JP7111488B2 - Speech volume integration device, customer service support device and program - Google Patents

Description

The present invention relates to a speech amount integrating device, a customer service support device, and a program.

Conventionally, there has been disclosed a technique related to a device that analyzes conversations during customer service as one type of marketing data, and calculates the conversation ratio between a salesperson and a customer, and the relationship between the conversation ratio and customer satisfaction.

Japanese Unexamined Patent Application Publication No. 2011-238028

Here, when using conversations during customer service as marketing data, if it is possible to analyze what topics are discussed between sales representatives and customers, such technology will It can contribute to the analysis of attributes and behaviors (eg, which of the customers has the right to make purchase decisions, the types of topics that lead to purchase motivation), etc.
However, according to the conventional technology as described above, since the relationship between the conversation ratio and customer satisfaction is simply calculated, it is possible to determine what topics the salesperson and the customer are talking about. I didn't know until For this reason, according to the conventional technology as described above, there is a problem that it cannot contribute to the analysis of customer attributes and behavior.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a speech volume integration device, a customer service support device, and a program that can contribute to the analysis of customer attributes and customer situations. .

An embodiment of the present invention comprises: a conversation data acquisition unit that acquires conversation data; a keyword extraction unit that extracts keywords from the conversation data acquired by the conversation data acquisition unit; and the conversation data acquiring unit acquires the category associated with the keyword extracted by the keyword extracting unit from a storage unit in which is stored as category information associated with an utterance place determination unit that determines an utterance place of the conversation indicated by the conversation data; and an appearance result of the keyword extracted by the keyword extraction unit, for each utterance place determination unit that determines the occurrence result of the keyword extracted by the keyword extraction unit, and the category. The speech amount accumulating device includes: an accumulating unit that accumulates for each of the categories acquired by the acquiring unit; and a presenting unit that presents, for each of the categories, the integration results accumulated by the accumulating unit.

In one embodiment of the present invention, in the above-described speech amount integrating device, the appearance result of the keyword includes the duration of the topic of the category, and the integrating unit includes the category associated with the keyword. is accumulated for each category.

In one embodiment of the present invention, in the above-described speech amount integration device, the appearance result of the keyword includes the frequency of appearance of the keyword, and the integration unit integrates the frequency of appearance of the keyword for each category. do.

In one embodiment of the present invention, in the above-described speech amount accumulating device, the presenting unit includes a first axis indicating the duration of the topic of the category accumulated by the accumulating unit, and the keywords accumulated by the accumulating unit. The cumulative results are presented along with a second axis that indicates the number of occurrences.

In one embodiment of the present invention, in the above-described speech amount integration device, the integration unit outputs a totalization result obtained by adding together the integration results of the same category acquired by the category acquisition unit, and the presentation unit presents the summation result output by the accumulator.

In one embodiment of the present invention, the speech amount integrating device described above further includes a speaker determination unit that determines a speaker of the conversation indicated by the conversation data acquired by the conversation data acquisition unit, wherein the integration unit includes: Appearance results of the keyword are integrated for each speaker determined by the speaker determination unit.

In one embodiment of the present invention, in the speech amount integrating device described above, the presentation unit presents the integration result for a specific speaker among the integration results.

In one embodiment of the present invention, in the speech amount integrating device described above, the presentation unit presents the integration result for a specific utterance place among the integration results.

One embodiment of the present invention is a customer service support device that includes the speech amount integration device described above, and the conversation data is data indicating conversation during customer service.

According to one embodiment of the present invention, a computer provided in a speech amount integrating device includes a conversation data acquisition step for acquiring conversation data, and a keyword extraction for extracting keywords from the conversation data acquired in the conversation data acquisition step. Category acquisition for acquiring the category associated with the keyword extracted in the keyword extraction step from a storage unit in which the step, the keyword, and the topic category are associated and stored as category information. an utterance place determination step of determining the utterance place of the conversation indicated by the conversation data acquired by the conversation data acquisition unit; Execution of an accumulating step of accumulating for each determined utterance place and for each category obtained in the category obtaining step, and a presenting step of presenting the accumulated result accumulated in the accumulating step for each category. It is a program for
An embodiment of the present invention comprises: a conversation data acquisition unit that acquires conversation data; a keyword extraction unit that extracts keywords from the conversation data acquired by the conversation data acquisition unit; extracted by a category acquisition unit that acquires the category associated with the keyword extracted by the keyword extraction unit from a storage unit that stores category information associated with an accumulating unit for accumulating the appearance result of the keyword for each of the categories acquired by the category acquiring unit; and a presenting unit for presenting the accumulation result accumulated by the accumulating unit for each category, wherein the appearance of the keyword is provided. The result includes the duration of the topic of the category, and the integration unit is an utterance amount integration device that integrates the duration of the topic of the category associated with the keyword for each category.
In one embodiment of the present invention, in the above-described speech amount accumulating apparatus, a first axis indicating the duration of the topic of the category accumulated by the accumulating unit and a second axis indicating the number of appearances of the keyword accumulated by the accumulating unit The integrated result is presented by two axes.

ADVANTAGE OF THE INVENTION According to this invention, the speech amount integration|stacking apparatus, customer-service assistance apparatus, and program which can contribute to the analysis of a customer's attribute and a customer's situation are provided.

It is a figure showing a schematic structure of a customer service support system concerning this embodiment. It is a figure which shows an example of the category information of this embodiment. It is a figure which shows an example of operation|movement of the speech amount integration|stacking apparatus of this embodiment. It is a figure which shows an example of the conversation data of this embodiment. It is a figure which shows an example of the extraction result of the keyword by the keyword extraction part of this embodiment. It is a figure which shows an example of the accumulation result by the accumulation part of this embodiment. It is a figure which shows an example of the presentation of the integration result by the presentation part of this embodiment.

[Embodiment]
BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. First, the overall configuration of this embodiment will be described with reference to FIG.
FIG. 1 is a diagram showing a schematic configuration of a speech amount integration system 1 according to this embodiment. This speech amount accumulation system 1 accumulates the amount of speech at product showrooms and mail-order call centers for products and services such as automobiles, insurance, housing equipment, and custom-built homes. For example, the speech amount accumulation system 1 analyzes and presents the content of the conversation between the salesperson SP and the customer C during the customer service, thereby serving as a customer service support system that assists the improvement of the customer service ability of the salesperson SP. Function. The speech amount integration system 1 includes a speech amount integration device 10 , a storage unit 20 and a display device 30 .

Speech amount integration device 10 includes conversation data acquisition unit 110 , keyword extraction unit 120 , category acquisition unit 130 , integration unit 140 , and presentation unit 150 .
Note that the speech amount integrating device 10 functions as a customer service support device when the conversation data DCV is data indicating conversation during customer service.

The conversation data acquisition unit 110 acquires conversation data DCV during customer service. In this example, the conversation between the salesperson SP and the customer C is recorded by a recording device RD such as an IC recorder carried by the salesperson SP. In this example, the conversation data acquisition unit 110 acquires conversation data DCV recorded by the recording device RD via a storage medium, wireless communication, or wired communication.
Note that the conversation between the salesperson SP and the customer C may be picked up by a sound pickup device (for example, a wall microphone) installed at the place where the customer is served. Also, the recording device RD may be a component of the speech amount integration system 1 .
Further, the conversation data DCV may include not only the conversation during the customer service, but also the conversation between the sales representatives SP before and after the customer service. Conversation data DCV may be data in which voice other than face-to-face format (for example, telephone) is collected, or conversation other than voice (for example, character data such as e-mail and chat) is recorded. It may be data that has been

In this embodiment, a case in which sales representative SP1 and customers C1 to C3 are having a conversation about a house in a merchandise showroom will be described as an example. In the following description, the utterance CV of the salesperson SP is also referred to as utterance CV-SP, and the utterance CV of customer C is also referred to as utterance CV-C. In particular, the utterance CV of the salesperson SPn (n is a natural number) is also described as utterance CV-SPn, and the utterance CV of the customer Cn (n is a natural number) is also described as utterance CV-Cn.
In this example, the case where there is one salesperson SP and three customers C will be described, but the number is not limited to this. The number of sales representatives SP may be plural, and the number of customers C may be two or less or four or more.
Further, in the present embodiment, the speech amount accumulation system 1 is described as being targeted for conversations about houses in a product showroom, but the present invention is not limited to this example. The speech volume accumulation system 1 can also be applied to product showrooms and mail-order call centers for products and services such as automobiles, insurance, housing equipment, and custom-built homes.

Keyword extraction unit 120 extracts keyword KW from conversation data DCV acquired by conversation data acquisition unit 110 . An example of the keyword KW will now be described with reference to FIG.

FIG. 2 is a diagram showing an example of the category information DCG of this embodiment. The category information DCG is information in which a keyword KW and a topic category CAT are associated with each other.
In the example shown in the figure, the category information DCG has pre-registered keywords KW such as "disaster", "disaster", "earthquake", "earthquake resistance", "vibration control", "kitchen", "IH heater", and "breakfast". Among these keywords KW, for example, the keyword KW "disaster" is associated with the category CAT (hierarchy 1) "disaster prevention/prevention" and the category CAT (hierarchy 2) "general". Further, for example, the keyword KW "earthquake" is associated with the category CAT (level 1) "disaster prevention/preparation" and the category CAT (level 2) "seismic performance". Further, for example, the keyword KW "kitchen" is associated with the category CAT (hierarchy 1) "meal" and the category CAT (hierarchy 2) "equipment". Keywords KW not exemplified here are also associated with categories CAT.
Note that the correspondence between the keyword KW and the topic category CAT is not limited to this example. For example, one and the same keyword KW may be associated with a plurality of different categories CAT. As an example, the keyword KW "all electrification" may be associated with the category CAT (tier 1) "disaster prevention/preparation" and the category CAT (tier 1) "meals".

Returning to FIG. 1, the keyword extraction unit 120 segments the utterance CV included in the conversation data DCV and extracts words obtained as a result of segmentation as keywords KW. The words extracted here include predetermined keywords KW registered in advance in the category information DCG and keywords KW not registered in the category information DCG.
Note that the keyword extracting unit 120 may collect (name collate) a plurality of keywords KW into one keyword KW based on information in a thesaurus database (not shown). For example, when "living room" and "living room" are extracted as the keywords KW, "living room" may be extracted as the keyword KW by combining these "living room" and "living room".

Category acquisition unit 130 acquires category CAT associated with keyword KW extracted by keyword extraction unit 120 . In this example, the storage unit 20 stores category information DCG. The category acquisition unit 130 searches the category information DCG stored in the storage unit 20 using the keyword KW extracted by the keyword extraction unit 120 as a search key. The category acquisition unit 130 acquires the category CAT associated with the keyword KW used as the search key as a search result.

The integration unit 140 integrates the appearance result RO of the keyword KW extracted by the keyword extraction unit 120 for each category CAT acquired by the category acquisition unit 130 . As a specific example, a case will be described in which the conversation data DCV includes the keyword KW "earthquake" twice and the keyword KW "kitchen" once. As described above, in the category information DCG, the keyword KW "earthquake" is associated with the category CAT (hierarchy 1) "disaster prevention/prevention" and the category CAT (hierarchy 2) "seismic performance". In the category information DCG, the keyword KW "kitchen" is associated with the category CAT (hierarchy 1) "meal" and the category CAT (hierarchy 2) "facility".
In this specific example, the accumulating unit 140 performs the category CAT (tier 1) “disaster prevention/preparation” twice, the category CAT (tier 2) “seismic performance” twice, and the category CAT (tier 1) “meals”. and once for the category CAT (level 2) "equipment" as the appearance result RO of the keyword KW.

The presentation unit 150 presents the integration result RS integrated by the integration unit 140 for each category CAT. Here, the presentation method includes an image method, an audio method, a vibration method, and the like. In this example, the presentation unit 150 outputs the integration result RS to the display device 30 to display the integration result RS on the display DP included in the display device 30 .

Note that the speech amount integrating device 10 may include a speaker determination unit 160 . The speaker determination unit 160 determines the speaker SPK of the conversation indicated by the conversation data DCV acquired by the conversation data acquisition unit 110 . In this case, the integration unit 140 integrates the appearance result RO of the keyword KW for each speaker SPK determined by the speaker determination unit 160 .

Further, the speech amount integrating device 10 may include a speech place determination unit 170 . The utterance place determination unit 170 determines the utterance place PL of the conversation indicated by the conversation data DCV acquired by the conversation data acquisition unit 110 . In this case, the integration unit 140 integrates the occurrence results RO of the keyword KW for each speech location PL determined by the speech location determination unit 170 .

[Operation of speech amount integrating device 10]
Next, an example of the operation of the speech amount integrating device 10 will be described with reference to FIG.
FIG. 3 is a diagram showing an example of the operation of the speech amount integrating device 10 of this embodiment.

(Step S10) The conversation data acquisition unit 110 acquires the conversation data DCV from the recording device RD. FIG. 4 shows an example of the conversation data DCV about the conversation between the sales representative SP and the customer C in this specific example.

FIG. 4 is a diagram showing an example of conversation data DCV of this embodiment. The salesperson SP1 utters "Explanation 1" at time t1. This "description 1" is recorded in the conversation data DCV as an utterance CV-SP11. Customer C1 utters "question 1" at time t2. This "question 1" is recorded in the conversation data DCV as an utterance CV-C11. Salesperson SP1 utters "Answer 1" at time t3. This "answer 1" is recorded in the conversation data DCV as the utterance CV-SP12.
Next, the salesperson SP1 utters "Explanation 2" at time t4. This "description 2" is recorded in the conversation data DCV as the utterance CV-SP13. Customer C2 utters "impression 2" at time t5. This "impression 2" is recorded in the conversation data DCV as an utterance CV-C21.
Next, salesperson SP1 utters "Explanation 3" at time t6. This "description 3" is recorded in the conversation data DCV as the utterance CV-SP14. Customer C3 utters "question 3" at time t7. This "question 3" is recorded in the conversation data DCV as an utterance CV-C31. Salesperson SP1 utters "Answer 2" at time t8. This "answer 2" is recorded in the conversation data DCV as the utterance CV-SP15. Customer C1 utters "question 4" at time t9. This "question 4" is recorded in the conversation data DCV as an utterance CV-C12. Salesperson SP1 utters "Answer 4" at time t10. This "answer 4" is recorded in the conversation data DCV as the utterance CV-SP16.
Note that the recording device RD may record only the sound period in which the salesperson SP or the customer C speaks, or may record the soundless period in which the salesperson SP or the customer C does not speak. may

(Step S20) Returning to FIG. 3, the keyword extraction unit 120 extracts the keyword KW from the conversation data DCV acquired by the conversation data acquisition unit 110. FIG. A specific example of the extraction result of the keyword KW extracted by the keyword extraction unit 120 will be described with reference to FIG.

FIG. 5 is a diagram showing an example of the extraction result of the keyword KW by the keyword extraction unit 120 of this embodiment. In this example, the keyword extraction unit 120 segments the utterance CV of the conversation data DCV and extracts the keyword KW from the conversation data DCV. The keyword extraction unit 120 arranges the extracted keywords KW in chronological order, and associates the keywords KW, the utterance IDs, and the utterance times of the keywords KW with each other.
In the specific example shown in the figure, the keyword KW "earthquake" is uttered twice in the utterance CV-SP11. The keyword extraction unit 120 associates the utterance ID "ID11", the keyword KW "earthquake", and the utterance time "time t1-1" with each other. The keyword extraction unit 120 associates the utterance ID "ID12", the keyword KW "earthquake", and the utterance time "time t1-2" with each other.
In the utterance CV-SP12, the keyword KW "earthquake" is uttered once, the keyword KW "earthquake resistance" is uttered once, and the keyword KW "disaster" is uttered once. The keyword extraction unit 120 associates the utterance ID "ID31", the keyword KW "earthquake", and the utterance time "time t3-1" with each other. The keyword extraction unit 120 associates the utterance ID "ID32", the keyword KW "earthquake resistance", and the utterance time "time t3-2" with each other. The keyword extraction unit 120 associates the utterance ID "ID33", the keyword KW "disaster", and the utterance time "time t3-3" with each other.
In the utterance CV-SP13, the keyword KW "breakfast" is uttered once, and the keyword KW "kitchen" is uttered once. The keyword extraction unit 120 associates the utterance ID "ID41", the keyword KW "breakfast", and the utterance time "time t4-1" with each other. The keyword extraction unit 120 associates the utterance ID "ID42", the keyword KW "kitchen", and the utterance time "time t4-2" with each other.

(Step S30) Returning to FIG. 3, the category acquisition unit 130 searches the category information DCG of the storage unit 20 to acquire the category CAT associated with the keyword KW extracted by the keyword extraction unit 120. FIG.
In the example shown in FIG. 5, the category acquisition unit 130 obtains the category CAT (hierarchy 1) “disaster prevention/preparation” and the category CAT (hierarchy 2) “seismic performance” for the keyword KW “earthquake” of the utterance ID “ID11”. ” and get. The category acquisition unit 130 acquires the category CAT (hierarchy 1) “disaster prevention/preparation” and the category CAT (hierarchy 2) “seismic performance” for the keyword KW “earthquake” of the utterance ID “ID12”. Similarly, category acquisition unit 130 acquires category CAT (hierarchy 1) and category CAT (hierarchy 2) for keyword KW for each utterance ID.

(Step S40) Returning to FIG. 3, the integrating section 140 determines the category CAT to which the keyword KW belongs based on the relationship between the keyword KW and the category CAT acquired in step S30. As an example, the accumulating unit 140 determines that the category CAT to which the keyword KW "earthquake" belongs is "disaster prevention/preparation" (tier 1) and "earthquake resistance performance" (tier 2).

(Step S50) The accumulation unit 140 accumulates the appearance results RO of the keyword KW for each category CAT. As an example, the appearance result RO of the keyword KW includes the number of appearances ON of the keyword KW. In this case, the accumulator 140 accumulates the appearance count ON of the keyword KW for each category CAT. FIG. 6 shows an example of the accumulation result of the appearance count ON of the keyword KW by the accumulation unit 140 .

FIG. 6 is a diagram showing an example of the integration result RS by the integration unit 140 of this embodiment. In the case of the specific example shown in FIG. 5 described above, the number of occurrences ON of the keyword KW "disaster" is one. In this case, the integration unit 140 calculates "once" as the integration result RS for the keyword KW "disaster". Also, in this specific example, the number of occurrences ON of the keyword KW "earthquake" is two. In this case, the integration unit 140 calculates "twice" as the integration result RS for the keyword KW "earthquake". Similarly, the integration unit 140 calculates the integration result RS for each keyword KW.

Further, the accumulating unit 140 may add up the accumulation results RS for which the number of appearances of the keyword KW is ON for each category CAT to which the keyword KW belongs. In the above specific example, the keyword KW "disaster", the keyword KW "earthquake", and the keyword KW "earthquake resistance" all belong to the category CAT (hierarchy 1) "disaster prevention/preparation". In this case, the accumulating unit 140 counts the keyword KW “disaster” once with the appearance count ON, the keyword KW “earthquake” with the appearance count ON three times, and the keyword KW “earthquake resistance” with the appearance count ON once. By summing up, "5 times" is calculated as the summing result RC.
That is, the integration unit 140 outputs the addition result RC obtained by adding together the integration results RS having the same category CAT acquired by the category acquisition unit 130 .

(Step S60) Returning to FIG. 3, the integrating section 140 integrates the topic duration DR of the category CAT associated with the keyword KW. In the above-described specific example, from time t1 to time t3, that is, from utterance CV-SP11 to utterance CV-SP12, the topic of "disaster prevention/preparation" continues. In this example, assuming that the period from time t1 to time t3 is 5 minutes, the accumulator 140 calculates the duration DR of the topic related to the category CAT (level 1) "disaster prevention/preparation" as shown in FIG. Calculate "5 minutes". Further, the topic of "meal" continues from time t4 to time t5, that is, from utterance CV-SP13 to utterance CV-C21. In this example, assuming that the period from time t4 to time t5 is 10 minutes, the accumulator 140 sets the duration DR of the topic related to the category CAT (level 1) "meals" to "10 minutes" as shown in FIG. ”.
Note that the integration unit 140 can calculate the duration of the topic using various techniques. For example, the accumulating unit 140 calculates the time from the start of one topic to the start of another topic as the duration of the topic. In this case, the accumulating unit 140 calculates the topic duration based on the switching timing of the category CAT of the utterance CV when each utterance CV included in the conversation data DCV is arranged in time series. As an example, the accumulating unit 140 calculates the time from when the category CAT of the utterance CV is switched to a certain category CAT to when it is switched to another category CAT as the duration of the topic. That is, the accumulating unit 140 calculates the duration of the same category CAT as the duration of the topic. In this case, if the duration of the same category CAT is shorter than a predetermined threshold value (for example, 10 seconds or less), the accumulator 140 assumes that the category CAT has not been switched, and that the topic has not changed. A duration may be calculated.

If the speech amount integrating device 10 includes the speaker determination unit 160, the speaker determination unit 160 associates the integration result RS with the speaker SPK. In the case of the specific example shown in FIG. 4 described above, the speaker determination unit 160 determines that the utterances CV-SP11 to SP16 are the utterances CV by the salesperson SP1. Further, the speaker determination unit 160 determines that the utterance CV-C11 is the utterance CV by the customer C1. In the same way, the speaker determination unit 160 determines the speaker SPK for each utterance CV.
As shown in FIG. 5, the speaker determination unit 160 reflects the result of determining the speaker SPK in the integration result RS. In this case, the integration unit 140 may integrate the appearance result RO of the keyword KW for each speaker SPK determined by the speaker determination unit 160 .

Further, when the speech amount integrating device 10 includes the speech place determination unit 170, the speech place determination unit 170 associates the integration result RS with the speech place PL. In the specific example shown in FIG. 4 described above, it is assumed that the conversation takes place in the living room from time t1 to time t3, and that the conversation takes place in the dining kitchen from time t4 to time t10. In this case, the utterance place determination unit 170 determines that the utterance place PL of the utterance CV from time t1 to time t3 is "living room (LB)", and that the utterance place PL of the utterance CV from time t4 to time t10 is " Dining Kitchen (DK)" is determined respectively.
As shown in FIG. 5, the utterance place determination unit 170 reflects the result of determining the utterance place PL in the integration result RS. In this case, the integrating section 140 may integrate the appearance result RO of the keyword KW for each utterance place PL determined by the utterance place determination section 170 .

(Step S70) Returning to FIG. 3, the presentation unit 150 presents the integration result RS integrated by the integration unit 140 for each category CAT. An example of presentation of the integration result RS by the presenting unit 150 will be described with reference to FIG.

FIG. 7 is a diagram showing an example of presentation of the integration result RS by the presentation unit 150 of the present embodiment. In this example, the presentation unit 150 includes a time axis AXT (first axis) indicating the duration DR of the topic of the category CAT accumulated by the accumulation unit 140, and a number indicating the appearance count ON of the keyword KW accumulated by the accumulation unit 140. The integration result RS is presented by the axis AXC (second axis).
That is, the presentation unit 150 presents the integration result RS with the topic duration DR plotted on the horizontal axis of the graph and the appearance count ON of the keyword KW related to the topic on the graph plotted on the vertical axis. As a result, the display DP of the display device 30 displays a figure (for example, a triangle) having a width corresponding to the duration DR of the topic and a height corresponding to the frequency ON of appearance of the keyword KW related to this topic.

Note that the presenting unit 150 calculates the number of appearances ON of the keyword KW for each keyword KW, the number of appearances ON of the keyword KW for each category CAT, and the topic duration DR may be presented in combination as appropriate. Further, when the category CAT is hierarchized, the presenting unit 150 may present these accumulation results RS for each hierarchy of the category CAT. For example, the presentation unit 150 may present the appearance count ON of the keyword KW related to the category CAT (level 1) and the appearance count ON of the keyword KW related to the category CAT (level 2).

Further, when the speech amount integrating device 10 includes the speaker determination unit 160, the presenting unit 150 may present the integration result RS for a specific speaker SPK among the integration results RS. As an example of this case, the presentation unit 150 presents the integration result RS of the salesperson SP without presenting the customer C's integration result RS.

Further, when the speech amount integrating device 10 includes the speech place determination unit 170, the presentation unit 150 may present the integration result RS for a specific speech place PL among the integration results RS. As an example of this case, the presentation unit 150 presents the integration result RS of the utterance CV in the dining kitchen without presenting the integration result RS of the utterance CV in the living room.

Further, in the example described above, the speech amount integrating device 10 presents the integration result RS to the salesperson SP after the conversation between the salesperson SP and the customer C has ended, that is, after the reception has ended. has been described, but it is not limited to this. The speech amount integrating device 10 may present the integration result RS to the salesperson SP while the salesperson SP is serving the customer. That is, the speech amount integrating device 10 may present the integration result RS in real time to the salesperson SP who is serving the customer.

[Summary of embodiment]
As described above, the speech amount integrating device 10 of the present embodiment analyzes topics of conversation during customer service, and presents appearance results RO of keywords KW for each topic category CAT. According to the speech amount integrating device 10 configured in this way, it is possible to indicate points in the conversation in which the customer C is interested by the presented integration result RS. That is, the speech amount integrating device 10 can contribute to the analysis of the customer's attribute and the customer's situation by providing the integration result RS. Further, according to the speech amount integrating device 10 of the present embodiment, when there is essential transmission information to be transmitted to the customer C by the salesperson SP, whether or not the essential transmission information is transmitted to the customer C during service is determined. can contribute to the analysis of
In addition, according to the speech amount integrating device 10 of the present embodiment, since it is possible to accumulate the integration results RS of a plurality of customer service opportunities, changes in the topic over time (for example, changes in topics in which the customer C is interested) can contribute to the analysis of

Further, the accumulator 140 of the speech amount accumulating device 10 of the present embodiment accumulates the topic duration DR of the category CAT associated with the keyword KW for each category CAT. According to the speech amount integrating device 10 configured in this manner, for example, a topic with a longer duration DR can be presented as a topic in which the customer C is interested.

In addition, the accumulating unit 140 of the speech amount accumulating device 10 of the present embodiment accumulates the appearance count ON of the keyword KW for each category CAT. According to the speech amount integrating device 10 configured in this way, it is possible to present a topic in which the keyword KW appears more often ON as a topic in which the customer C is interested.

In addition, the presentation unit 150 of the speech amount accumulation device 10 of the present embodiment has a time axis AXT (first axis) indicating the duration DR of the topic of the category CAT accumulated by the accumulation unit 140, and the keywords accumulated by the accumulation unit 140. The integrated result RS is presented along with a frequency axis AXC (second axis) indicating the number of occurrences ON of KW. According to the speech amount integrating device 10 configured in this way, when a conversation includes a plurality of topics, the topics can be compared by the area of the graph. , the integration result RS can be presented in a more comprehensible manner.

In addition, the presenting unit 150 of the speech amount integrating device 10 of the present embodiment presents a result obtained by summing the integration results RS having the same category CAT acquired by the category acquiring unit 130 .
Here, in general, the finer the type of category CAT is classified, the more detailed analysis of the results is possible. On the other hand, when the types of category CAT are finely classified, if the accumulation result RS finely accumulated corresponding to the type of category CAT is presented, the accumulation result RS becomes difficult for the viewer to understand. There is
According to the speech amount integrating device 10 of the present embodiment, since the integration results RS having the same category CAT are added together, the integration results RS can be made easier to understand even if the categories CAT are finely classified. can be presented.
In addition, the speech amount integrating device 10 of the present embodiment has hierarchical categories CAT, and can present the integration result RS for any hierarchy of the hierarchical categories CAT. That is, according to the speech amount integrating device 10, it is possible to present a rough integration result RS obtained by integrating the upper hierarchy category CAT and a detailed integration result RS obtained by integrating the lower hierarchy category CAT. According to the speech amount integrating device 10 configured in this way, it is possible to achieve both the comprehensibility of the integration result RS and the utilization of the integration result RS for more detailed analysis.

In addition, the speech amount integrating device 10 of this embodiment includes a speaker determination unit 160 . According to the speech amount integrating device 10 configured in this manner, for example, it is possible to analyze who is the key person in the group of the customer C (for example, the key person is the person who speaks more often). can contribute. Further, according to the speech amount integrating device 10, for example, when there are a plurality of sales representatives SP, it is possible to contribute to the analysis of trends in topics for each sales representative SP.
In addition, by specifying the customer C who is in conversation, the speech amount integrating device 10 can contribute to, for example, analysis of whether or not the customer C has returned at a later date (that is, whether or not the customer is a repeat customer). can. In the case where customer C is a repeat customer, by analyzing the topics of the first visit to customer C, it is possible to analyze what kind of topics are effective in making the customer a repeat customer. The speech amount integrating device 10 can contribute to the analysis of the topic of repeat customers by providing the integration result RS.

Further, in the speech amount integrating apparatus 10 of the present embodiment, the presentation unit 150 presents the integration result RS for a specific speaker SPK among the integration results RS. According to the speech amount integrating apparatus 10 configured in this way, for example, only the integration result RS of the salesperson SP and the customer C can be selected and presented. Further, for example, when a skilled salesperson SP and a new salesperson SP are serving customers together, the speech amount integrating device 10 presents only the integration result RS of the skilled salesperson SP. , presenting only the integration result RS of the new salesperson SP, or presenting these two integration results RS at the same time.

Further, the speech amount integrating device 10 of the present embodiment includes a speech place determination unit 170 . According to the speech amount integrating device 10 configured in this way, for example, by providing the integration result RS, it contributes to the analysis of whether or not the topic to be conveyed to the customer C is being conveyed at a specific speech place PL. can do.

In addition, the presentation unit 150 of the speech amount integrating device 10 of the present embodiment presents the integration result RS for a specific utterance place PL among the integration results RS. According to the speech amount integrating apparatus 10 configured in this manner, by providing the integrated result RS, the topic to be communicated from the salesperson SP to the customer C is conveyed to the customer C in a specific place such as a living room or a dining kitchen, for example. can contribute to the analysis of whether or not

Although embodiments of the present invention and variations thereof have been described above, these embodiments and variations thereof are presented as examples and are not intended to limit the scope of the invention. These embodiments and modifications thereof can be implemented in various other forms, and various omissions, replacements, and modifications can be made without departing from the scope of the invention. These embodiments and their modifications are included in the scope and gist of the invention, and are included in the scope of the invention described in the claims and equivalents thereof.

Each of the devices described above has a computer inside. The process of each process of each device described above is stored in a computer-readable recording medium in the form of a program, and the above process is performed by reading and executing this program by a computer. Here, the computer-readable recording medium refers to magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, semiconductor memories, and the like. Alternatively, the computer program may be distributed to a computer via a communication line, and the computer receiving the distribution may execute the program.

Further, the program may be for realizing part of the functions described above.
Further, it may be a so-called difference file (difference program) that can realize the above-described functions in combination with a program already recorded in the computer system.

DESCRIPTION OF SYMBOLS 1... Speech amount integration system (customer service support system), 10... Speech amount integration device (customer service support apparatus), 110... Conversation data acquisition part, 120... Keyword extraction part, 130... Category acquisition part, 140... Integration part, 150... Presentation unit 160 Speaker determination unit 170 Speaking place determination unit 20 Storage unit 30 Display device AXC Frequency axis AXT Time axis C Customer CAT Category DCG Category information DCV...conversation data, DR...duration, KW...keyword, ON...number of appearances, PL...utterance location, RC...total result, RS...accumulation result, RO...appearance result, SP...salesperson, SPK...speaker

Claims (12)

a conversation data acquisition unit that acquires conversation data;
a keyword extraction unit for extracting keywords from the conversation data acquired by the conversation data acquisition unit;
a category acquisition unit for acquiring the category associated with the keyword extracted by the keyword extraction unit from a storage unit in which the keyword and topic category are associated and stored as category information;
an utterance place determination unit that determines the utterance place of the conversation indicated by the conversation data acquired by the conversation data acquisition unit;
an accumulating unit for accumulating the appearance result of the keyword extracted by the keyword extracting unit for each utterance place determined by the utterance place determining unit and for each category acquired by the category acquiring unit;
a presenting unit that presents the integration results integrated by the integrating unit for each of the categories;
A speech amount integrating device. The occurrence results of the keyword include the duration of the topic of the category,
The integration unit
The speech amount integrating device according to claim 1, wherein the duration of topics in the categories associated with the keywords is integrated for each category. The appearance result of the keyword includes the number of appearances of the keyword,
The integration unit
3. The speech amount integrating device according to claim 1, wherein the number of appearances of the keyword is integrated for each category. The presentation unit
The integrated result is presented by a first axis indicating the duration of the topic of the category integrated by the integrating unit and a second axis indicating the frequency of appearance of the keyword integrated by the integrating unit. Item 4. The speech amount integration device according to any one of items 3 . The integration unit
outputting a total result obtained by adding together the integration results having the same category acquired by the category acquiring unit;
The presentation unit
The speech amount integration device according to any one of claims 1 to 4, wherein the integration result output by the integration unit is presented. a speaker determination unit that determines a speaker of the conversation indicated by the conversation data acquired by the conversation data acquisition unit;
The integration unit
The speech amount integrating device according to any one of claims 1 to 5, wherein the appearance result of the keyword is integrated for each speaker determined by the speaker determination unit. The presentation unit
The speech amount integrating device according to any one of claims 1 to 6, wherein, among the integrated results, the integrated result for a specific speaker is presented. The presentation unit
The speech amount integrating device according to any one of claims 1 to 7, wherein the integration result for a specific speech location is presented among the integration results. Equipped with the speech amount integrating device according to any one of claims 1 to 8 ,
The said conversation data is the data which show the conversation in service. Customer service assistance apparatus. In the computer equipped with the speech amount integration device,
a conversation data acquisition step for acquiring conversation data;
a keyword extraction step of extracting keywords from the conversation data acquired in the conversation data acquisition step;
a category obtaining step of obtaining the category associated with the keyword extracted in the keyword extracting step from a storage unit in which the keyword and topic category are associated with each other and stored as category information;
an utterance place determination step of determining the utterance place of the conversation indicated by the conversation data acquired in the conversation data acquisition step;
an accumulating step of accumulating the appearance result of the keyword extracted in the keyword extracting step for each utterance place determined in the utterance place determining step and for each category obtained in the category obtaining step;
a presenting step of presenting the integration result integrated in the integrating step for each category;
program to run the a conversation data acquisition unit that acquires conversation data;
a keyword extraction unit for extracting keywords from the conversation data acquired by the conversation data acquisition unit;
a category acquisition unit for acquiring the category associated with the keyword extracted by the keyword extraction unit from a storage unit in which the keyword and topic category are associated and stored as category information;
an accumulating unit for accumulating the occurrence results of the keywords extracted by the keyword extracting unit for each of the categories acquired by the category acquiring unit;
a presenting unit that presents the integration results integrated by the integrating unit for each of the categories;
with
The occurrence results of the keyword include the duration of the topic of the category,
The integration unit
A speech amount integrating device that integrates, for each category, the duration of topics of the categories associated with the keywords. The presentation unit
The accumulation result is presented by a first axis indicating the duration of the topic of the category accumulated by the accumulation unit and a second axis indicating the number of appearances of the keyword accumulated by the accumulation unit.
The speech amount integration device according to claim 11.

Images