付与する（ステップＳ１５）。例えば、第３付与部１２１は、上記第３評価ポイントとして、１０ポイントを付与する。なお、第３付与部１２１は、このような意見がないと判断したときは（ステップＳ１４でＮＯ）、第３評価ポイントを付与しない。 Subsequently, the third assignment unit 121, based on the result of each of the persons P11 to P33 stored by the storage control unit 110, for each of the groups G1 to G3, the opinion given by a person in the group is At the timing immediately after being done, it is determined whether the above-mentioned sumo wrestling is being performed by another person (step S14). When it is determined that there is such an opinion (YES in step S14), the third giving unit 121 further gives the third evaluation point to the person who made the opinion (step S15). For example, the third applying unit 121 applies 10 points as the third evaluation point. When it is determined that there is no such opinion (NO in step S14), the third giving unit 121 does not give the third evaluation point.
　この第１変形例によれば、意見の直後に連続して他の人により相槌がされており、他の人の関心を惹き付けた良質な意見であると想定される意見等については、このような意見に、当該相槌がされなかった他の意見よりも高い評価が与えられることになるため、良質と想定される意見に対して高評価を的確に与えることが可能になる。 According to the first modification, the opinion, etc. assumed to be a good opinion which is being followed up by another person immediately after the opinion and which has attracted the interest of the other person Since such an opinion will be given a higher evaluation than other opinions that have not been countered, it is possible to appropriately give a high evaluation to an opinion that is supposed to be good.
　次に、情報処理装置１による会議参加者の評価処理の第２変形例について説明する。図６は情報処理装置１による会議参加者の評価処理の第２変形例を示すフローチャートである。なお、当該第２変形例の説明では、第１実施形態及び第１変形例と同様の処理については説明を省略する。 Next, a second modified example of the process of evaluating the meeting participants by the information processing device 1 will be described. FIG. 6 is a flow chart showing a second modified example of the evaluation processing of the conference participant by the information processing device 1. In the description of the second modification, the description of the same processes as those of the first embodiment and the first modification will be omitted.
　第２変形例は、第１実施形態におけるステップＳ１～ステップＳ１０の後、又は第１変形例におけるステップＳ１１～ステップＳ１６の後に行われる。 The second modification is performed after steps S1 to S10 in the first embodiment or after steps S11 to S16 in the first modification.
　テキスト化部１２２は、上記音声データに含まれる上記各人の発言の内容をキャラクター変換してテキスト化する（ステップＳ２０）。 The text conversion unit 122 converts the contents of the speech of each person included in the voice data into characters and converts them into text (step S20).
　加算付与部１２３は、テキスト化部１２２によってテキスト化されたテキストデータに基づいて、当該テキストデータに含まれる上記各人Ｐ１１～Ｐ３３による発言が予め定められたキーワードを含むか否かを判断する（ステップＳ２１）。 Based on the text data converted into text by the text conversion unit 122, the addition application unit 123 determines whether the utterance by each of the persons P11 to P33 included in the text data includes a predetermined keyword (refer to FIG. Step S21).
　加算付与部１２３が、上記テキストデータに含まれる上記各人Ｐ１１～Ｐ３３による発言が予め定められたキーワードを含むと判断したとき（ステップＳ２１でＹＥＳ）、加算付与部１２３は、当該キーワードを含むと判断した発言に加算ポイントを付与する（ステップＳ２２）。上記テキストデータには、各人による複数の発言が含まれるが、加算付与部１２３が、当該全ての発言に対して、上記キーワードを含むか否かの判断を行い、加算付与部１２３は、当該キーワードを含むと判断された全ての発言に加算ポイントを付与する。 When the addition giving unit 123 determines that the utterance by each of the persons P11 to P33 included in the text data contains a predetermined keyword (YES in step S21), the addition giving unit 123 includes the keyword. An addition point is added to the determined speech (step S22). Although the text data includes a plurality of utterances by each person, the addition imparting unit 123 determines whether all the utterances include the keyword, and the addition imparting unit 123 Add points to all messages judged to contain a keyword.
　なお、加算付与部１２３が、上記各人Ｐ１１～Ｐ３３による発言が予め定められたキーワードを含まないと判断したとき（ステップＳ２１でＮＯ）、加算付与部１２３は、当該加算ポイントを付与しない。 When the addition giving unit 123 determines that the remarks by each of the persons P11 to P33 do not include a predetermined keyword (NO in step S21), the addition giving unit 123 does not give the addition points.
　この後、第４付与部１２４は、ステップＳ２２で加算付与部１２３によって付与された上記各発言に対する加算ポイントを、人Ｐ１１～Ｐ３３の各人についてそれぞれ集計し（ステップＳ２３）、当該加算ポイントの集計値を第４評価ポイントとして、当該集計の対象とされている各人に対して更に付与する（ステップＳ２４）。 Thereafter, the fourth adding unit 124 adds up the added points for each of the above-mentioned statements added by the adding up unit 123 in step S22 for each of the persons P11 to P33 (step S23), and adds up the added points. A value is further assigned as a fourth evaluation point to each person targeted for the aggregation (step S24).
　例えば、加算付与部１２３は、各発言に対して、これに含まれる上記キーワードが出現する度に上記加算ポイント（例えば、１ポイント）を付与し、人Ｐ１１～Ｐ３３の人毎にそれぞれ集計する。そして、加算付与部１２３は、上記キーワード含む発言を行った人に対して、各人についての当該集計値を第４評価ポイントとして付与する。 For example, the addition giving unit 123 gives the above-mentioned addition points (for example, 1 point) to the respective utterances each time the above-mentioned keyword included in the utterance appears, and adds up each of the persons P11 to P33. Then, the addition giving unit 123 gives the total value of each person as the fourth evaluation point to the person who made the statement including the keyword.
　この第２変形例によれば、会議の主催者等が、貢献度が高いとして想定していたキーワードを発言した人に対して、上記第１乃至第３評価ポイントに加えて更に第４評価ポイントが付与されるので、当該ポイントの合計値に基づけば、この会議に真に貢献している人Ｐ１１～Ｐ３３を正確に判定することが可能になる。 According to the second modified example, the fourth evaluation point is added to the above first to third evaluation points for the person who has made the keyword for which the organizer of the meeting or the like assumed that the degree of contribution is high. Therefore, based on the total value of the points, it is possible to accurately determine the persons P11 to P33 who are truly contributing to the conference.
　次に、情報処理装置１による会議参加者の評価処理の第３変形例について説明する。図７は情報処理装置１による会議参加者の評価処理の第３変形例を示すフローチャートである。なお、当該第３変形例の説明では、第１実施形態、第１変形例、及び第２変形例と同様の処理については説明を省略する。 Next, a third modified example of the evaluation processing of the conference participant by the information processing device 1 will be described. FIG. 7 is a flow chart showing a third modification of the evaluation processing of the conference participant by the information processing apparatus 1. In the description of the third modification, the description of the same processes as those of the first embodiment, the first modification, and the second modification will be omitted.
　第３変形例では、第２変形例と同様に、第４付与部１２４は、ステップＳ２２で加算付与部１２３によって付与された上記各発言に対する加算ポイントを、人Ｐ１１～Ｐ３３の各人についてそれぞれ集計した後（ステップＳ２３）、更に、第５算出部１２５が、加算付与部１２３によって付与された上記人毎の加算ポイントを用いて、複数のグループに属する各人全員分の加算ポイントの合計値を算出する（ステップＳ３１）。 In the third modification, as in the second modification, the fourth adding unit 124 adds up the addition points for each of the utterances added by the addition adding unit 123 in step S22 for each of the persons P11 to P33. After that (step S23), the fifth calculation unit 125 further uses the addition points for each person given by the addition giving unit 123 to calculate the total value of the addition points for all persons belonging to a plurality of groups. It calculates (step S31).
　続いて、第４付与部１２４は、第５算出部１２５によって算出された合計値に対する、各人に対する加算ポイントの集計値の割合を算出し（ステップＳ３２）、当該算出した割合が高い人ほど、第４評価ポイントを高く設定して、対象となる人に付与する（ステップＳ３３）。 Subsequently, the fourth adding unit 124 calculates the ratio of the total value of added points to each person to the total value calculated by the fifth calculating unit 125 (step S32), and the higher the calculated ratio, the more The fourth evaluation point is set high and given to the target person (step S33).
　例えば、第４付与部１２４は、当該割合が０～２０％未満の場合は集計値の２０％を加算、２０％以上４０％未満の場合は集計値の４０％を加算、４０％以上６０％未満の場合は集計値の６０％を加算、６０％以上８０％未満の場合は集計値の８０％を加算、８０％以上１００％までの場合は集計値の１００％を加算とし、当該加算後の集計値を第４評価ポイントとして付与する。 For example, the fourth giving unit 124 adds 20% of the total value when the ratio is less than 0 to 20%, adds 40% of the total value when 20% or more and less than 40%, 40% or more and 60% In the case of less than 60% of the aggregate value is added, in the case of 60% or more and less than 80%, 80% of the aggregate value is added, and in the case of 80% or more to 100%, 100% of the aggregate value is added. The aggregate value of is given as the fourth evaluation point.
　この第３変形例によれば、上記キーワードを発言した人の発言度合をグループ内におけるキーワード発言状況を勘案して客観的な要素を入れた上で、キーワードを発言した人を評価することが可能になる。 According to the third modification, it is possible to evaluate the person who has made a comment on the keyword after adding an objective element in consideration of the state of the user who made a statement in the group and the degree of speech of the person who made the above keyword become.
　次に、情報処理装置１による会議参加者の評価処理の第４変形例について説明する。図８は情報処理装置１による会議参加者の評価処理の第４変形例を示すフローチャートである。なお、当該第４変形例の説明では、第１実施形態、及び第１乃至第３変形例と同様の処理については説明を省略する。 Next, a fourth modified example of the process of evaluating a conference participant by the information processing device 1 will be described. FIG. 8 is a flow chart showing a fourth modification of the evaluation processing of the conference participant by the information processing apparatus 1. In the description of the fourth modification, the description of the same processes as those of the first embodiment and the first to third modifications will be omitted.
　第４変形例は、第３変形例の後に行われる。第３変形例と同様に、第４評価ポイントを高く設定して対象となる人に付与した後（ステップＳ３３）、第６算出部１２６は、グループＧ１～Ｇ３のそれぞれのグループ単位で、グループに属する各人の加算ポイントを集計して合計値を算出する（ステップＳ３４）。 The fourth modification is performed after the third modification. As in the third modification, after setting the fourth evaluation point high and giving it to the target person (step S33), the sixth calculator 126 adds the fourth evaluation point to each of the groups G1 to G3 as a group. The total points are calculated by totaling addition points of each person who belongs (step S34).
　そして、第４付与部１２４は、第６算出部１２６によって算出された合計値に対する、各人に対する加算ポイントの集計値の割合をそれぞれに算出し（ステップＳ３５）、当該算出した割合が高い人ほど、第４評価ポイントを高く設定して、対象となる人に更に付与する（ステップＳ３６）。 Then, the fourth assignment unit 124 calculates the ratio of the total value of the added points to each person to the total value calculated by the sixth calculation unit 126 (step S35), and the person with the calculated ratio is higher The fourth evaluation point is set high, and is further given to the target person (step S36).
　例えば、第４付与部１２４は、当該割合が０～２０％未満の場合は集計値の１０％を加算、２０％以上４０％未満の場合は集計値の２０％を加算、４０％以上６０％未満の場合は集計値の３０％を加算、６０％以上８０％未満の場合は集計値の４０％を加算、８０％以上１００％までの場合は集計値の５０％を加算とし、当該加算後の集計値を第４評価ポイントとして付与する。 For example, the fourth giving unit 124 adds 10% of the total value when the ratio is less than 0 to 20%, adds 20% of the total value when 20% or more and less than 40%, 40% or more and 60% In the case of less than 30% of the aggregate value is added, in the case of 60% or more and less than 80%, 40% of the aggregate value is added, and in the case of 80% or more to 100%, 50% of the aggregate value is added. The aggregate value of is given as the fourth evaluation point.
　第４付与部１２４は、グループ全体の合計値に対する割合（第３の割合）に対する第４評価ポイントを、グループ単位での合計値に対する割合（第４の割合）に対する第４評価ポイントよりも高く設定して第４評価ポイントを付与することが好ましい。各人による上記キーワードの出現回数の割合は、個別のグループ内の出現回数の合計に対する割合よりも、全グループにおける人Ｐ１１～Ｐ３３の全員の出現回数の合計に対する割合が高い方が、当該会議における貢献度が高いと思われるためである。 The fourth assignment unit 124 sets the fourth evaluation point for the ratio (third ratio) to the total value of the entire group higher than the fourth evaluation point for the ratio (fourth ratio) to the total value in group units. It is preferable to give a fourth evaluation point. The ratio of the number of occurrences of the above keyword by each person is the one in which the ratio of the number of occurrences of all the persons P11 to P33 in all groups is higher than the ratio to the total of the number of appearances in the individual group It is because it seems that the degree of contribution is high.
　この第４変形例によれば、上記キーワードを発言した人の発言度合を、グループ内における発言状況に加えて、グループ全体の全員によるキーワード発言状況を勘案して、より客観的な要素を入れた上で、キーワードを発言した人を評価することが可能になる。 According to the fourth modified example, in addition to the situation of speech in the group, the degree of speech of the person who has said the keyword is more objective in consideration of the state of speech of the keyword by all members of the whole group. Above, it will be possible to evaluate the person who said the keyword.
　次に、第２実施形態に係る情報処理装置、及びグループ再構成方法について図面を参照して説明する。なお、第１実施形態に係る情報処理装置等と同様の構成及び処理については説明を省略する。 Next, an information processing apparatus and a group reconfiguring method according to the second embodiment will be described with reference to the drawings. Descriptions of configurations and processes similar to those of the information processing apparatus and the like according to the first embodiment will be omitted.
　第２実施形態に係る情報処理装置１Ａは、第１実施形態に係る情報処理装置１と同様にして、会話グループＧ１～Ｇ３に属する各人が発した音声を音声データとして取得する。 Similar to the information processing apparatus 1 according to the first embodiment, the information processing apparatus 1A according to the second embodiment acquires voices uttered by persons belonging to the conversation groups G1 to G3 as voice data.
　続いて、第２実施形態に係る情報処理装置１Ａの構成を説明する。図９は、第２実施形態に係る情報処理装置１Ａの内部構成の概略を示すブロック図である。第１実施形態に係る情報処理装置１と同様の処理については説明を省略する。 Subsequently, the configuration of the information processing apparatus 1A according to the second embodiment will be described. FIG. 9 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1A according to the second embodiment. Descriptions of processes similar to those of the information processing apparatus 1 according to the first embodiment will be omitted.
　第２実施形態に係る情報処理装置１Ａの制御ユニット１０は、ＨＤＤ１１４に記憶されているグループ再構成プログラムが上記のプロセッサーで実行されることにより、制御部１００、第１検出部１０１、第１集計部１０２、再構成部１３０、テキスト化部１２２、第２検出部１３１、及び第２集計部１３２として機能する。なお、制御部１００、第１検出部１０１、第１集計部１０２、再構成部１３０、テキスト化部１２２、第２検出部１３１、及び第２集計部１３２は、ハード回路により構成されてもよい。 The control unit 10 of the information processing apparatus 1A according to the second embodiment executes the group reconfiguration program stored in the HDD 114 by the above processor, whereby the control unit 100, the first detection unit 101, and the first aggregation are performed. It functions as the unit 102, the reconstruction unit 130, the textification unit 122, the second detection unit 131, and the second aggregation unit 132. The control unit 100, the first detection unit 101, the first aggregation unit 102, the reconstruction unit 130, the textification unit 122, the second detection unit 131, and the second aggregation unit 132 may be configured by a hardware circuit. .
　再構成部（グループ再構成部）１３０は、第１集計部１０２によって集計された各人の発言時間に基づいて、各グループ構成員を再構成する。本実施形態では、再構成部１３０は、第１集計部１０２によって集計された人Ｐ１１～Ｐ３３のそれぞれの発言時間に基づいて、グループＧ１～Ｇ３の各構成員を再構成する。 The reconfiguration unit (group reconfiguration unit) 130 reconfigures each group member based on the speech time of each person counted by the first counting unit 102. In the present embodiment, the reconfiguration unit 130 reconfigures the members of the groups G1 to G3 based on the speech times of the persons P11 to P33 counted by the first counting unit 102.
　テキスト化部１２２は、上記音声データに含まれる各人の発言の内容をテキスト化する。 The text conversion unit 122 converts the contents of each person's utterance included in the voice data into text.
　第２検出部（キーワード検出部）１３１は、テキスト化部１２２によってテキスト化されたテキストデータに基づいて、上記各人による発言が予め定められたキーワードを含むか否かを判断する。 The second detection unit (keyword detection unit) 131 determines, based on the text data converted into text by the text conversion unit 122, whether the utterance by each person includes a predetermined keyword.
　第２集計部（グループ毎集計部）１３２は、第１集計部１０２によって集計された人Ｐ１１～Ｐ３３のそれぞれの発言時間を、再構成部１３０により再構成されたグループ毎に集計する。 The second tabulating unit (tabulation unit for each group) 132 tabulates the speech time of each of the persons P11 to P33 tabulated by the first tabulating unit 102 for each group reconstructed by the reconstructing unit 130.
　次に、情報処理装置１Ａによるグループ再構成処理を説明する。図１０は情報処理装置１Ａによるグループ再構成処理を示すフローチャートである。 Next, group reconfiguration processing by the information processing apparatus 1A will be described. FIG. 10 is a flowchart showing group reconfiguration processing by the information processing apparatus 1A.
　当該評価が行われる場面は、第１実施形態に係る情報処理装置１の場合と同様で、図１を参照して説明したように、会話グループＧ１～Ｇ３に属する各人が、それぞれのグループ毎に会議を行っている場面である。人Ｐ１１～Ｐ３３は、各自がヘッドセット２を装着し、各ヘッドセット２は上述したように情報処理装置１Ａに対して通信可能に接続されている。この状態で、人Ｐ１１～Ｐ３３は、各自が属するそれぞれのグループにおいて会議中に発話を行う。人Ｐ１１～Ｐ３３によって発話された音声は、人Ｐ１１～Ｐ３３のそれぞれのヘッドセット２により集音され、情報処理装置１Ａに出力される。 The scene in which the evaluation is performed is the same as in the case of the information processing apparatus 1 according to the first embodiment, and as described with reference to FIG. In a meeting. Each of the persons P11 to P33 wears the headset 2, and each headset 2 is communicably connected to the information processing apparatus 1A as described above. In this state, the people P11 to P33 speak during the meeting in the respective groups to which they belong. The voice uttered by the people P11 to P33 is collected by the headset 2 of each of the people P11 to P33, and is output to the information processing apparatus 1A.
　情報処理装置１Ａは、通信インターフェイス１１８を介して、各ヘッドセット２から音声データを取得する（ステップＳ１０１）。第１検出部１０１は、人Ｐ１１～Ｐ３３毎に、すなわち、ＨＤＤ１１４に記憶されている識別情報に対応付けて、当該音声データをＨＤＤ１１４に記憶させる。 The information processing apparatus 1A acquires audio data from each headset 2 via the communication interface 118 (step S101). The first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
　続いて、第１検出部１０１は、人Ｐ１１～Ｐ３３毎にＨＤＤ１１４に記憶された音声データのそれぞれから、当該音声データが示す発話の中でされている各発言を、上述したようにして抽出する（ステップＳ１０２）。そして、第１検出部１０１は、当該抽出した各発言の発言時間を検出する（ステップＳ１０３）。 Subsequently, the first detection unit 101 extracts, as described above, each utterance in the speech indicated by the voice data from each of the voice data stored in the HDD 114 for each of the persons P11 to P33. (Step S102). Then, the first detection unit 101 detects the utterance time of each of the extracted utterances (step S103).
　更に、第１集計部１０２は、第１検出部１０１によって検出された上記各発言についての発言時間を上記人Ｐ１１～Ｐ３３のそれぞれについて個別に集計する（ステップＳ１０４）。 Furthermore, the first aggregation unit 102 individually aggregates, for each of the persons P11 to P33, the speech time for each of the utterances detected by the first detection unit 101 (step S104).
　続いて、再構成部１３０は、第１集計部１０２によって集計された発言時間が長い人から順に順位付けを行う（ステップＳ１０５）。そして、再構成部１３０は、順位上位から予め定められた人数毎にグループ分けを行って、グループを再構成する（ステップＳ１０６）。 Subsequently, the reconfiguration unit 130 performs ranking in order from the person with the longest utterance time counted by the first counting unit 102 (step S105). Then, the reconstruction unit 130 performs grouping into groups of predetermined numbers from the top ranks to reconstruct a group (step S106).
　例えば、再構成部１３０による、人Ｐ１１～Ｐ３３の発言時間長さの順位付けが、順位上位から、Ｐ３１、Ｐ２１、Ｐ１１、Ｐ２２、Ｐ２３、Ｐ１２、Ｐ１３、Ｐ３３、Ｐ３２であったとする。この場合、再構成部１３０は、上記予め定められた人数を例えば３名として、グループＧ１の構成員をＰ３１、Ｐ２１、Ｐ１１とし、グループＧ２の構成員をＰ２２、Ｐ２３、Ｐ１２とし、グループＧ３の構成員をＰ１３、Ｐ３３、Ｐ３２として、各グループの再構成を行う。 For example, it is assumed that the reordering unit 130 ranks the speaking time lengths of the people P11 to P33 from P31, P21, P11, P22, P23, P12, P12, P13, P33, and P32 from the top. In this case, the reconstruction unit 130 sets the predetermined number of people as, for example, three, sets members of the group G1 as P31, P21, and P11, and sets members of the group G2 as P22, P23, and P12. Each group is reconfigured with the members as P13, P33, and P32.
　制御部１００は、ＨＤＤ１１４にそれぞれ対応付けて記憶されている識別情報及びグループ情報を、再構成部１３０による再構成の結果が反映されるように更新する。この場合、ＨＤＤ１１４は、Ｐ３１、Ｐ２１、Ｐ１１がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ１を特定するグループ情報と対応付けて記憶する。ＨＤＤ１１４はまた、Ｐ２２、Ｐ２３、Ｐ１２がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ２を特定するグループ情報と対応付けて記憶する。ＨＤＤ１１４はさらに、Ｐ１３、Ｐ３３、Ｐ３２がそれぞれ使用しているヘッドセット２の識別情報を、グループＧ３を特定するグループ情報と対応付けて記憶する。 The control unit 100 updates the identification information and the group information stored in association with each other in the HDD 114 so that the result of the reconfiguration by the reconfiguration unit 130 is reflected. In this case, the HDD 114 stores identification information of the headset 2 used by each of P31, P21, and P11 in association with group information specifying the group G1. The HDD 114 also stores identification information of the headset 2 used by each of P22, P23, and P12 in association with group information specifying the group G2. The HDD 114 further stores identification information of the headset 2 used by each of P13, P33, and P32 in association with group information specifying the group G3.
　そして、第２集計部１３２は、第１集計部１０２によって集計されている各人の発言時間を、上記再構成されたグループＧ１～Ｇ３毎に集計する（ステップＳ１０７）。 Then, the second totaling unit 132 totals the speech time of each person counted by the first totaling unit 102 for each of the reconfigured groups G1 to G3 (step S107).
　制御部１００は、ＨＤＤ１１４に記憶されている識別情報及びグループ情報と、第２集計部１３２による集計結果に基づいて、再構成されたグループＧ１～Ｇ３及び各グループの構成員と、Ｓ１０７で集計された上記グループＧ１～Ｇ３毎の発言時間とを表示部１１５に表示させる（ステップＳ１０８）。 Based on the identification information and the group information stored in HDD 114 and the counting result by second counting unit 132, control unit 100 counts the rearranged groups G1 to G3 and members of each group in S107. The display time is displayed on the display unit 115 for each of the groups G1 to G3 (step S108).
　この第２実施形態によれば、過去に行われた会議における各人の発言時間長さ（貢献度）に応じて、発言時間の長い人同士、及び発言時間の短い人同士等が同じグループになるように、グループを作り直すことになるので、過去に行われた会議の結果に基づいて、より効果的な討論が行われることが期待できる新たなグループを構成することが可能になる。 According to this second embodiment, according to the speech time length (contribution) of each person in a conference held in the past, people with long talk time, and people with short talk time, etc. are in the same group As a result, it will be possible to construct a new group that can be expected to be more effectively discussed based on the results of meetings held in the past, since the group will be rebuilt.
　また、再構成されたグループ毎に、構成員による発言時間の合計時間が算出されて表示されるので、再構成された新たなグルーブで会議を行った場合に、どのような結果になるかを予想することが可能になる。 In addition, since the total time of the speaking time by members is calculated and displayed for each of the reconstructed groups, what kind of result will be obtained if the meeting is performed in the new, restructured groove It becomes possible to predict.
　なお、本実施形態では、ステップＳ１０６において、以下の処理を行うようにしてもよい。例えば、再構成部１３０は、第１集計部１０２によって集計された発言時間が長い人から順に順位付けがされた後（ステップＳ１０５）、順位上位から予め定められた人数毎にグループ分けを行ってグループを再構成するが、このとき、発言時間が予め定められた規定時間よりも長い人同士を同じグループとせずにグループを再構成する。例えば、再構成部１３０は、順位上位から上記予め定められた人数毎にグループ分けを行ってグループを再構成したときに、同一のグループに、発言時間が上記規定時間よりも長い人が複数存在する場合には、当該人を、上記順位が下位の人から構成されるグループに属する人であって、発言時間が上記規定時間よりも長くない人と入れ替える。これにより、発言の長い人同士が同じグループになることが回避されるので、会議を有意義なものとなる可能性を高めることができる。 In the present embodiment, the following processing may be performed in step S106. For example, after the reordering unit 130 is ranked in order from the person with the longest utterance time tabulated by the first tabulating unit 102 (step S105), grouping is performed for each predetermined number of people from the top ranks A group is reconfigured, but at this time, a group is reconfigured without making people whose speaking time is longer than a predetermined time set in advance into the same group. For example, when the reconstruction unit 130 performs grouping into groups according to the predetermined number of people from the top of the ranking and reconstructs a group, there are a plurality of persons whose speaking time is longer than the specified time in the same group. In the case of doing this, the person concerned is replaced with a person who belongs to a group constituted by persons of lower ranks and whose speaking time is not longer than the specified time. This prevents people with long utterances from being in the same group, which can increase the possibility of making a meeting meaningful.
　次に、第２実施形態に係る情報処理装置１Ａによるグループ再構成処理の第１変形例について説明する。図１１は情報処理装置１Ａによるグループ再構成処理の第１変形例を示すフローチャートである。なお、第２実施形態と同様の処理については説明を省略する。 Next, a first modification of the group reconfiguration process by the information processing apparatus 1A according to the second embodiment will be described. FIG. 11 is a flowchart showing a first modified example of the group reconfiguration processing by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
　第１変形例では、第１集計部１０２が上記各発言についての発言時間を上記人Ｐ１１～Ｐ３３のそれぞれについて個別に集計した後（ステップＳ２０４）、再構成部１３０は、予め定められたグループ分けルールに従って、人Ｐ１１～Ｐ３３の各人を組み合わせて作成可能な全てのグループを作成する（ステップＳ２０５）。例えば、予め定められたグループ分けルールが、全構成員９人の場合、１グループ３人ずつで、３グループを作成する、であるとする。この場合、再構成部１３０は、人Ｐ１１～Ｐ３３を３人ずつ３つのグループＧ１～Ｇ３に振り分け、このように振り分け可能な全ての組み合わせを網羅し、振り分け可能な全ての組み合わせについてグループＧ１～Ｇ３を作成する。 In the first modification, after the first tabulating unit 102 tabulates the utterance time for each of the utterances individually for each of the persons P11 to P33 (step S204), the reconfiguring unit 130 divides the utterance into predetermined groupings. According to the rules, each of the persons P11 to P33 is combined to create all the groups that can be created (step S205). For example, it is assumed that, in the case where all members are nine, a predetermined grouping rule is to create three groups with three members in each group. In this case, the reconstruction unit 130 divides the people P11 to P33 into three groups G1 to G3 by three people, covers all combinations that can be distributed in this way, and groups G1 to G3 for all combinations that can be distributed. Create
　続いて、再構成部１３０は、上記のようにして作成した、全てのグループについて、所属する各人の発言時間を集計する（ステップＳ２０６）。 Subsequently, the reconstructing unit 130 counts the utterance time of each member who belongs to all the groups created as described above (step S206).
　更に、再構成部１３０は、上記のようにして作成した、各グルーブＧ１～Ｇ３について、各グループ間におけるグループ発言時間の差を算出し、当該差が最小となっているグループＧ１～Ｇ３を抽出する（ステップＳ２０７）。 Furthermore, for each of the grooves G1 to G3 created as described above, the reconstruction unit 130 calculates the difference in group utterance time between the groups, and extracts the group G1 to G3 in which the difference is the smallest. (Step S207).
　制御部１００は、ステップＳ２０７で抽出されたグループＧ１～Ｇ３と、抽出された当該グループＧ１～Ｇ３それぞれについての各人の発言時間の集計を表示部１１５に表示させる（ステップＳ２０８）。 The control unit 100 causes the display unit 115 to display a tabulation of the utterance time of each of the groups G1 to G3 extracted in step S207 and the extracted groups G1 to G3 (step S208).
　この第１変形例によれば、過去に行われた会議における各人の発言時間長さ（貢献度）に応じて、各グループでの発言時間合計が均等となるように調整するため、あるグループだけが発言の多い人のみで構成され、別のグループは発言の少ない人のみで構成されるといった、グループ構成の偏りがなくなる。 According to the first modification, in order to adjust the total speaking time in each group to be equal according to the speaking time length (contribution) of each person in the conference held in the past, a certain group There is no bias in the group structure, such as only people with many utterances, and other groups with only few people.
　次に、第２実施形態に係る情報処理装置１Ａによるグループ再構成処理の第２変形例について説明する。図１２は情報処理装置１Ａによるグループ再構成処理の第２変形例を示すフローチャートである。なお、第２実施形態と同様の処理については説明を省略する。 Next, a second modified example of the group reconfiguration processing by the information processing apparatus 1A according to the second embodiment will be described. FIG. 12 is a flow chart showing a second modification of the group reconstruction processing by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
　第２変形例では、第２実施形態と同様にして、再構成部１３０が、第１集計部１０２によって集計された発言時間が長い人から順に順位付けを行った後に（ステップＳ３０５）、再構成部１３０は、最低順位から予め定められた順位までの人と、最高順位から予め定められた順位までの人とを同一のグループにするという条件を満たした上で、予め定められた人数毎にグループを再構成する（ステップＳ３０６）。 In the second modification, similarly to the second embodiment, the reconstruction unit 130 sequentially ranks in order from the person with the longest utterance time counted by the first aggregation unit 102 (step S305), and then reconstructs it. Section 130 satisfies the condition that the people from the lowest rank to the predetermined rank and the people from the highest rank to the predetermined rank are in the same group, and for each predetermined number of people. The group is reconfigured (step S306).
　例えば、再構成部１３０による、人Ｐ１１～Ｐ３３の発言時間長さの順位付けが、順位上位から、Ｐ３１、Ｐ２１、Ｐ１１、Ｐ２２、Ｐ２３、Ｐ１２、Ｐ１３、Ｐ３３、Ｐ３２であったとする。また、上記予め定められた人数を例えば３名とする。そして、例えば、再構成部１３０は、最高順位１名の者のみを上記「最高順位から予め定められた順位まで」とし、最低順位から２名までの者を「最低順位から予め定められた順位まで」として、各グループの再構成を行う。この場合、再構成部１３０は、まず、最高順位のＰ３１と、最低順位から２名のＰ３３、Ｐ３２をグループＧ１の構成員とする。続いて、再構成部１３０は、当該３名（Ｐ３１、Ｐ３３、Ｐ３２）を除いて、最高順位となるＰ２１と、最低順位から２名のＰ１２、Ｐ１３をグループＧ２の構成員とする。同様にして、再構成部１３０は、当該３名（Ｂ２１、Ｐ１２、Ｐ１３）を更に除いて、最高順位となるＰ１１と、最低順位から２名のＰ２２、Ｐ２３をグループＧ３の構成員とする。 For example, it is assumed that the reordering unit 130 ranks the speaking time lengths of the people P11 to P33 from P31, P21, P11, P22, P23, P12, P12, P13, P33, and P32 from the top. Further, the predetermined number of people is, for example, three. Then, for example, the reconstruction unit 130 sets only one person with the highest rank to the above “from the highest rank to a predetermined rank”, and “up to two persons from the lowest rank” to the “predetermined rank from the lowest rank Reconfigure each group as "up". In this case, the reconfiguration unit 130 first sets P31 of the highest rank and P33 and P32 of the lowest rank to two members of the group G1. Subsequently, except for the three persons (P31, P33, P32), the reconfiguration unit 130 sets P21 having the highest rank and two P12 and P13 from the lowest rank as members of the group G2. Similarly, the reconstruction unit 130 excludes the three persons (B21, P12, and P13), and sets P11 having the highest rank and two P22 and P23 from the lowest rank as members of the group G3.
　そして、第２集計部１３２は、第１集計部１０２によって集計された各人の発言時間を、上記再構成された上記グループＧ１～Ｇ３のそれぞれ毎に集計する（ステップＳ３０７）。 Then, the second totaling unit 132 totals the speech time of each person counted by the first totaling unit 102 for each of the reconstructed groups G1 to G3 (step S307).
　制御部１００は、再構成された上記グループＧ１～Ｇ３と、集計された上記グループＧ１～Ｇ３別の上記発言時間の合計とを表示部１１５に表示させる（ステップＳ３０８）。 The control unit 100 causes the display unit 115 to display the reconstructed groups G1 to G3 and the sum total of the utterance times of the groups G1 to G3 (step S308).
　この第２変形例によれば、過去に行われた会議における各人の発言時間長さ（貢献度）に応じて、各グループに、比較的積極的に発言を行う人と比較的書極的に発言を行う人とを組み合わせて、グループを構成できるため、あるグループだけが発言の多い人のみで構成され、別のグループは発言の少ない人のみで構成されるといった、グループ構成の偏りがなくなる。 According to this second modified example, each group is relatively actively written with a person who speaks relatively positively according to the speech time length (contribution) of each person in a conference held in the past. Groups can be configured by combining with people who make a comment, so that there is no bias in group configuration, such that only one group is composed of only people who speak a lot, and another group is composed only of people who have a few voices. .
　次に、第２実施形態に係る情報処理装置１Ａによるグループ再構成処理の第３変形例について説明する。図１３は情報処理装置１Ａによるグループ再構成処理の第３変形例を示すフローチャートである。なお、第２実施形態と同様の処理については説明を省略する。 Next, a third modification of the group reconfiguration process by the information processing apparatus 1A according to the second embodiment will be described. FIG. 13 is a flowchart showing a third modification of the group reconfiguration process by the information processing apparatus 1A. The description of the same processing as that of the second embodiment will be omitted.
　第３変形例は、第２実施形態、並びに第２実施形態の第１及び第２の変形例によりグループ再構成が行われた後に実行される処理である。 The third modification is processing executed after group reconfiguration is performed in the second embodiment and the first and second modifications of the second embodiment.
　グループ再構成処理が終了すると、テキスト化部１２２は、上記音声データに含まれる上記各人の発言の内容をキャラクター変換してテキスト化する（ステップＳ４０１）。 When the group reconstruction processing is completed, the text conversion unit 122 converts the contents of the utterances of the above persons included in the audio data into characters and converts them into text (step S401).
　続いて、第２検出部１３１は、当該テキストデータに基づいて、予め定められたキーワードを含む発言と、当該発言をした人を特定する（ステップＳ４０２）。 Subsequently, the second detection unit 131 specifies an utterance including a predetermined keyword and a person who made the utterance based on the text data (step S402).
　そして、再構成部１３０は、上記再構成を終えたグループＧ１～Ｇ３において、上記キーワードを含む発言を行った複数の人が同一のグループの構成員となっているか否かを判断する（ステップＳ４０３）。ここで、再構成部１３０は、上記再構成を終えたグループＧ１～Ｇ３に、上記キーワードを含む発言を行った複数の人が同一のグループの構成員となっていないと判断した場合は（ステップＳ４０３でＮＯ）、この時点で構成されているグループＧ１～Ｇ３を維持して、処理を終了する。 Then, in the groups G1 to G3 for which the reconfiguration has been completed, the reconfiguration unit 130 determines whether or not a plurality of persons who have made a statement including the keyword are members of the same group (step S403). ). Here, if the reconfiguration unit 130 determines that a plurality of persons who have made a remark including the above-mentioned keyword are not members of the same group in the groups G1 to G3 after the above-mentioned reconfiguration (Step The process ends while maintaining the groups G1 to G3 configured at this point in time).
　一方、再構成部１３０は、上記再構成を終えたグループＧ１～Ｇ３において、上記キーワードを含む発言を行った複数の人が同じグループにおいて共に構成員となっていると判断した場合は（ステップＳ４０３でＹＥＳ）、グループＧ１～Ｇ３の中に、当該キーワードを含む発言をしていない人のみで構成されるグループがあるか否かを判断する（ステップＳ４０４）。 On the other hand, when it is determined that, in the groups G1 to G3 for which the reconfiguration has been completed, the reconfiguration unit 130 has a plurality of persons who have made a statement including the keyword become members of the same group (step S403 YES), it is determined whether or not there is a group consisting of only those who do not speak including the keyword among the groups G1 to G3 (step S404).
　ここで、再構成部１３０が、グループＧ１～Ｇ３の中に、当該当該キーワードを含む発言をしていない人のみで構成されるグループがあると判断した場合（ステップＳ４０４でＹＥＳ）、再構成部１３０は、上記同じグループにおいて共に構成員（上記キーワードを含む発言を行った人）とされている人のうちの一人を、当該キーワードを含む発言をしていない人のみで構成されるグループの一人と入れ替えて、グループを再構成する（ステップＳ４０５）。なお、再構成部１３０は、グループＧ１～Ｇ３の中に、当該キーワードを含む発言をしていない人のみで構成されるグループがないと判断した場合は（ステップＳ４０４でＮＯ）、グループの組み替えを行うことなく、この時点で構成されているグループＧ１～Ｇ３を維持して、処理を終了する。 Here, when reconfiguration unit 130 determines that there is a group consisting of only those who do not speak including the keyword, among the groups G1 to G3 (YES in step S404), the reconfiguration unit 130 is one of a group consisting of only one person among those who are both members (people who have made a statement including the above keyword) in the same group but who are not making a statement including that keyword And the group is reconstructed (step S405). When it is determined that there is no group consisting of only those who do not speak including the keyword among the groups G1 to G3 (NO in step S404), the reconfiguration unit 130 rearranges the group. The process is terminated without maintaining the groups G1 to G3 configured at this time without performing the process.
　この第３変形例によれば、同じような発言内容を行う人同士が同じグループに属する事態を避け、各自が偏らない内容の発言を行うグループ分けを実現することができる。 According to the third modified example, it is possible to avoid grouping situations in which persons who perform similar utterance content belong to the same group, and to realize grouping into which utterances of non-biased contents are made.
　次に、第３実施形態に係る情報処理装置１Ｂの構成を説明する。図１４は、第３実施形態に係る情報処理装置１Ｂの内部構成の概略を示すブロック図である。第１実施形態に係る情報処理装置１又は第２実施形態に係る情報処理装置１Ａと同様の処理については説明を省略する。 Next, the configuration of the information processing apparatus 1B according to the third embodiment will be described. FIG. 14 is a block diagram showing an outline of an internal configuration of the information processing apparatus 1B according to the third embodiment. Descriptions of processes similar to those of the information processing apparatus 1 according to the first embodiment or the information processing apparatus 1A according to the second embodiment will be omitted.
　第３実施形態に係る情報処理装置１Ｂの制御ユニット１０は、ＨＤＤ１１４に記憶されている評価プログラムが上記のプロセッサーで実行されることにより、制御部１００、第３検出部１３５、作成部１３６、評価部１３７、テキスト化部１２２、及び第２検出部１３１として機能する。なお、制御部１００、第３検出部１３５、作成部１３６、評価部１３７、テキスト化部１２２、及び第２検出部１３１は、ハード回路により構成されてもよい。 In the control unit 10 of the information processing apparatus 1B according to the third embodiment, the evaluation program stored in the HDD 114 is executed by the processor, whereby the control unit 100, the third detection unit 135, the creation unit 136, the evaluation It functions as the unit 137, the text conversion unit 122, and the second detection unit 131. The control unit 100, the third detection unit 135, the creation unit 136, the evaluation unit 137, the text conversion unit 122, and the second detection unit 131 may be configured by a hardware circuit.
　第３検出部（発言検出部）１３５は、ＨＤＤ１１４に記憶されている人Ｐ１１～Ｐ３３の９人についての音声データのそれぞれから、当該発話の中でされている各発言を、各発言を行った人と、各発言が行われた時刻と共に検出する。 The third detection unit (speech detection unit) 135 made each of the speeches in the speech from each of the speech data of nine persons P11 to P33 stored in the HDD 114. Detect with the person and the time each utterance was made.
　作成部（発言分布作成部）１３６は、第３検出部１３５によって検出された各発言及びその時刻を用いて、時間経過に応じた発言量の変化を示す発言分布図を作成する。 The creation unit (speech distribution creation unit) 136 creates a statement distribution map showing a change in the amount of statement according to the passage of time, using each statement detected by the third detection unit 135 and its time.
　評価部１３７は、作成部１３６によって作成された発言分布図における、発言量が増加傾向を示す時間帯において最初に発せられた発言をした人の評価レベルを、予め定められた第１の高レベルと判定する処理を行う。 The evaluation unit 137 sets the evaluation level of the person who made the first utterance in the time zone in which the amount of utterance increases in the utterance distribution map created by the creator 136 to a predetermined first high level. Perform processing to determine that
　テキスト化部１２２は、上記音声データに含まれる各人の発言の内容をテキスト化する。 The text conversion unit 122 converts the contents of each person's utterance included in the voice data into text.
　第２検出部１３１は、テキスト化部１２２によってテキスト化されたテキストデータに基づいて、上記各人による発言が予め定められたキーワードを含むか否かを判断する。 The second detection unit 131 determines, based on the text data converted into text by the text conversion unit 122, whether the utterance by each person includes a predetermined keyword.
　次に、第３実施形態に係る情報処理装置１Ｂによる評価処理について説明する。図１５は情報処理装置１Ｂによる評価処理を示すフローチャートである。 Next, evaluation processing by the information processing apparatus 1B according to the third embodiment will be described. FIG. 15 is a flowchart showing evaluation processing by the information processing apparatus 1B.
　当該評価処理が行われる場面は、第１実施形態に係る情報処理装置１の場合と同様で、図１を参照して説明したように、会話グループＧ１～Ｇ３に属する各人が、それぞれのグループ毎に会議を行っている場面である。情報処理装置１Ｂは、通信インターフェイス１１８を介して、各ヘッドセット２から音声データを取得する（ステップＳ５０１）。第１検出部１０１は、人Ｐ１１～Ｐ３３のそれぞれ毎に、すなわち、ＨＤＤ１１４に記憶されている識別情報に対応付けて、当該音声データをＨＤＤ１１４に記憶させる。 The scene where the evaluation process is performed is the same as in the case of the information processing apparatus 1 according to the first embodiment, and each person belonging to the conversation groups G1 to G3 is a group as described with reference to FIG. It is a scene where we have a meeting every time. The information processing apparatus 1B acquires audio data from each headset 2 via the communication interface 118 (step S501). The first detection unit 101 stores the voice data in the HDD 114 for each of the persons P11 to P33, that is, in association with the identification information stored in the HDD 114.
　続いて、第３検出部１３５は、人Ｐ１１～Ｐ３３毎にＨＤＤ１１４に記憶されている音声データのそれぞれから、当該音声データが示す発話の中でされている各発言を、各発言の時刻と共に、上述したようにして抽出する（ステップＳ５０２）。ここで、「時刻」とは、音声データ開始時からの経過時間のことである。 Subsequently, the third detection unit 135 determines, from the voice data stored in the HDD 114 for each of the persons P11 to P33, each utterance being performed in the utterance indicated by the voice data, together with the time of each utterance. Extraction is performed as described above (step S502). Here, "time" is an elapsed time from the start of audio data.
　更に、作成部１３６は、第３検出部１３５によって検出された各発言及び各発言の時刻を用いて、時間経過に応じた発言量の変化を示す発言分布図を作成する（ステップＳ５０３）。例えば、図１６に示すように、会議開始からの時間経過に応じて、グループＧ１～Ｇ３毎に、例えば１分経過毎に発言の数を示した折れ線グラフを作成する。 Furthermore, the creating unit 136 creates a statement distribution map showing a change in the amount of messages according to the passage of time, using the utterance and the time of each utterance detected by the third detection unit 135 (step S503). For example, as shown in FIG. 16, in accordance with the passage of time from the start of the conference, a line graph indicating the number of utterances per one minute, for example, is created for each of the groups G1 to G3.
　続いて、評価部１３７は、作成部１３６によって作成された発言分布図における、発言量が増加傾向を示す時間帯を抽出する（ステップＳ５０４）。例えば、評価部１３７は、上記作成した発言分布図の折れ線が示す傾斜が、予め定められた角度以上になっているか否かを、例えば、予め定められた時間（例えば、１分間）での増加率（微分値）が予め定められた値（発言数増加率５０％）以上となっている時間帯を抽出する。例えば、図１６に示す発言分布図の場合、評価部１３７は、ポイントＰ１から１分後のポイントＰ２までの傾き（増加率、微分値）は、予め定められた値の５０％以上となっているため、当該時間帯を抽出する。 Subsequently, the evaluation unit 137 extracts a time zone in which the amount of speech has a tendency to increase in the speech distribution map generated by the generation unit 136 (step S504). For example, the evaluation unit 137 increases, for example, in a predetermined time (for example, one minute) whether the inclination indicated by the broken line in the above-mentioned statement distribution map is equal to or more than a predetermined angle. A time zone in which the rate (differential value) is equal to or more than a predetermined value (50% increase rate in the number of messages) is extracted. For example, in the case of the utterance distribution map shown in FIG. 16, the evaluation unit 137 determines that the slope (increase rate, differential value) from point P1 to point P2 one minute after is 50% or more of the predetermined value. Therefore, the relevant time zone is extracted.
　そして、評価部１３７は、抽出した時間帯において、最初に発せられた発言及び当該発言をした人を判定する（ステップＳ５０５）。例えば、評価部１３７は、ポイントＰ１（開始から１１分経過時）の時点経過後、最初に行われた発言及び発言者を特定する。 Then, in the extracted time zone, the evaluation unit 137 determines the first utterance and the person who made the utterance (step S505). For example, after the time point of point P1 (when 11 minutes has elapsed from the start), the evaluation unit 137 specifies the first speech and speaker.
　更に、評価部１３７は、当該最初に行われた発言の発言者の評価レベルを、予め定められた第１の高レベルと判定する（ステップＳ５０６）。この第１の高レベルとは、会議において、発言を行った後に他の発言が増加したことにより会議に貢献している人、或いは、それに近似する人として、当該人に与えられる高い評価レベルである。 Furthermore, the evaluation unit 137 determines that the evaluation level of the speaker of the first-to-be-sent utterance is the first high level determined in advance (step S506). The first high level is a high evaluation level given to the person as a person who contributes to or is similar to the person at the meeting after making a statement and increasing other remarks. is there.
　この第３実施形態によれば、自身の発言により他の発言を増加させた人を上記第１の高レベルと判定して評価することにより、あるメンバーによる発言が他の発言に与える影響に応じた評価を、当該メンバーに与えることが可能になる。 According to the third embodiment, the person who has caused another utterance to increase according to his / her own utterance is judged to be the first high level and evaluated, whereby the influence of the utterance of a certain member on the other utterance is made. Can be given to the members concerned.
　次に、第３実施形態に係る情報処理装置１Ｂによる評価処理の第１変形例について説明する。図１７は情報処理装置１Ｂによる評価処理の第１変形例を示すフローチャートである。なお、第３実施形態と同様の処理は説明を省略する。 Next, a first modification of the evaluation process by the information processing apparatus 1B according to the third embodiment will be described. FIG. 17 is a flowchart showing a first modification of the evaluation process by the information processing apparatus 1B. The same processes as in the third embodiment will not be described.
　第１変形例では、評価部１３７が、上記人を第１の高レベルに判定した後、（ステップＳ６０６）、テキスト化部１２２は、上記音声データに含まれる上記各人の発言の内容をキャラクター変換してテキスト化する（ステップＳ６０７）。 In the first modification, after the evaluation unit 137 determines that the person is at the first high level (step S606), the textification unit 122 converts the contents of the utterance of each person included in the voice data into a character Convert and convert into text (step S 607).
　続いて、評価部１３７は、上記抽出した発言量が増加傾向を示す上記時間帯において上記最初に発せられた発言に含まれる各単語を、上記テキスト化部１２２によりテキスト化されたテキストデータから抽出する（ステップＳ６０８）。 Subsequently, the evaluation unit 137 extracts, from the text data converted into text by the text conversion unit 122, each word included in the first utterance made in the time zone in which the extracted utterance amount shows an increasing tendency. (Step S608).
　制御部１００は、評価部１３７により抽出された上記各単語を表示部１１５に表示させる（ステップＳ６０９）。 The control unit 100 causes the display unit 115 to display the words extracted by the evaluation unit 137 (step S609).
　この第１変形例によれば、他の発言を増加させた発言に含まれる単語を抽出するため、どのような単語がその後に会議を活発化させたかが明瞭になる。 According to this first modification, it becomes clear which word has subsequently activated the conference, in order to extract words included in the utterances in which other utterances are increased.
　次に、第３実施形態に係る情報処理装置１Ｂによる評価処理の第２変形例について説明する。図１８は情報処理装置１Ｂによる評価処理の第２変形例を示すフローチャートである。なお、第３実施形態又は第３実施形態の第１変形例と同様の処理は説明を省略する。 Next, a second modification of the evaluation process by the information processing apparatus 1B according to the third embodiment will be described. FIG. 18 is a flowchart showing a second modified example of the evaluation process by the information processing apparatus 1B. The same processes as those of the third embodiment or the first modification of the third embodiment will not be described.
　第２変形例では、制御部１００が、評価部１３７により抽出された上記単語を表示部１１５に表示させた後（ステップＳ７０９）、評価部１３７は、上記最初に発せられた発言に含まれる各単語が、上記抽出された時間帯において当該最初の発言の後に行われた他の発言にも含まれて出現数が増加しているかを判断する（ステップＳ７１０）。例えば、評価部１３７は、上記最初に発せられた発言に含まれる各単語が、上記抽出された時間帯において当該最初の発言の後に行われた他の発言（全ての発言）が上記テキスト化部１２２によりテキスト化されたテキストデータに出現するかを判定する。評価部１３７は、当該各単語が出現する場合には、上記各単語のそれぞれについて出現数を検出し、各単語のそれぞれについて、上記最初に発せられた発言における出現数よりも増加しているかを判断する。 In the second modification, after the control unit 100 causes the display unit 115 to display the word extracted by the evaluation unit 137 (step S709), the evaluation unit 137 determines each of the words included in the first utterance. It is determined whether the word is included in the other utterances performed after the first utterance during the extracted time zone to increase the appearance number (step S710). For example, the evaluation unit 137 may be configured such that each word included in the first utterance is the other utterance (all utterances) performed after the first utterance during the extracted time zone. It is determined whether the data appears in the text data converted into text data. When the respective words appear, the evaluation unit 137 detects the number of appearances of each of the words, and determines whether the number of appearances of each of the words is greater than the number of appearances in the first utterance. to decide.
　ここで、評価部１３７が、上記最初に発せられた発言に含まれる単語が、上記抽出された時間帯において当該発言の後に行われた他の発言にも含まれて出現数が増加していると判断した場合（ステップＳ７１０でＹＥＳ）、上記最初に発せられた発言を行った人の評価レベルを、第１の高レベルよりも更に高い第２の評価レベルと判定する（ステップＳ７１１）。例えば、評価部１３７は、上記各単語の内、１つでも出現数が増加していれば、ステップＳ７１０においては「増加している」と判断する。この第２の高レベルとは、会議において、ある単語を発した後に他の発言が増加し、かつ、当該単語の出現量が増加したことにより会議に貢献している人、或いは、それに近似する人として、当該人に与えられる高い評価レベルである。 Here, the evaluation unit 137 indicates that the word included in the first utterance is included in the other utterances performed after the utterance during the extracted time zone, and the appearance number is increased. If it is determined (YES in step S710), the evaluation level of the person who made the first utterance is determined to be a second evaluation level higher than the first high level (step S711). For example, if the number of occurrences of even one of the above-described words increases, the evaluation unit 137 determines that the number of occurrences is “increased” in step S710. This second high level means that people who contribute to the meeting or have an approximation of the increase in the amount of occurrence of the word after the utterance of a certain word increases in the meeting. As a person, it is a high evaluation level given to the person.
　この第２変形例によれば、自身の発した言葉（単語）により、その言葉が後に議論において他者にも多用されることになり、会議を活発化させる要因となる言葉を発した者を的確に見つけ出して評価することが可能になる。 According to the second modification, the person's words (words) are later frequently used by others in the discussion, and those who give the words that cause the activation of the conference It will be possible to accurately find out and evaluate.
　なお、制御部１００は、指示入力部１１９を介してユーザーによって入力される指示に応じて、評価部１３７による評価結果を表示部１１５に表示可能に構成されていてもよい。評価結果としては、評価結果を示す情報であれば特に限定されないが、例えば、各人について、第１の評価レベル及び第２の評価レベルのうちの少なくともいずれかが付与されているか否かを示す情報がある。 Control unit 100 may be configured to be able to display the evaluation result by evaluation unit 137 on display unit 115 in accordance with an instruction input by the user via instruction input unit 119. The evaluation result is not particularly limited as long as it is information indicating an evaluation result, but indicates, for example, whether or not at least one of the first evaluation level and the second evaluation level is given to each person. There is information.
　なお、上記第１乃至第３の実施の形態において、発言を行なった人の特定は、ヘッドセット２に付されている識別情報に基づいて行なわれたが、本発明はそのような実施の形態に限定されない。例えば、一般的な話者識別技術を用いて、発言を行なった人の特定が行なわれてもよい。 In the first to third embodiments, the identification of the person who made the utterance was performed based on the identification information attached to the headset 2. However, the present invention relates to such an embodiment. It is not limited to. For example, a general speaker identification technique may be used to identify the person who made the utterance.
　また、上記実施形態では、図１乃至図１８を用いて上記実施形態により示した構成及び処理は、本発明の一実施形態に過ぎず、本発明を当該構成及び処理に限定する趣旨ではない。 Further, in the above embodiment, the configurations and processes shown by the above embodiment using FIGS. 1 to 18 are only one embodiment of the present invention, and the present invention is not limited to the configurations and processes.

Claims (11)

  1. 　予め定められた複数人数からなる複数のグループにおける各人の発話が記録された音声データから、当該発話の中でされている各発言についての発言時間を検出する第１検出部と、
    　前記第１検出部によって検出された前記各発言についての発言時間を前記各人についてそれぞれ集計する第１集計部と、
    　前記第１集計部によって集計された前記各人の前記発言時間に基づいて、前記グループを再構成する再構成部と、を備える情報処理装置。     A first detection unit that detects a speech time for each utterance in the utterance from voice data in which the utterance of each person in a plurality of predetermined groups of people is recorded;
    A first counting unit that counts, for each person, the speech time for each of the utterances detected by the first detection unit;
    An reconstructing unit configured to reconstruct the group based on the speaking time of each person aggregated by the first aggregating unit.
  2. 　前記再構成部は、前記第１集計部によって集計された前記発言時間が長い人から順に、予め定められた人数毎にグループ分けを行って、前記グループを再構成する請求項１に記載の情報処理装置。 The information according to claim 1, wherein the reconstruction unit rearranges the group by performing grouping into groups according to a predetermined number of people in order from the person with the longest utterance time counted by the first tabulation unit. Processing unit.
  3. 　前記再構成部は、前記第１集計部によって集計された前記発言時間が長い人同士を同一のグループとせずに、前記グループを再構成する請求項１に記載の情報処理装置。 The information processing apparatus according to claim 1, wherein the reconstruction unit reconstructs the group without setting persons having a long speech time counted by the first aggregation unit as the same group.
  4. 　前記再構成部は、前記第１集計部によって集計された各人の前記発言時間のグループ合計を予め定められた差までに収めて、予め定められた人数毎に前記グループを再構成する請求項１に記載の情報処理装置。 The reconstruction unit is configured to reorganize the group for each predetermined number of persons by putting the group total of the utterance time of each person aggregated by the first aggregation unit to a predetermined difference. The information processing apparatus according to 1.
  5. 　前記再構成部は、前記第１集計部によって集計された前記発言時間に応じて各人を順位付けし、最低順位から予め定められた順位までの人と、最高順位から予め定められた順位までの人とを同一のグループにするという条件を満たした上で、予め定められた人数毎に前記グループを再構成する請求項１に記載の情報処理装置。 The reconstruction unit ranks each person according to the utterance time counted by the first counting unit, and the person from the lowest rank to a predetermined rank and the highest rank to a predetermined rank The information processing apparatus according to claim 1, wherein the group is reconfigured for each of a predetermined number of people, while satisfying a condition of making the same group of people.
  6. 　前記音声データに含まれる前記各人の発言の内容をテキスト化するテキスト化部と、
    　前記テキスト化部によってテキスト化されたテキストデータに基づいて、前記各人による発言が予め定められたキーワードを含むか否かを判断する第２検出部と、
    　前記再構成部は、前記第２検出部による検出結果に基づいて、同一のキーワードを含む発言をした各人を同一のグループにしないという条件を満たした上で、予め定められた人数毎に前記グループを再構成する請求項２に記載の情報処理装置。     A textification unit for converting the contents of each person's utterance included in the voice data into text;
    A second detection unit that determines whether the utterance by each person includes a predetermined keyword based on text data converted into text by the text conversion unit;
    The reconstruction unit satisfies the condition that each person who has made a speech including the same keyword is not to be in the same group based on the detection result by the second detection unit, and the above-mentioned reconstruction unit is configured for each predetermined number of people. The information processing apparatus according to claim 2, which reconfigures a group.
  7. 　前記第１検出部は、前記再構成部により再構成されたグループに属する各人の発話が記録された音声データから、当該発話の中でされている各発言についての発言時間を検出し、
    　前記第１集計部は、前記第１検出部によって検出された前記各発言についての発言時間を前記各人についてそれぞれ集計し、
    　前記第１集計部によって集計された前記各人の前記発言時間を、前記再構成されたグループ毎に集計する第２集計部を更に備える請求項１に記載の情報処理装置。     The first detection unit detects, from voice data in which an utterance of each person belonging to the group reconstructed by the reconstruction unit is recorded, a speech time for each utterance in the utterance,
    The first totaling unit totals, for each person, the speech time for each of the utterances detected by the first detection unit,
    The information processing apparatus according to claim 1, further comprising a second aggregation unit that aggregates the utterance time of each person aggregated by the first aggregation unit for each of the reconfigured groups.
  8. 　予め定められた複数人数からなる複数のグループにおける各人の発話が記録された音声データから、当該発話の中でされている各発言についての発言時間を検出する発言時間検出ステップと、
    　前記発言時間検出ステップで検出された前記各発言についての発言時間を前記各人についてそれぞれ集計する各人別発言時間集計ステップと、
    　前記各人別発言時間集計ステップで集計された前記各人の前記発言時間に基づいて、前記グループを再構成するグループ再構成ステップと、を備えるグループ再構成方法。     A speech time detection step of detecting a speech time for each speech in the speech from speech data in which the speech of each person in a plurality of groups of a predetermined number of people is recorded;
    Individual-speaking-time totaling step for counting, for each person, the speaking time for each of the utterances detected in the speaking-time detecting step;
    And D. a group reconfiguring step of reconfiguring the group based on the speaking time of each of the persons tabulated in the per-person speaking time tabulating step.
  9. 　音声を示す電気信号が入力される音声入力部と、
    　前記音声入力部に前記電気信号が入力される度に、入力された前記電気信号に基づく音声データを、前記音声を発した人ごとに記憶する記憶部と、
    　プロセッサーを含み、当該プロセッサーがグループ構成プログラムを実行することにより、
    　　前記音声データから発言に対応する部分を抽出し、前記発言が続いている時間を発言時間として検出する第１検出部と、
    　　少なくとも１つの前記発言時間を、前記人ごとに集計して、各人の発言時間を算出する第１集計部と、
    　　前記各人の発言時間に基づいて、前記各人が所属するグループを構成する構成部と、して機能する制御ユニットと、を備える情報処理装置。     A voice input unit to which an electrical signal indicating voice is input;
    A storage unit that stores voice data based on the input electrical signal for each person who has issued the voice each time the electrical signal is input to the voice input unit;
    By including a processor, the processor executes a group configuration program,
    A first detection unit which extracts a portion corresponding to a speech from the voice data and detects a time during which the speech continues as a speech time;
    A first counting unit that counts at least one of the utterance times for each person and calculates the utterance time of each person;
    An information processing apparatus, comprising: a component that constitutes a group to which each person belongs, and a control unit that functions as a group to which each person belongs based on the speaking time of each person.
  10. 　表示部をさらに備え、
    　前記制御ユニットはさらに、
    　　前記各人の前記発言時間を、前記構成部によって構成された前記グループ毎に集計して、グループ毎の発言時間を算出する第２集計部と、
    　　前記構成部によって構成された前記グループを示す情報、前記グループに所属する前記人を示す情報、及び、前記グループ毎の発言時間を示す情報を、前記表示部に表示させる制御部と、して機能する、請求項９に記載の情報処理装置。     Further comprising a display unit,
    The control unit further
    A second aggregation unit that calculates the utterance time of each group by totaling the utterance times of the respective persons for each of the groups configured by the configuration unit;
    A control unit that causes the display unit to display information indicating the group configured by the configuration unit, information indicating the person belonging to the group, and information indicating a speaking time for each group The information processing apparatus according to claim 9.
  11. 　前記記憶部はさらに、前記人を特定するための識別情報と、前記グループを特定するためのグループ情報とを対応付けて記憶し、
    　前記制御ユニットはさらに、前記構成部によって前記グループが構成されると、前記構成の結果が反映されるように、前記記憶部に記憶されている前記識別情報及び前記グループ情報を更新する制御部として機能する、請求項９に記載の情報処理装置。     The storage unit further stores identification information for identifying the person and group information for identifying the group in association with each other.
    The control unit further updates the identification information and the group information stored in the storage unit so as to reflect the result of the configuration when the group is configured by the configuration unit. The information processing apparatus according to claim 9, which functions.
PCT/JP2018/040838 2017-12-25 2018-11-02 Information processing device and group reconstruction method WO2019130818A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2017248458 2017-12-25
JP2017-248458 2017-12-25

Publications (1)

Publication Number Publication Date
WO2019130818A1 true WO2019130818A1 (en) 2019-07-04

Family

ID=67063412

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2018/040838 WO2019130818A1 (en) 2017-12-25 2018-11-02 Information processing device and group reconstruction method

Country Status (1)

Country Link
WO (1) WO2019130818A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016117070A1 (en) * 2015-01-22 2016-07-28 楽天株式会社 Information processing device, information processing method, program, and recording medium
JP2016162339A (en) * 2015-03-04 2016-09-05 Kddi株式会社 Program, terminal, and system for estimating activation of debate for each group

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016117070A1 (en) * 2015-01-22 2016-07-28 楽天株式会社 Information processing device, information processing method, program, and recording medium
JP2016162339A (en) * 2015-03-04 2016-09-05 Kddi株式会社 Program, terminal, and system for estimating activation of debate for each group

Similar Documents

Publication Publication Date Title
CN108900725B (en) Voiceprint recognition method and device, terminal equipment and storage medium
EP3254478B1 (en) Scheduling playback of audio in a virtual acoustic space
JP6755304B2 (en) Information processing device
EP3254453B1 (en) Conference segmentation based on conversational dynamics
CN113140216B (en) Selective meeting abstract
US8078470B2 (en) System for indicating emotional attitudes through intonation analysis and methods thereof
Zezario et al. STOI-Net: A deep learning based non-intrusive speech intelligibility assessment model
JP6401079B2 (en) Apparatus and program for estimating activation of discussion for each group
JP2006267465A (en) Uttering condition evaluating device, uttering condition evaluating program, and program storage medium
Völker et al. Modifications of the MUlti stimulus test with Hidden Reference and Anchor (MUSHRA) for use in audiology
CN114566187B (en) Method of operating a system comprising an electronic device, electronic device and system thereof
WO2019130816A1 (en) Information processing device and evaluation method
JP2022028539A (en) Information processing program, information processing method, and information processing apparatus
WO2019130818A1 (en) Information processing device and group reconstruction method
Ikhwanuddin et al. Library soundscape: higher education students' perception
WO2019130815A1 (en) Information processing device and evaluation method
Richter et al. EARS: An Anechoic Fullband Speech Dataset Benchmarked for Speech Enhancement and Dereverberation
US20210335352A1 (en) Information processing apparatus
Aletta et al. Exploring associations between soundscape assessment, perceived safety and well-being: A pilot field study in Granary Square, London
Walden Toward a model clinical-trials protocol for substantiating hearing aid user-benefit claims
Laskowski Predicting, detecting and explaining the occurrence of vocal activity in multi-party conversation
CN118475940A (en) Meeting containment and mixed workplace insight
JP2018049140A (en) Voice analyzing program, voice analyzer, and voice analysis method
JP7449577B2 (en) Information processing device, information processing method, and program
Ruiz Women's Perceptions on Feminism

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18894192

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18894192

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP