JP6299877B2 - Analysis device, analysis method, and program - Google Patents

Description

  The present invention relates to an analysis apparatus, an analysis method, and a program.

  A customer service clerk or the like needs to make various statements in accordance with a predetermined manual or the like, for example, during accounting processing. Patent Documents 1 and 2 disclose a technique for analyzing voice data of a voice uttered by a store clerk and determining whether or not a predetermined statement is made.

JP 2004-192545 A Japanese Patent Laid-Open No. 7-37170

  By applying the techniques described in Patent Documents 1 and 2, it is possible to determine whether or not the store clerk is making a predetermined statement and evaluate the store clerk according to the result. Naturally, the higher the predetermined speech, the higher the evaluation. However, some of the predetermined statements to be made by the store clerk are not required depending on the actions on the customer side.

  For example, the store clerk may have to confirm with the customer whether or not there is a point card for receiving a point service or a discount coupon during the accounting process. However, if the customer voluntarily presents a point card or discount coupon before confirmation, the store clerk does not need to make a remark for confirmation.

  In addition, for example, the store clerk may have to make a statement recommending a promotional product during the accounting process. However, if the customer voluntarily orders the promotional product or possesses it for purchase before making a recommendation, the store clerk does not need to make a statement to recommend the promotional product.

  A similar situation can occur outside of the accounting process. For example, a similar situation may occur when processing an order from a customer using an input terminal such as a so-called handy terminal in a restaurant or a tavern.

  In spite of this situation, the store clerk is evaluated only depending on whether a predetermined statement (eg, a statement confirming the presence or absence of a point card or a statement recommending a promotional product) has been made. And proper evaluation is not possible. An object of the present invention is to provide a technique for analyzing whether or not a predetermined utterance is made in a predetermined case by a clerk or the like (for example, an operator of a POS terminal).

According to the present invention,
Voice data acquisition means for acquiring voice data of voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
A speech content determination means for analyzing the voice data at the time of the processing determined that the predetermined information has not been input, and determining whether or not a predetermined speech content is spoken;
Is provided.

Moreover, according to the present invention,
Computer
A voice data acquisition step of acquiring voice data of a voice issued by an operator operating the input terminal;
In the process for each customer, an input information determination step for determining whether or not predetermined information is input via the input terminal;
A speech content determination step of analyzing the audio data at the time of the processing determined to have the predetermined information not input and determining whether or not a predetermined speech content is spoken;
An analysis method for performing the above is provided.

Moreover, according to the present invention,
Computer
Voice data acquisition means for acquiring voice data of a voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
An utterance content determination unit that analyzes the voice data at the time of the processing when it is determined that the predetermined information has not been input, and determines whether or not a predetermined utterance content is uttered,
A program for functioning as a server is provided.

  According to the present invention, a technique for analyzing whether or not a predetermined statement is made in a predetermined case by a store clerk or the like (eg, an operator of a POS terminal) is realized.

  The above-described object and other objects, features, and advantages will become more apparent from the preferred embodiments described below and the accompanying drawings.

It is a figure which shows notionally an example of the hardware constitutions of the apparatus of this embodiment. It is an example of the functional block diagram of the analyzer of this embodiment. It is a figure which shows typically an example of the sales promotion product list information of this embodiment. It is a figure which shows typically an example of the utterance information of this embodiment. It is a figure which shows typically an example of the determination result of this embodiment. It is a figure which shows typically an example of the determination result of this embodiment. It is a flowchart which shows an example of the flow of a process of the analysis method of this embodiment. It is an example of the functional block diagram of the analyzer of this embodiment. It is a figure which shows typically an example of the evaluation criteria of this embodiment. It is a figure which shows typically an example of the utterance information of this embodiment. It is a figure which shows typically an example of the evaluation criteria of this embodiment. It is a figure which shows typically an example of the evaluation criteria of this embodiment. It is a figure which shows typically an example of the determination result of this embodiment. It is an example of the functional block diagram of the analyzer of this embodiment. It is a flowchart which shows an example of the flow of a process of the analysis method of this embodiment.

  First, an example of the hardware configuration of the apparatus according to the present embodiment will be described. Each unit included in the apparatus of the present embodiment includes a CPU (Central Processing Unit), a memory, and a program loaded in the memory (a program stored in the memory from the stage of shipping the apparatus in advance, a CD ( Compact Disc) and other storage media, and programs downloaded from servers on the Internet), storage units such as hard disks that store the programs, and any combination of hardware and software, centering on the network connection interface Realized. It will be understood by those skilled in the art that there are various modifications to the implementation method and apparatus.

  FIG. 1 is a diagram conceptually illustrating an example of a hardware configuration of an apparatus according to the present embodiment. As shown in the figure, the apparatus of the present embodiment includes, for example, a CPU 1A, a RAM (Random Access Memory) 2A, a ROM (Read Only Memory) 3A, a display control unit 4A, a display 5A, and an operation reception connected to each other via a bus 10A. Unit 6A, operation unit 7A, communication unit 8A, auxiliary storage device 9A, and the like. Although not shown, other elements such as an imaging device, an input / output interface connected to an external device by wire, a microphone, and a speaker may be provided.

  The CPU 1A controls the entire computer of the apparatus together with each element. The ROM 3A includes an area for storing programs for operating the computer, various application programs, various setting data used when these programs operate. The RAM 2A includes an area for temporarily storing data, such as a work area for operating a program. The auxiliary storage device 9A is an HDD (Hard Disc Drive), for example, and can store a large amount of data.

  The display 5A is, for example, a display device (LED (Light Emitting Diode) display, liquid crystal display, organic EL (Electro Luminescence) display, etc.). The display 5A may be a touch panel display integrated with a touch pad. The display control unit 4A reads data stored in a VRAM (Video RAM), performs predetermined processing on the read data, and then sends the data to the display 5A to display various screens. The operation reception unit 6A receives various operations via the operation unit 7A. The operation unit 7A includes operation keys, operation buttons, switches, a jog dial, a touch panel display, a keyboard, and the like. The communication unit 8A is wired and / or wirelessly connected to a network such as the Internet or a LAN (Local Area Network) and communicates with other electronic devices.

  Hereinafter, this embodiment will be described. Note that the functional block diagram used in the following description of the embodiment shows functional unit blocks rather than hardware unit configurations. In these drawings, each device is described as being realized by one device, but the means for realizing it is not limited to this. That is, it may be a physically separated configuration or a logically separated configuration. In addition, the same code | symbol is attached | subjected to the same component and description is abbreviate | omitted suitably.

<First Embodiment>
First, an outline of the present embodiment will be described. The analysis apparatus of this embodiment processes the processing for each customer as one unit (processing unit). In the present embodiment, an example will be described in which the analysis apparatus processes the accounting process for each customer as one unit (processing unit). Hereinafter, the accounting process for each customer may be simply referred to as “accounting process”.

  First, in the accounting process for each customer, the analysis apparatus according to the present embodiment determines whether or not predetermined information is input via a predetermined input terminal. In the present embodiment, the input terminal will be described as a so-called POS (Point of Sales) terminal (POS register). Predetermined predetermined information corresponds to information input to a POS terminal by scanning a point card or discount coupon presented by each customer, product information (product name, price, etc.) of a promotional product, etc. .

  When the analysis apparatus according to the present embodiment detects an accounting process in which predetermined information is not input, the analysis apparatus analyzes voice data of a voice uttered by an operator operating the POS terminal during the accounting process, and determines a predetermined message. Determine whether the content is spoken. The content of the remark is, for example, a remark for inquiring whether or not a point card or a discount coupon is possessed, a remark for recommending a promotional product, or the like.

  As described above, the analysis apparatus according to the present embodiment simply analyzes voice data and does not determine whether or not a predetermined statement has been made, but determines whether or not predetermined information is input via the POS terminal. Determination is made for each accounting process, and when an accounting process that has not been input is detected, the voice data at the time of the accounting process is analyzed to determine whether or not a predetermined statement has been made.

  FIG. 2 shows an example of a functional block diagram of the analysis apparatus 10 of the present embodiment. As illustrated, the analysis device 10 includes a voice data acquisition unit 11, an input information determination unit 12, and a statement content determination unit 13. Note that the analysis device 10 of the present embodiment is a POS terminal (POS register).

  The voice data acquisition unit 11 acquires voice data of a voice uttered by an operator who operates the POS terminal (analysis device 10).

  For example, a microphone is provided in the POS terminal (analysis device 10), and the voice of an operator who operates the POS terminal (analysis device 10) is input via the microphone, and is accumulated in the POS terminal (analysis device 10). May be. In addition, a microphone independent of the POS terminal (analysis device 10) may be connected to the POS terminal (analysis device 10) in a wired and / or wireless manner. The voice of the operator who operates the POS terminal (analysis device 10) may be input via the microphone and stored in the POS terminal (analysis device 10). For example, voice data may be accumulated while any operator is logged in to the POS terminal (analysis device 10). And the audio | voice data acquisition part 11 may acquire the audio | voice data accumulate | stored in the POS terminal (analysis apparatus 10).

  In addition, a recorder including a microphone and a storage device may be installed in the vicinity of the POS terminal (analysis device 10), or may be carried by an operator who operates the POS terminal. The operator's voice may be input to the recorder, and voice data may be accumulated. For example, the operator may log in to the POS terminal (analysis device 10), turn on the recorder while operating the POS terminal (analysis device 10), and accumulate voice data. In the case of the example, the audio data accumulated in the recorder are collectively input to the POS terminal (analysis device 10) at an arbitrary timing. After the input, the audio data acquisition unit 11 can acquire such audio data.

  Note that the audio data is associated with information that can specify which accounting process the audio is. For example, time information, date information, etc. are associated.

  The input information determination unit 12 determines whether or not predetermined information is input via the POS terminal (analysis device 10) in the accounting process for each customer. When an accounting process in which predetermined information is not input is detected, information specifying the accounting process, for example, information indicating the start time and end time of the accounting process is input to the comment content determination unit 13 described below. To do.

  For example, the input information determination unit 12 may analyze input information input via the POS terminal (analysis device 10) and determine whether or not predetermined information has been input. Moreover, the input information determination part 12 may specify the start time and end time of each accounting process based on the input information. The input information includes product information, service information, and the like.

  The product information is the name, price, etc. of the product to be accounted for (purchased). The merchandise information is input to the POS terminal (analysis device 10) via a scanner, operation keys, or the like provided in the POS terminal (analysis device 10). For example, by scanning a barcode or the like attached to a product using a scanner, the product information of the product is input to the POS terminal (analysis device 10). In addition, by pressing a predetermined operation key provided in the POS terminal (analysis device 10), product information of a product associated with the operation key is input to the POS terminal (analysis device 10).

  The service information is information relating to services provided at the store. The service corresponds to, for example, a point service in which points are accumulated as shopping is performed, a discount service in which discounts can be received for shopping at the store, and the like. Specifically, the service information of the point service may be information for identifying each customer, accumulated points, or the like. Specifically, the service information of the discount service may be identification information of a discount target product (service), a discount amount (discount rate), or the like.

  The service information is input to the POS terminal (analysis device 10) by scanning a service card possessed by the customer or scanning information (barcode or the like) printed on the discount coupon. In addition, the service information may be stored in a mobile terminal such as a mobile phone or a smartphone owned by the customer. In this case, for example, service information is input to the POS terminal (analysis device 10) by performing short-range wireless communication between the portable terminal and the POS terminal (analysis device 10).

  The input information may include a plurality of accounting processes for a plurality of customers, but may be described so that each accounting process for each customer can be distinguished. Furthermore, the input information may be associated with the start time and end time of accounting processing for each customer and date information. The start timing and the end timing of the accounting process for each customer are timings when a predetermined input is made to the POS terminal (analysis device 10). The input that is the start timing of the accounting process and the input that is the end timing of the accounting process may be any input. For example, the predetermined input serving as the start timing and end timing of the accounting process may be determined based on the design of the POS terminal (analysis device 10). For example, the first input of product information, service information, etc. is the start timing, and the predetermined input (eg: pressing a subtotal button, input of customer information (gender, age group, etc.) after the deposit amount is input is the end timing Also good. There is also a POS terminal (analysis device 10) configured to display a change amount, to open a drawer, or to change in a predetermined position by a predetermined input after the deposit amount is input. For example, the time when such a predetermined input is made is described in the input information as the start time and end time of the accounting process for each customer.

  Such input information is stored in each POS terminal (analysis device 10), or is connected to each POS terminal (analysis device 10) to a management device that is communicably connected to each POS terminal (analysis device 10). It is accumulated in association.

  The predetermined information for determining whether or not the information is input can be at least one of service information and product information of a predetermined product. The predetermined product is, for example, a sales promotion product. Promotional product list information as shown in FIG. 3 may be registered in the input information determination unit 12. And the input information determination part 12 may determine whether the merchandise information (predetermined information) of a predetermined | prescribed product is input based on sales promotion merchandise list information. The sales promotion product list information may be rewritable at an arbitrary timing.

  The input information determination unit 12 may batch process the input information at an arbitrary timing and perform the above determination, or may process the input information every time the accounting process for each customer is completed and perform the above determination. .

  When determining whether or not predetermined information is input every time the accounting process for the customer is completed, the input information determination unit 12 may perform the determination by a method different from the analysis of the input information. it can. For example, the input information determination unit 12 may monitor the input via the POS terminal (analysis device 10), and may determine whether or not the input content matches predetermined information each time an input is detected. When a content that matches the predetermined information is input, a predetermined flag may be set. The input information determination unit 12 may perform the above determination by confirming the state of the flag every time the accounting process for each customer is completed. Further, the input information determination unit 12 analyzes the content of the input via the POS terminal (analysis device 10), and detects that a predetermined input that is the start timing and the end timing of the accounting process described above is made. The start time and the end time of the accounting process for each customer (the time when a predetermined input for starting and ending the accounting process is made) may be specified.

  The statement content determination unit 13 analyzes the voice data at the time of accounting processing in which it is determined by the input information determination unit 12 that predetermined information has not been input, and determines whether or not a predetermined statement content has been said. To do.

  As described above, the input information determination unit 12 provides the statement content determination unit 13 with information identifying the accounting process that is determined to have not been input with the predetermined information, for example, information indicating the start time and end time of the accounting process. input. Based on such information, the statement content determination unit 13 identifies voice data at the time of the accounting process from the voice data acquired by the voice data acquisition unit 11 and sets it as a processing target. And the input information determination part 12 analyzes the audio | voice data of a process target using a speech recognition technique, and determines whether the predetermined speech content is said.

  For example, when the input information determination unit 12 determines “whether or not information related to a predetermined service has been input”, the statement content determination unit 13 may determine whether or not “an inquiry about a predetermined service” has been made. . The message for inquiring about a predetermined service includes an inquiry as to whether or not a point card is possessed, an inquiry as to whether or not a discount coupon is possessed, and the like.

  As another example, when the input information determination unit 12 determines “whether or not product information of a predetermined product (for example, sales promotion product) has been input”, the message content determination unit 13 It may be determined whether or not a “suggestion to recommend a product” has been made.

  The statement content determination unit 13 determines whether a keyword such as “point card”, “discount coupon”, “meat bun (predetermined product)” has been spoken, for example, It may be determined whether or not a statement recommending a predetermined product has been made. When the predetermined keyword is spoken, it can be determined that the predetermined content is spoken.

  When there is only one predetermined information determined by the input information determination unit 12, the statement content determination unit 13 stores the statement content (keyword), and determines whether the statement content is spoken. May be.

  On the other hand, when there are a plurality of predetermined information determined by the input information determination unit 12, the speech content determination unit 13 stores speech information (see FIG. 4) in which the predetermined speech content is associated with each predetermined information. You may keep it. Then, when the input information determination unit 12 determines that the predetermined information is not input, the speech content determination unit 13 searches the speech information (see FIG. 4) and associates it with the predetermined information that has not been input. The content of the remarks may be extracted, and it may be determined whether or not the extracted remark content has been remarked. In FIG. 4, a plurality of keywords are connected by “or” in a cell of one message content. In this case, if any one of a plurality of keywords connected by “or” is spoken, the input information determination unit 12 determines that the speech has been made.

  Note that a plurality of utterance contents may be associated with one predetermined information. FIG. 10 shows an example of such utterance information. According to the utterance information shown in FIG. 10, the input information determination unit 12 “an utterance that inquires about a predetermined service” and “an utterance that recommends membership in the predetermined service” when the information about the point service is not input It will be determined whether or not. In FIG. 10, words such as “enrollment” are described as keywords of “a remark that recommends membership in a predetermined service”. In this case, the input information determination unit 12 determines whether or not “an utterance inquiring about a predetermined service” and “an utterance recommending membership in the predetermined service” are made, and obtains a determination result for each utterance.

  By the way, instead of the process of determining whether or not a predetermined speech content is spoken by detecting a keyword, the speech content determination unit 13 parses the content the operator has uttered, and performs the above inquiry or a predetermined process. It may be determined whether or not a statement recommending a product has been made.

  According to the analysis apparatus 10 of the present embodiment described above, the determination results of the input information determination unit 12 and the statement content determination unit 13 can be summarized and a determination result as shown in FIG. 5 can be output.

  The determination results shown in FIG. 5 are the identification information (accounting process ID column) of the accounting process for each customer, the identification information of the operator who performed each accounting process (operator ID column), and predetermined information by the input information determination unit 12. The determination result (information column) of whether or not the information is input is associated with the determination result (remark column) of whether or not a predetermined statement is made by the statement content determination unit 13. .

  The operator who performed each accounting process, for example, login information to each POS terminal (analysis device 10) (information indicating a time zone in which each operator was logged in) and each POS terminal (analysis device 10). It can be specified based on the start time and end time of each accounting process performed.

  “Yes” in the information column shown in FIG. 5 indicates that the predetermined information has been input, and “No” indicates that the predetermined information has not been input. “Yes” in the remarks column indicates that a predetermined remark has occurred, “None” indicates that a predetermined remark has not occurred, and “−” indicates that determination by the remark content determination unit 13 has not been made. .

  The accounting process with the accounting process ID “000001” is performed by the operator with the operator ID “000A”, and it is understood that predetermined information has been input. Further, the accounting process with the accounting process ID “000002” is performed by the operator with the operator ID “000A”, and it can be seen that there was no input of predetermined information and that there was a predetermined remark.

  The determination result shown in FIG. 5 is obtained when the predetermined information and the predetermined utterance are one type. When at least one of the predetermined information and the predetermined utterance is two or more, for example, a determination result as shown in FIG. 6 is obtained.

  The determination result shown in FIG. 6 includes the identification information (accounting process ID column) of the accounting process for each customer, the identification information of the operator who performed each accounting process (column of operator ID), and a predetermined information by the input information determination unit 12. The determination result of whether or not the first information is input (information 1 column) and the determination result whether or not the predetermined second information is input by the input information determination unit 12 (information 2 column) ), A determination result whether or not a predetermined first utterance is made by the utterance content determination unit 13 (a column of the utterance 1), and whether or not a predetermined second utterance is made by the statement content determination unit 13 And the like (remark 2 column) and the like are associated with each other.

  FIG. 13 shows another example of the determination result. As shown in FIG. 13, the start time and end time (time column) of each accounting process may be associated with the identification information of the accounting process for each customer.

  Next, an example of the processing flow of the analysis method performed by the analysis apparatus 10 according to this embodiment will be described with reference to the flowchart of FIG.

  First, an example in which the analysis apparatus 10 processes input information and audio data by batch processing will be described. The voice data acquisition unit 11 acquires voice data to be processed, for example, voice data of a voice uttered by an operator who operates a POS terminal (analysis device 10) on a predetermined day. The audio data includes data corresponding to a plurality of accounting processes. Further, the input information determination unit 12 acquires input information to be processed, for example, input information via the POS terminal (analysis device 10) on the predetermined day. The input information includes information corresponding to a plurality of accounting processes (S10).

  The input information determination unit 12 specifies information for one accounting process from the acquired input information (S11). Then, the input information determination unit 12 determines whether or not predetermined information is input via the POS terminal (analysis device 10) based on the specified input information.

  If it has been input (Yes in S12), the input information determination unit 12 outputs the determination result (S15). For example, the input information determination unit 12 accesses a file indicating the determination result as illustrated in FIG. 5 and describes the determination result in association with the accounting process to be processed.

  On the other hand, if it has not been input (No in S12), the input information determination unit 12 inputs information (start time and end time of the accounting process) specifying the accounting process to the statement content determination unit 13. Then, the statement content determination unit 13 specifies the audio data at the time of the accounting process from the audio data acquired in S10 based on the information (accounting process start time and end time) specifying the accounting process ( S13).

  Thereafter, the speech content determination unit 13 performs speech recognition on the speech data identified in S13, and determines whether or not a predetermined speech has been made (S14). And the statement content determination part 13 outputs the determination result (S15). For example, the statement content determination unit 13 accesses a file indicating the determination result as illustrated in FIG. 5 and describes the determination result in association with the accounting process to be processed.

  Thereafter, the input information determination unit 12 determines whether or not there is an accounting process not specified in S11 in the input information acquired in S10. If there is (Yes in S16), the process returns to S11 and the process is continued. On the other hand, if not (No in S16), the process is terminated.

  The analysis device 10 that is a POS terminal can also process input information and voice data each time the accounting process for each customer is completed. In this case, in S10, the audio data acquisition unit 11 acquires audio data by real-time processing. The input information determination unit 12 acquires input information every time real-time processing or accounting processing for each customer is completed. If there is no other accounting process in S16, the input information determination unit 12 waits for input information of another accounting process.

  According to the present embodiment described above, it is not necessary to simply analyze voice data and determine whether or not a predetermined utterance has been made, but to determine whether or not predetermined information has been input via the POS terminal. When an accounting process that is not input is detected, the voice data at the time of the accounting process is analyzed to determine whether or not a predetermined statement has been made.

  Thus, according to the present embodiment, it is possible to determine whether or not predetermined information is input via the POS terminal. Then, the subsequent processing can be divided according to the determination result. If a predetermined condition is satisfied, it can be further determined whether or not a predetermined statement has been made. Based on the result of such a two-step determination in the present embodiment, for example, based on the fact that the predetermined remark was not made despite the fact that the predetermined remark was unnecessary, the operator's supervisor, for example Can avoid the inconvenience of negatively evaluating the operator.

  In addition, when performing voice recognition on all recorded and stored voice data as a processing target and determining whether or not a predetermined statement has been made, the amount of data to be processed becomes enormous and the processing load on the computer increases. . In the case of the present embodiment, not all of the recorded and accumulated audio data is processed, but only the audio data at the time of accounting processing in which predetermined information is not input can be processed. For this reason, the processing burden on the computer can be reduced.

<Second Embodiment>
The analysis apparatus 10 of this embodiment is an apparatus different from the POS terminal. The analysis apparatus 10 is installed in, for example, an office of each store, a data center that manages a plurality of stores, or the like. The analysis device 10 acquires input information, login information, voice data, and the like from a POS terminal, a recorder, a management device that communicates with a plurality of POS terminals, etc., by any input / communication means. Then, the same processing as in the first embodiment is performed. Also in this embodiment, the same effect as the first embodiment can be realized.

<Third Embodiment>
The analysis apparatus 10 of this embodiment performs customer service evaluation for each operator based on the determination results by the input information determination unit 12 and the statement content determination unit 13.

  FIG. 8 shows an example of a functional block diagram of the analysis apparatus 10 of the present embodiment. As illustrated, the audio data acquisition unit 11, the input information determination unit 12, the speech content determination unit 13, and the evaluation unit 14 are included. The configurations of the voice data acquisition unit 11, the input information determination unit 12, and the speech content determination unit 13 are the same as those in the first and second embodiments.

  The evaluation unit 14 performs customer service evaluation for each operator based on the determination results by the input information determination unit 12 and the statement content determination unit 13 (see FIGS. 5, 6, and 13). The evaluation unit 14 sets a positive evaluation as the predetermined information is input and as the predetermined remark is made. In addition, the evaluation part 14 is set as negative | minus evaluation, when predetermined information related to the said predetermined information is not made when the predetermined information is not input. However, when predetermined information is input, even if a predetermined remark related to the predetermined information is not made, the evaluation is not negative.

  For example, in a case where information related to a predetermined service is not input, a negative evaluation may be made if a message for inquiring about the predetermined service is not made. However, when information regarding a predetermined service is input, a negative evaluation may not be made even if a message for inquiring about the predetermined service is not made.

  Furthermore, when information regarding a predetermined service is not input, a positive evaluation may be performed if a message for inquiring about the predetermined service is made. In such a case, a positive evaluation may be made if a message recommending membership in a predetermined service is made.

  As another example, when product information of a predetermined product is not input, a negative evaluation may be made if a statement recommending the predetermined product is not made. However, when product information of a predetermined product is input, even if a statement recommending the predetermined product is not made, negative evaluation may not be performed. Furthermore, when the product information of the predetermined product is not input, a positive evaluation may be performed if a statement recommending the predetermined product is made.

  There are various specific evaluation methods, but an example will be described below. For example, the evaluation unit 14 holds an evaluation standard as shown in FIG. And the evaluation part 14 may calculate the point of each operator based on the determination result of the said evaluation reference | standard, the input information determination part 12, and the statement content determination part 13. FIG. For example, the evaluation unit 14 may calculate points (average value) per transaction process as points for each operator. In the case of the evaluation criteria shown in FIG. 9, the higher the point, the higher the evaluation.

  In addition to the evaluation based on the determination results of the input information determination unit 12 and the statement content determination unit 13, the evaluation unit 14 separately processes the voice data at the time of each accounting process, and further evaluates the operator based on the processing result. Also good. For example, the presence / absence of intonation may be determined, and a positive evaluation may be made when there is inflection, and a negative evaluation may be made when there is no inflection. Also, if the emotion is judged and the emotion is favorable for customer service (eg, joy, normality, etc.), the evaluation is positive. If the emotion is not favorable for customer service (eg, anger, sadness, etc.), the evaluation is negative. Good.

  According to the present embodiment described above, the same operational effects as those of the first and second embodiments can be realized. In addition, each operator and the supervisor who supervises the operator can easily check the customer service situation of each operator, that is, whether or not a predetermined manual is being executed, by confirming the evaluation result by the evaluation unit 14. Can grasp.

  In addition, the evaluation part 14 is set as negative | minus evaluation, when predetermined information related to the said predetermined information is not made when the predetermined information is not input. However, when predetermined information is input, even if a predetermined remark related to the predetermined information is not made, the evaluation is not negative. According to this embodiment, it is possible to avoid the inconvenience of negatively evaluating the operator with the fact that the predetermined statement is not made even though the predetermined statement is unnecessary. it can.

<Fourth Embodiment>
In the present embodiment, the evaluation criteria for the customer service evaluation of the operator are different for each time zone and / or for every crowded store. For this reason, according to the time slot | zone at the time of an accounting process and / or the congestion condition of a shop, an operator's customer service can be evaluated appropriately.

  An example of a functional block diagram of the analysis apparatus 10 of the present embodiment is shown in FIG. 8 as in the third embodiment. The configurations of the voice data acquisition unit 11, the input information determination unit 12, and the speech content determination unit 13 are the same as those in the first to third embodiments.

  The evaluation unit 14 is an operator based on evaluation standard information that defines an evaluation standard for evaluation based on the determination result by the input information determination unit 12 and the statement content determination unit 13 for each time period and / or for every crowded store. The third embodiment differs from the third embodiment in that customer service evaluation is performed every time. Other configurations are the same as those of the third embodiment.

  FIG. 11 and FIG. 12 show an example of evaluation criteria for each time zone. Comparing the evaluation criteria of FIG. 11 and FIG. 12, the points of “Speech 2-Yes” and “Speech 2-No” are different. For example, a customer who recommends joining a predetermined service in a time zone in which stores are crowded and there are many customers waiting for the order of cash register is not necessarily a correct customer service. As described above, there are some message contents that the operator should speak, which are more appropriate not to say under a predetermined situation.

  In this embodiment, in view of such a point, an appropriate evaluation standard is set for each time zone and / or the degree of congestion in a store, and an operator is appropriately evaluated.

  In FIG. 11 and FIG. 12, the business hours are divided into the first time zone and the second time zone, and the operator is evaluated based on two evaluation criteria, but it may be divided into more time zones. For example, an evaluation standard for each time zone may be prepared by dividing into a time zone in which stores are crowded and a time zone in which stores are not crowded.

  In addition, the evaluation part 14 can also judge the degree of congestion of the store at the time of each accounting process, and can also evaluate each accounting process with the evaluation criteria according to the judged degree of congestion. The degree of congestion at a store depends on, for example, the time between consecutive accounting processes (hereinafter sometimes referred to as “accounting interval”), that is, the time from the end time of the previous accounting process to the start time of the immediately following accounting process May be shown. And several evaluation criteria according to this time (accounting interval) may be defined. Naturally, the smaller the accounting interval, the more crowded the stores.

  For example, the evaluation unit 14 may calculate an accounting interval between the first accounting process and the second accounting process immediately after the first accounting process as the degree of congestion at the store during the first accounting process. Then, the operator at the time of the first accounting process may be evaluated based on the evaluation standard corresponding to the accounting interval. As another example, the evaluation unit 14 calculates the average of the accounting intervals from the first accounting process to the accounting process after a predetermined number (for example, 5 after) as the degree of crowded stores during the first accounting process. May be. And based on the evaluation criteria corresponding to the said average time, you may evaluate the operator at the time of a 1st accounting process.

  In addition, the crowded state of the store may be indicated by the number of constituent members in the column waiting for the accounting process. And a plurality of evaluation criteria according to this number may be defined.

  For example, you may comprise so that the row | line | column which waits for an accounting process may be image | photographed with an imaging device. And the evaluation part 14 may analyze the picked-up image which image | photographed the row | line | column at the time of 1st accounting process, and may calculate the number of the structural members of the said row | line | column as the congestion degree of the store at the time of 1st accounting process. . Then, the operator at the time of the first accounting process may be evaluated based on the evaluation standard corresponding to the calculated number.

  According to the present embodiment described above, the same function and effect as those of the first to third embodiments can be realized. Moreover, according to this embodiment, an operator can be evaluated appropriately according to a time slot | zone, the congestion state of a shop, etc.

<Fifth Embodiment>
The analysis device 10 of this embodiment is a POS terminal. The POS terminal (analysis device 10) performs determination by the input information determination unit 12 and the statement content determination unit 13 every time the accounting process for each customer is completed. When the POS terminal (analyzing device 10) determines that the predetermined information is not input by the input information determination unit 12 and the predetermined content is not determined by the statement content determination unit 13, A predetermined warning is output to the operator.

  According to this embodiment as described above, a predetermined determination can be made every time the accounting process for each customer is completed, and feedback can be made to the operator as necessary.

  In FIG. 14, an example of the functional block diagram of the analyzer 10 of this embodiment is shown. As illustrated, the analysis apparatus 10 includes a voice data acquisition unit 11, an input information determination unit 12, a statement content determination unit 13, and a warning unit 15. The analysis device 10 may further include an evaluation unit 14. The voice data acquisition unit 11, the input information determination unit 12, the speech content determination unit 13, and the evaluation unit 14 are the same as those in the first, third, and fourth embodiments.

  The input information determination unit 12 and the statement content determination unit 13 make the above determination every time the accounting process for each customer is completed.

  “Determine every time the accounting process for each customer is completed” means that the accounting process for each customer is completed (a predetermined input to end the accounting process is made at the POS terminal (analyzer 10)) This means that if the data (input information, audio data, etc.) is prepared, the determination is performed at an arbitrary timing thereafter. From the viewpoint of ensuring the immediateness of feedback to the operator, it is preferable that the determination is made as soon as possible after the data to be processed is prepared.

  The warning unit 15 issues a warning when it is determined that the predetermined information is not input by the input information determination unit 12 and the predetermined content of the speech is not spoken by the speech content determination unit 13. To emit.

  The warning unit 15 can issue a warning by using a display, a speaker, a warning lamp, or the like. For example, information such as “Let's check for the presence of a service card”, “Let's check for the presence of a discount coupon”, “Let us recommend a promotional product” may be output via a display or a speaker. A warning lamp may be lit.

  In addition, when the warning unit 15 issues the warning, the identification information of the POS terminal (analyzing device 10) or the POS terminal (analyzing device 10) at that time is addressed to a predetermined mail address accordingly. The identification information of the logged-in operator, warning contents, and the like may be transmitted. For example, by setting an e-mail address that can be always checked by the operator's supervisor, the supervisor can grasp in real time that the operator's customer service situation is undesirable.

  From the viewpoint of ensuring the immediacy of the warning, the warning unit 15 should issue a warning within 5 minutes, preferably within 1 minute, and more preferably within 30 seconds from the end of the accounting process for a certain customer. It is good to configure.

  Next, an example of the processing flow of the analysis method by the POS terminal (analysis device 10) of this embodiment will be described using the flowchart of FIG.

  The audio data acquisition unit 11 acquires and accumulates audio data by real-time processing. The input information determination unit 12 acquires and accumulates input information by real-time processing.

  And the input information determination part 12 will identify the said input information as a process target, if the input information for one accounting process is accumulate | stored (S20).

  Then, the input information determination unit 12 determines whether or not predetermined information is input via the POS terminal (analysis device 10) based on the specified input information to be processed.

  If it has been input (Yes in S21), if there is no input to end the processing (POS terminal (analyzer 10) power off, etc.) (Yes in S25), the process returns to S20. Then, the input information determination unit 12 waits for input information for the next accounting process to be accumulated. Note that, before returning to S20, as described in the first embodiment, the determination result in S21 may be output to a predetermined file.

  On the other hand, if it has not been input (No in S21), the input information determination unit 12 inputs information (start time and end time of the accounting process) specifying the accounting process to the statement content determination unit 13. Then, the statement content determination unit 13 specifies voice data at the time of the accounting process from the stored voice data based on the information (accounting process start time and end time) specifying the accounting process ( S22).

  Thereafter, the speech content determination unit 13 performs speech recognition on the speech data specified in S22, and determines whether or not a predetermined speech has been made (S23).

  If a predetermined statement has been made (Yes in S23), if there is no input to end the processing (Yes in S25), the process returns to S20. Then, the input information determination unit 12 waits for input information for the next accounting process to be accumulated. Before returning to S20, as described in the first embodiment, the determination result in S23 may be output to a predetermined file.

  On the other hand, when a predetermined statement is not made (No in S23), the warning unit 15 issues a predetermined warning (S24). Thereafter, if there is no input for ending the process (Yes in S25), the process returns to S20. Then, the input information determination unit 12 waits for input information for the next accounting process to be accumulated. Before returning to S20, as described in the first embodiment, the determination result in S23 may be output to a predetermined file. If the input which completes a process is received by S25 (No of S25), a process will be complete | finished.

  According to the present embodiment described above, the same operational effects as those of the first to fourth embodiments can be realized. Moreover, according to this embodiment, when an operator's evaluation (the determination result by the input information determination part 12 and the statement content determination part 13) is unpreferable, it can provide feedback to an operator immediately. For this reason, improvement of customer service by an operator after that is expected.

<Sixth Embodiment>
The input terminal of this embodiment is a terminal different from the POS terminal. For example, it may be an input terminal such as a so-called handy terminal used when processing an order from a customer in a restaurant or a pub, or may be another input terminal. The input terminal and the analysis device 10 of the present embodiment may be physically and / or logically separated or may be integrated.

  The voice data acquisition unit 11 of the present embodiment acquires voice data of voices uttered by an operator who operates the input terminal as described above.

  For example, the input terminal may be provided with a microphone, and the voice of an operator who operates the input terminal may be input through the microphone and stored in the input terminal, or configured to be communicable with the input terminal. May be stored in a storage device. In addition, a microphone independent of the input terminal may be connected to be communicable with the input terminal and / or another storage device by wire and / or wireless. The voice of the operator who operates the input terminal may be input via the microphone and stored in the input terminal and / or other storage device. For example, voice data may be accumulated while any operator logs in to the input terminal. And the audio | voice data acquisition part 11 may acquire the audio | voice data accumulate | stored in the input terminal and / or another memory | storage device.

  In addition, the operator may have a recorder including a microphone and a storage device. The operator's voice may be input to the recorder, and voice data may be accumulated. For example, the operator may log in to the input terminal, turn on the recorder while operating the input terminal, and accumulate voice data. In the case of this example, the audio data accumulated in the recorder are collectively input to the audio data acquisition unit 11 of the analysis device 10 at an arbitrary timing.

  Note that the audio data is associated with information that can specify which audio is being processed. For example, time information, date information, operator identification information, and the like are associated with each other.

  The input information determination unit 12 determines whether or not predetermined information is input through the input terminal in the process for each customer. The process for each customer is an arbitrary process using an input terminal, for example, a process for taking an order from a customer at a restaurant or a tavern.

  The input information determination unit 12 can analyze information input via the input terminal and determine whether or not predetermined information is input during each process. For example, the input information determination unit 12 specifies a predetermined input made via the input terminal as the start timing of the predetermined process, and specifies a predetermined input made via the input terminal as the end timing of the predetermined process. Then, it is determined whether or not predetermined information is input between the start timing and the end timing. The type of input used as the start timing and end timing can be arbitrarily determined according to the design of the input terminal.

  The speech content determination unit 13 analyzes the voice data during the processing in which it is determined that predetermined information has not been input, and determines whether or not a predetermined speech content has been spoken. For example, if the input information determination unit 12 determines that predetermined information is not input during a certain process, the utterance content determination unit 13 acquires time information of the start timing and end timing during the process from the input information determination unit 12. To do. In addition, the statement content determination unit 13 acquires identification information of the operator who has operated the input terminal. There are various means for obtaining the identification information of the operator who has operated the input terminal. An example will be described below. For example, as a premise for operating the input terminal, the user may be required to log in. The information input through the input terminal may be further associated with the identification information of the operator based on the login information. And the statement content determination part 13 may acquire the information which identifies the operator at the time of the process from which the predetermined information is not input from the input information determination part 12. FIG. Thereafter, the speech content determination unit 13 analyzes the speech data from the start timing to the end timing in the speech data corresponding to the operator, and determines whether or not a predetermined speech content has been said.

  Other configurations of the analysis apparatus 10 of the present embodiment are the same as those of the first to fifth embodiments. That is, the analysis apparatus 10 of the present embodiment also includes the evaluation unit 14 and the warning unit 15 in addition to the voice data acquisition unit 11, the input information determination unit 12, and the speech content determination unit 13, as in the first to fifth embodiments. At least one of the following. According to the analysis device 10 of the present embodiment, the same operational effects as those of the first to fifth embodiments can be realized.

Hereinafter, examples of the reference form will be added.
1. Voice data acquisition means for acquiring voice data of voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
A speech content determination means for analyzing the voice data at the time of the processing determined that the predetermined information has not been input, and determining whether or not a predetermined speech content is spoken;
Analyzing device.
2. In the analysis apparatus according to 1,
The input information determination means determines whether information relating to a predetermined service has been input,
If it is determined that the information regarding the predetermined service has not been input, the content determination unit determines whether or not an inquiry about the predetermined service has been made.
3. In the analysis apparatus according to 2,
The speech content determination means is an analysis device that determines whether or not a speech recommending membership in the predetermined service is made in addition to a speech inquiring about the predetermined service.
4). In the analysis device according to any one of 1 to 3,
The input information determination means determines whether or not product information of a predetermined predetermined product has been input,
The speech content determination unit is an analysis device that determines whether or not a speech recommending the predetermined product is made when it is determined that the product information is not input.
5. In the analysis device according to any one of 1 to 4,
The input information determination unit is an analysis device that analyzes input information input via the input terminal and determines whether or not the predetermined information is input.
6). In the analysis device according to any one of 1 to 5,
The speech content determination means stores speech information in which the predetermined speech content is associated with each of the predetermined information,
When the input information determination unit determines that the predetermined information has not been input, the statement content determination unit searches the statement information and displays the statement content associated with the predetermined information that has not been input. An analysis device that extracts and determines whether the extracted content of the remark is remarked.
7). In the analysis device according to any one of 1 to 6,
An analysis apparatus further comprising an evaluation unit that performs customer service evaluation for each operator based on determination results by the input information determination unit and the statement content determination unit.
8). In the analysis apparatus according to 7,
The said evaluation means is an analyzer which performs customer service evaluation for every operator based on the evaluation criteria defined for every time slot | zone and / or every congestion of a store.
9. In the analysis device according to any one of 1 to 8,
The analysis device, wherein the input terminal is a POS terminal.
10. In the analysis device according to any one of 1 to 9,
The input information determination unit and the statement content determination unit perform the determination every time the processing for each customer is completed,
Warning means for issuing a warning when it is determined by the input information determination means that the predetermined information has not been input, and the speech content determination means determines that the predetermined speech content has not been spoken An analysis device further comprising:
11. Computer
A voice data acquisition step of acquiring voice data of a voice issued by an operator operating the input terminal;
In the process for each customer, an input information determination step for determining whether or not predetermined information is input via the input terminal;
A speech content determination step of analyzing the audio data at the time of the processing determined to have the predetermined information not input and determining whether or not a predetermined speech content is spoken;
Analysis method to execute.
11-2. In the analysis method according to 11,
In the input information determination step, it is determined whether or not information related to a predetermined service is input,
An analysis method for determining whether or not a message for inquiring about the predetermined service has been made in the statement content determination step when it is determined that information regarding the predetermined service has not been input.
11-3. In the analysis method described in 11-2,
In the statement content determination step, in addition to a statement for inquiring about the predetermined service, an analysis method for determining whether or not a statement recommending membership in the predetermined service is made.
11-4. In the analysis method according to any one of 11 to 11-3,
In the input information determination step, it is determined whether or not product information of a predetermined predetermined product is input,
An analysis method for determining whether or not an utterance recommending the predetermined product is made in the utterance content determination step when it is determined that the product information is not input.
11-5. In the analysis method according to any one of 11 to 11-4,
In the input information determination step, an analysis method of analyzing input information input via the input terminal and determining whether or not the predetermined information is input.
11-6. In the analysis method according to any one of 11 to 11-5,
In the speech content determination step, speech information in which the predetermined content of speech is associated with each of the predetermined information can be used.
In the statement content determination step, when it is determined that the predetermined information is not input in the input information determination step, the statement content associated with the predetermined information that has not been input by searching for the statement information is obtained. An analysis method for extracting and determining whether or not the extracted content of the remark is remarked.
11-7. In the analysis method according to any one of 11 to 11-6,
An analysis method in which the computer further executes an evaluation step of performing customer service evaluation for each operator based on determination results of the input information determination step and the statement content determination step.
11-8. In the analysis method according to 11-7,
In the evaluation step, an analysis method for performing customer service evaluation for each operator based on an evaluation standard determined for each time zone and / or for every crowded store.
11-9. In the analysis method according to any one of 11 to 11-8,
The analysis method wherein the input terminal is a POS terminal.
11-10. In the analysis method according to any one of 11 to 11-9,
In the input information determination step and the statement content determination step, each time the processing for each customer is completed, the determination is performed,
When the computer determines that the predetermined information has not been input in the input information determination step, and determines that the predetermined speech content is not spoken in the speech content determination step, a warning An analysis method for further executing a warning process for issuing a warning.
12 Computer
Voice data acquisition means for acquiring voice data of a voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
An utterance content determination unit that analyzes the voice data at the time of the processing when it is determined that the predetermined information has not been input, and determines whether or not a predetermined utterance content is uttered,
Program to function as.
12-2. In the program described in 12,
Let the input information determination means determine the presence or absence of input of information about a predetermined service;
A program for causing the message content determination unit to determine whether or not a message for inquiring about the predetermined service has been made when it is determined that information regarding the predetermined service has not been input.
12-3. In the program described in 12-2,
A program for causing the statement content determination means to determine whether or not a statement recommending membership in the predetermined service is made in addition to a statement inquiring about the predetermined service.
12-4. In the program according to any one of 12 to 12-3,
Let the input information determination means determine whether or not product information of a predetermined product has been input;
A program for causing the message content determination means to determine whether or not a message recommending the predetermined product has been made when it is determined that the product information has not been input.
12-5. In the program according to any one of 12 to 12-4,
A program for causing the input information determination means to analyze input information input via the input terminal and determine whether or not the predetermined information has been input.
12-6. In the program according to any one of 12 to 12-5,
The speech content determination means stores speech information in which the predetermined content of speech is associated with each of the predetermined information,
When the input information determination unit determines that the predetermined information is not input to the statement content determination unit, the statement content associated with the predetermined information that has not been input by searching the statement information is displayed. A program for extracting and determining whether or not the extracted content of the remark is remarked.
12-7. In the program according to any one of 12 to 12-6,
A program that causes the computer to further function as an evaluation unit that performs customer service evaluation for each operator based on determination results by the input information determination unit and the statement content determination unit.
12-8. In the program described in 12-7,
The program which makes the said evaluation means perform customer service evaluation for every operator based on the evaluation criteria defined for every time slot | zone and / or every congestion of a store.
12-9. In the program according to any one of 12 to 12-8,
The program in which the input terminal is a POS terminal.
12-10. In the program according to any one of 12 to 12-9,
The input information determination means and the statement content determination means, each time the processing for each customer is completed, to make the determination,
If the computer determines that the predetermined information has not been input by the input information determination means, and the speech content determination means determines that the predetermined speech content has not been spoken, a warning A program that further functions as a warning means for issuing a warning.

  This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2014-201361 for which it applied on September 30, 2014, and takes in those the indications of all here.

Claims (12)

Voice data acquisition means for acquiring voice data of voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
A speech content determination means for analyzing the voice data at the time of the processing determined that the predetermined information has not been input, and determining whether or not a predetermined speech content is spoken;
Analyzing device. The analysis device according to claim 1,
The input information determination means determines whether information relating to a predetermined service has been input,
If it is determined that the information regarding the predetermined service has not been input, the content determination unit determines whether or not an inquiry about the predetermined service has been made. The analysis device according to claim 2,
The speech content determination means is an analysis device that determines whether or not a speech recommending membership in the predetermined service is made in addition to a speech inquiring about the predetermined service. In the analysis device according to any one of claims 1 to 3,
The input information determination means determines whether or not product information of a predetermined predetermined product has been input,
The speech content determination unit is an analysis device that determines whether or not a speech recommending the predetermined product is made when it is determined that the product information is not input. In the analysis device according to any one of claims 1 to 4,
The input information determination unit is an analysis device that analyzes input information input via the input terminal and determines whether or not the predetermined information is input. In the analysis device according to any one of claims 1 to 5,
The speech content determination means stores speech information in which the predetermined speech content is associated with each of the predetermined information,
When the input information determination unit determines that the predetermined information has not been input, the statement content determination unit searches the statement information and displays the statement content associated with the predetermined information that has not been input. An analysis device that extracts and determines whether the extracted content of the remark is remarked. In the analysis device according to any one of claims 1 to 6,
An analysis apparatus further comprising an evaluation unit that performs customer service evaluation for each operator based on determination results by the input information determination unit and the statement content determination unit. The analysis device according to claim 7,
The said evaluation means is an analyzer which performs customer service evaluation for every operator based on the evaluation criteria defined for every time slot | zone and / or every congestion of a store. In the analysis device according to any one of claims 1 to 8,
The analysis device, wherein the input terminal is a POS terminal. In the analysis device according to any one of claims 1 to 9,
The input information determination unit and the statement content determination unit perform the determination every time the processing for each customer is completed,
Warning means for issuing a warning when it is determined by the input information determination means that the predetermined information has not been input, and the speech content determination means determines that the predetermined speech content has not been spoken An analysis device further comprising: Computer
A voice data acquisition step of acquiring voice data of a voice issued by an operator operating the input terminal;
In the process for each customer, an input information determination step for determining whether or not predetermined information is input via the input terminal;
A speech content determination step of analyzing the audio data at the time of the processing determined to have the predetermined information not input and determining whether or not a predetermined speech content is spoken;
Analysis method to execute. Computer
Voice data acquisition means for acquiring voice data of a voice uttered by an operator operating the input terminal;
In the process for each customer, input information determination means for determining whether or not predetermined information is input via the input terminal;
An utterance content determination unit that analyzes the voice data at the time of the processing when it is determined that the predetermined information has not been input, and determines whether or not a predetermined utterance content is uttered,
Program to function as.

Images