JP2002297646A - System, method, and program for service - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a person in charge of support to input a keyword to a database without operating a keyboard. SOLUTION: The person 20 in charge of support rearranges telephone consultation contents of a user 10 and vocally repeats them in proper timing by using specific keyword terms. A voice recognition part 31 recognizes and converts the repetition voice into consultation contents data. A term extraction part 32 extracts keyword data from the consultation contents data and after a DB information decision part 33 judges whether the extracted data are already registered, the data are inputted as a keyword to a database 50. The repetition voice when the service provider confirms the consultation contents is effectively used to input a handling method and a support history. Further, the voice recognition part 31 previously recognizes and learns the voice of each person in charge of support and the recognition rate can be increased.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a service system, a method, and a program to be executed by a computer in a service system suitable for use when a user consults a service center regarding a defect of a product. is there.

[0002]

2. Description of the Related Art Conventionally, when a problem has occurred in a product such as an information processing device owned by a user, the user inquires of a service center by telephone, and a support person there responds to the content of the problem and Information about the user is retrieved, and this information is used as a keyword to search a database or to register the database. Then, the cause and the solution of the problem are obtained by searching, and are transmitted to the user. In this case, the support staff operates the keyboard of the PC and inputs necessary information in accordance with the content spoken by the user.

[0003]

However, when the support staff operates the keyboard to input necessary information, it takes time because the user operates the keyboard after the user has finished speaking. In addition, since a user cannot often explain the situation of a product on a telephone in a coherent manner, input information is likely to be insufficient or redundant, and there is a problem that input cannot be performed smoothly. Therefore, it is conceivable that the voices of the users are recognized by voice and directly input to the database. However, it is difficult to accurately recognize the voices of various users with the current voice recognition technology.

[0004] In addition, some support personnel can efficiently use the support history information accumulated in the database to provide efficient support, and others who are not accustomed to support work. Due to differences in the level of service, it was not always possible to always provide high service quality. Furthermore, if the response to the consultation is prolonged, the support technician does not remember the final conclusion (resolution) of the consultation until after the response, so when entering the support history, Sometimes it was incomplete.

The present invention has been made in view of the above circumstances, and has as its object to provide a service system capable of promptly providing high-quality support services to users.

[0006]

In order to achieve the above object, a service system according to the present invention recognizes a voice spoken by a support person of a consultation content of a user and converts it into consultation content data. Means, information extraction means for extracting information such as text data about a predetermined word from the consultation content data, and input means for inputting information such as the extracted word data to a database as a search key It is.

[0007] The service method according to the present invention recognizes the voice spoken by the support staff and converts the consultation content of the user into consultation content data, and extracts predetermined word data from the consultation content data. Word data is entered as a keyword into a database.

[0008] The program according to the present invention recognizes the voice of a user who consults a consultation content and recognizes the voice, and converts the content into consultation content data. Is a program for causing a computer to execute an information extraction process of extracting information of the above type and an input process of inputting the extracted information into a database as a search key.

[0009]

Therefore, according to the present invention, a high recognition rate can be obtained by making the voice recognition means recognize and learn the voice of the support person in advance, so that it can be used as a search key for accessing the database. Key words can be obtained with high precision, keyword input can be done quickly and accurately, and support personnel do not need to operate the keyboard, just summarize the user's inquiry and repeat it, Input can be done easily. In addition, since only keyword data is extracted, access to the database can be ensured,
No useless data is accumulated. Further, the support history information can be accumulated by the voice at the time of repetition.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a service system according to the first embodiment of the present invention. In the present embodiment, a case will be described as an example where a user makes an inquiry to a service center by telephone about a defect of a product and the like, and a support person who performs a support service there responds.

In FIG. 1, a user telephone set 11 used by a user 10 and a support person 2 for consulting the user.
0 is connected via a public line (not shown) to the telephone number 21 in charge. The person in charge telephone 21 is connected to a support system 30 used by the support person 20. The support system 30 includes the network 4
0 is connected to another support system 30 and a database (hereinafter abbreviated as DB) 50. The other support system 30 is provided with a person-in-charge telephone 21 of another support person.

In the support system 30, the voice recognition unit 31 sends the support person 20 from the person in charge telephone 21.
Is input, the input voice is converted to text and text data representing the content of the consultation (hereinafter referred to as consultation content data) is output. In addition, the speech recognition unit 31 extracts a predetermined word as a technical term serving as a keyword for accessing the DB 50 from the consultation content data so that a speech can be input according to the support skill level of the support staff. A part 32 is provided.

When registering information based on the consultation content data input from the voice recognition unit 31 in the DB 50, a database information determination unit (hereinafter referred to as a DB information determination unit) 33 makes a determination for avoiding duplicate registration. The information search request unit 34 performs a process of searching the DB 50 for the information required by the support person 20 or searching the DB 50 based on the problem data extracted by the problem extraction unit 35 described later. Problem extraction unit 35
Extracts problem data relating to a product defect from the consultation content data from the voice recognition unit 31. Solution search part 3
6 is a DB for storing the solution of the problem extracted by the problem extracting unit 35.
Search and acquire 50.

The network information transmission / reception unit 37 exchanges information with the DB 50 and other support systems 30 via the network 40 in response to requests from the above units. The information output unit 38 outputs information from the DB 50 and other support systems 30 to the support staff 20. DB
50 includes a solution to the above problem, support history information,
User information, technical terms and their similar words, and other necessary information are stored.

Next, the operation of the above configuration will be described. When there is an inquiry from the user 10 through the user telephone 11, the support person 20 responds using the person in charge telephone 21, and the user explains a defect of the product or the like. At that time, the support person 20 arranges the contents explained by the user while listening to the story of the user, summarizes the contents, and repeats it at an appropriate timing using technical terms (keyword terms that can be used as search keys). Check with the user. The voice of the support staff who repeats the speech is input from the staff phone 21 to the speech recognition unit 31 and is recognized. in this case,
The voice recognition unit 31 can identify the speaker by preliminarily recognizing and learning the voices of a limited number of support staff.
Speech recognition tailored to the speaker can be performed with high accuracy of about 100%. It should be noted that the support staff should use a terminology as much as possible when inputting.

Necessary word data is extracted by the term extraction unit 32 from the consultation content data whose speech has been recognized. For example, when the voice at the time of repetition is "I'm XX," only XX is extracted. The extracted data is stored in a DB
It is determined whether or not the data has already been registered in B50, and only the unregistered data is
Is input to the DB 50 via the network 40.

The term extraction unit 32 responds to a case where a low-level support technician speaks another similar word (including a synonym or the like) without repetition using the correct technical term. It has a function of searching for a technical term to be performed, and replacing the similar word part of the data with the technical term and outputting it.

On the other hand, the problem extracting unit 35 extracts predetermined problem data indicating a problem such as a product defect from the consultation content data, and the solution searching unit 36 searches the DB 50 for the problem based on the extracted problem data. Search for cause and solution. The retrieved cause / solution is notified from the information output unit 38 to the support person 20 by display or the like. The support person can inform the user of the displayed cause and solution by telephone. Further, the voice of the support staff at that time can be recognized by voice and the information can be recorded in the DB 50 as a support history. Thereby, the support history information can be accumulated every time the user is supported.

According to the above-described embodiment, since the voice of the repetition by the support person can be recognized with high accuracy, the voice at the time of repetition can be effectively used without wasting, and the voice of the repetition can be effectively used. By extracting a predetermined word and using it as a keyword, access to the DB 50 can be performed quickly and accurately. Therefore, the support person does not need to perform any keyboard operation, and the input operation can be performed extremely easily. In addition, the function of the term extraction unit 32 enables even a low-level support person who is not familiar with technical terms to deal with repetition without using special terms.

Further, the DB information determination unit 33
It is possible to prevent the useless information overlapping with 0 from being registered. In addition, the problem extraction unit 35 automatically extracts a problem from the voice of the support staff, obtains a cause and a solution based on the extracted problem, and informs the support staff.

Since information is sequentially input to the DB 50 during support, support history information is accumulated and can be used for the next support. Further, a veteran support staff can use the accumulated support history information to educate a low-level support staff via a network.

FIG. 2 is a block diagram showing a service system according to a second embodiment of the present invention. Parts corresponding to those in FIG. 1 are denoted by the same reference numerals, and redundant description is omitted.
In FIG. 2, the term extraction unit 3 of the speech recognition unit 31 in FIG.
2 is omitted, and a person in charge level determination unit 39 is used instead.
Is provided.

When a low-level support technician speaks in non-technical terms and determines that the level is low, the staff level determination section 39 provides a service through the information output section 38 when the level is understood to be low. Request that the person 20 paraphrase using technical terms. For this purpose, the staff level determination unit 39 has a search function for understanding the words of the support staff.

Therefore, according to the present embodiment, a low-level support person can be trained to become proficient in technical terms, and the level of the support person can be improved. As a result, the rate of inputting appropriate terms into the information search request unit 34 and the problem extraction unit 35 is increased, and registration and search can be more reliably speeded up.

Next, a third embodiment of the present invention will be described. In this embodiment, in the service center, a plurality of support staff members provide support for each user's consultation by telephone, but if the support staff member cannot solve the problem, another support staff member It can be replaced with

For example, when a certain period of time has passed since the start of support, or in response to a request from a support person who is currently responding to the user, another support person (a vacant person, a veteran person, The person who responded last time). In that case, the first support person may continue with the support. For example, another support staff member can grasp the support process by viewing the response status converted into the text, and can be replaced at an appropriate time.

FIG. 3 is a flowchart showing an operation related to the change of two support personnel A and B. here,
For example, since the person in charge B has a higher level than the person in charge A, B is assumed to be in charge of A in priority. First, person in charge A,
In the standby state of B (S1), a call for an inquiry is received from the user, and when B picks up the telephone (when B switches) or when B is not busy (S2, YE).
S), B is in charge of the inquiry (S3).
When there is no switching operation (when A picks up the telephone) or when B is busy (S2, NO), A is in charge, but a predetermined time elapses in that state, or A
Alternatively, if B switches arbitrarily (S4, YES), A
Is replaced with B (S5). If any switching operation is performed while B is in charge (S6, YES), B can replace A (S7). If NO in S4, S
Return to 2.

Therefore, according to the present embodiment, a plurality of support personnel can cooperate with each other via a network to provide a support service to one user, so that a more reliable solution can be found. Can be. In addition, it is possible to switch while continuing the support without having to temporarily hang up during the support and then change to another support person.

Next, a program according to an embodiment of the present invention will be described. The processing by the operation described above in FIG. 1 and FIG.
The program executed by the CPU in the above constitutes the program according to the present invention.

As a recording medium for recording this program, a magneto-optical disk, an optical disk, a semiconductor memory, a magnetic recording medium, or the like can be used.
M, RAM, CD-ROM, floppy (registered trademark) disk, memory card, and the like may be used.

The recording medium is a fixed medium such as a volatile memory such as a RAM inside a computer system serving as a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. The one that holds the time program is also included.

The above program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium.
The transmission medium refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.

The above program may be for realizing a part of the functions described above. Furthermore, what can realize the above-described function in combination with a program already recorded in the computer system, that is, a so-called difference file (difference program) may be used.

Therefore, this program is used in a system or apparatus different from the system or apparatus shown in FIGS. 1 and 2, and the computer of the system or apparatus executes this program, so that the functions described in each embodiment can be achieved. The same function and effect as those described above can be obtained, and the object of the present invention can be achieved.

[0035]

According to the first, ninth and eleventh aspects of the present invention, information such as predetermined word data is obtained from the consultation content data obtained by voice recognition of the voice of the user who repeats the consultation content. Extracting and inputting the word data into the database enables high-recognition speech recognition, which eliminates the need for support personnel to operate the keyboard, and simply reduces the consultation content of the user. By simply summarizing and repeating, a keyword can be quickly and reliably input when registering or searching in the database.

Since the support content is repeated during the support, the support status is automatically input, and complete support history information can be accumulated. Therefore,
By analyzing this history information, the shortest resolution process can be derived for the next inquiry.
Furthermore, since only data that is a predetermined keyword is input, access to the database can be reliably performed, and useless data is not accumulated.

According to the third and eleventh aspects of the present invention, problem data is extracted from consultation content data, a database is searched based on the problem data, and a solution to the problem is obtained. A solution can be automatically obtained, and prompt service can be provided.

According to the fourth and twelfth aspects of the present invention, since the searched solution is notified to the support person, the support person can know this and refer to it. According to the fifth and thirteenth aspects of the invention, the support person repeats the notified solution information, voice-recognizes the voice to obtain consultation content data, and stores the data extracted from the consultation content data in the database. Of the service performed this time can be stored as history information.

According to the invention of claims 6 and 14, when the support person repeats a word similar to the predetermined word, the word is replaced with the corresponding predetermined word.
Support staff with a low level of technical support can respond accurately and provide consistent services. According to the seventh and fifteenth aspects of the present invention, when the support staff repeats a word similar to the predetermined word, the support staff requests the support staff to rephrase the word to the predetermined word. While working, the level can be gradually increased without awareness.

According to the eighth aspect of the present invention, the support staff can be replaced with another support staff, so that the consultation can be performed in cooperation with the other support staff without interrupting the consultation with the user. And a solution can be reliably found. As described above, according to the present invention,
We can always provide high quality service.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a service system according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a service system according to a second embodiment of the present invention.

FIG. 3 is a flowchart illustrating an operation of a service system according to a third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 User 11 User telephone 20 Support person 21 Person in charge telephone 30 Support system 31 Voice recognition part 32 Term extraction part 33 DB information judgment part 34 Information search request part 35 Problem extraction part 36 Solution search part 37 Network information Transmission / reception section 38 Information output section 39 Person in charge level determination section 40 Network 50 Database

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 3/16 340 G06F 3/16 340A

Claims (15)

[Claims] 1. A voice recognition means for recognizing a voice spoken by a support person of a consultation content of a user and converting it into consultation content data; an information extracting means for extracting predetermined information from the consultation content data; An input unit for inputting the extracted information to a database as a search key. 2. The service system according to claim 1, wherein the consultation content data is text data, and the information is textualized word data. 3. A problem extracting means for extracting problem data corresponding to a problem in the consultation content from the consultation content data, and searching a database based on the extracted problem data to obtain solution information indicating a problem solving method. 3. The service system according to claim 1, further comprising a solution search means for acquiring. 4. The service system according to claim 3, further comprising a solution information notifying unit for notifying the support person of the searched solution information. 5. The support person repeats the notified solution information, the voice recognition means recognizes the voice, converts the voice into consultation content data, and converts the consultation content data to the input means through the information extraction means. 5. The service system according to claim 4, wherein the information is registered as support history information in a database. 6. The information extraction means according to claim 2, wherein when the support person repeats a word similar to the predetermined word, the information extracting means replaces the word data with a corresponding predetermined word. The service system according to claim 1. 7. A notifying means for notifying a support person to rephrase a word similar to the predetermined word to the predetermined word when the support person recites a word similar to the predetermined word. 7. The service system according to any one of 6. 8. A plurality of support personnel can be connected via a network, and each of the support personnel responds to a user's consultation by using an information terminal. Claims that support is provided in response to a request from another person, in response to a request from another support person, or when a predetermined time has elapsed from the start of consultation, by replacing or cooperating with another support person. The service system according to any one of claims 1 to 7. 9. A consultation content of a user recognizes a voice reproduced by a support person and converts it into consultation content data, and extracts data relating to predetermined information from the consultation content data.
A service method characterized by inputting data relating to extracted information to a database as a search key. 10. A voice recognition process for recognizing a voice spoken by a support person of a consultation content of a user and converting it into consultation content data, and an information extraction process for extracting data relating to predetermined information from the consultation content data. A program for causing a computer to execute an input process of inputting data relating to the extracted information to a database as a search key. 11. A problem extracting process for extracting problem data corresponding to a problem in the content of an inquiry from the voice data, searching a database based on the extracted problem data, and obtaining solution information indicating a problem solving method. The program according to claim 10, further comprising a solution search process. 12. The non-transitory computer-readable storage medium according to claim 11, further comprising a solution information notifying process for notifying a support person of the searched solution information. 13. A voice recognition process for recognizing voice when the support person repeats the notified solution information and converting the voice into consultation content data, and a word extraction process for extracting a predetermined word from the consultation content data. 13. The program according to claim 10, further comprising a registration process for registering the extracted word data as support history information in a database. 14. The word extracting process according to claim 1, further comprising the step of: when a support person repeats a word similar to the predetermined word, replacing the voice with a corresponding predetermined word. The program according to any one of claims 10 to 13. 15. A notifying process for notifying a support technician that the support technician rephrases a word similar to the predetermined word so as to rephrase the word to the predetermined vocabulary. 15. The program according to any one of 14.