JP2011170274A - Accident restoration training device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable a trainee to easily know, in an evaluation carried out later on, whether the communication to relevant parties during accident restoration training is carried out based on the standpoint of other parties. <P>SOLUTION: A simulated control panel 101 including a display 102 and a simulated telephone 103 are prepared, simulated event information including accident information is displayed on the display 102 with time, and communication time that is a current time when using the simulated telephone and the destination of the communication are displayed on the display 102 together with the event information with time. Voices input in the simulated telephone 103 are associated with the communication time and records of the destination and recorded, and speech recognition of the voices is carried out to determine whether NG words exist. When a specific record is specified, the associated voices are reproduced and, if NG words are contained, a warning is issued. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an accident recovery training device used for accident recovery training in various plants and the like.

  In a plant such as a power plant, if an accident occurs, it must be promptly restored. In addition, timely contact with relevant parties is essential for such recovery. For this reason, the operator who corresponds to the recovery of the accident is required to have a corresponding skill. Since it is difficult to acquire such skills simply by learning on the desk, it is important to carry out practical training in advance. Therefore, in plants such as power plants, an accident recovery training device is prepared so that operators can experience from the accident to the recovery, and acquire the skills necessary to complete the quick recovery after the accident. I try to make it.

  As an example of an accident recovery training device, although it is not necessarily publicly used as a conventional technology, a simulated operation panel imitating an actual operation panel including a display is prepared, and a simulated event including accident information is prepared A device is used that displays information over time while practicing information. The accident recovery training apparatus can display an operation history made according to the event information displayed on the display, or leave it in a record. The operation history left in the record is utilized as a later reflection material or evaluation material by referring to the worker or the manager later.

  In addition to this, Patent Document 1 discloses a technique in which a voice spoken by an operator at the time of training is captured together with a moving image, the voice is voice-recognized, converted into text data, and stored. And this patent document 1 also describes displaying the text data after voice recognition synchronously with time.

JP 2004-102116 A

  The problem in training using the above-mentioned accident recovery training apparatus is the communication with the relevant places that should be performed in the event of accident recovery. That is, in the accident recovery training, there are a case where training is performed with an instructor and a case where voluntary training is performed. When an instructor is attached, the instructor can determine at that time whether or not the worker has contacted the relevant places in a timely manner, and this can be recorded. On the other hand, in the case of self-training, there is no record of contact with relevant places, so later verification is impossible.

Here, the items to be noted about the communication to the relevant places that should be performed at the time of accident recovery are as follows:
(1) Presence / absence of communication (2) Whether the communication was timely (3) Appropriateness of the content of communication (4) Whether the communication was made from the standpoint of the other party. Therefore, if all the items (1) to (4) can be confirmed after the accident recovery training, the accident recovery training will be more fruitful.

  In this regard, according to the technique described in Patent Document 1, text data based on the voice spoken by the operator during training is displayed in synchronization with the time, so that a record of contact with relevant places can be kept with the time. . Therefore, by adopting this technique, it is possible to confirm the items (1) to (4) after the accident recovery training.

  However, even if the text data based on the voice spoken by the operator at the time of training is displayed synchronously with the time, or even when the video with the voice is reproduced at that time, It is difficult to judge whether or not In particular, workers in plants such as power plants are highly technical and tend to use technical terms, so for people in related places who do not require such expertise, the content of communication may be difficult to understand, In some cases it may not be understood at all. For this reason, even if the technology as described in Patent Document 1 is adopted, it is obvious whether the communication to the various places made during the accident recovery is really the communication from the standpoint of the other party. It will not be.

  The present invention has been made in view of the above points, and it can be easily understood at the time of later reflection whether or not the communication to the related places performed at the time of accident recovery is the communication from the standpoint of the other party. Is to do so.

  According to the first aspect of the present invention, in the accident recovery training apparatus, a simulated operation panel including a display, a simulated telephone for performing a simulated telephone operation, and simulated event information including accident information are sequentially displayed on the display. When the display means for displaying and the simulated telephone make a simulated off-hook and connect to a predetermined destination, the current time at that time and the destination are displayed over time along with the event information. A recovery display means for displaying on the display; a recording means for recording the voice input to the simulated telephone connected to the simulated destination in association with the contact time and the destination record; and A designation means for designating the desired contact time and destination record by operation on the simulated operation panel, and correspondence to the designated contact time and destination record Reproducing means for reproducing the recorded voice, voice recognition means for recognizing the recorded voice, and determining whether or not a pre-registered NG word is included in the result of the voice recognition The above problem is solved by providing NG recognition means and warning means for displaying a warning on the display when the voice is reproduced when an NG word is included in the result of the voice recognition.

  According to a second aspect of the present invention, in the accident recovery training apparatus according to the first aspect, the recording means records voice as digital data.

  According to a third aspect of the present invention, in the accident recovery training apparatus according to the first or second aspect, the voice recognition unit performs voice recognition when the reproduction unit reproduces the voice.

  According to a fourth aspect of the present invention, in the accident recovery training apparatus according to the second aspect, the voice recognition unit performs voice recognition when recording the voice by the recording unit, and the NG recognition unit performs the voice recognition. At the same time, the presence / absence of an NG word is determined and the result is stored.

  According to a fifth aspect of the present invention, in the accident recovery training apparatus according to the fourth aspect, when an NG word is included in the result of the voice recognition, a warning is displayed on the display when the voice is recorded. Two warning means are provided.

  According to the first aspect of the present invention, if an NG word is included in the voice recognition result of the voice recorded in association with the designated contact time and destination record, the voice is recorded. Since the warning is displayed on the display during playback, it is easy to know at the time of later reflection whether or not the contact to the relevant places during the accident recovery training is from the standpoint of the other party Can do.

  According to the second aspect of the present invention, since voice is recorded as digital data, it is possible to improve ease of handling in subsequent voice recognition processing or the like.

  According to the third aspect of the present invention, since voice recognition is performed when the reproduction means reproduces the voice, it is possible to reduce the burden on the hardware during the accident recovery training, and to smoothly save the training contents. I can.

  According to the fourth aspect of the present invention, since voice recognition is performed when the voice is recorded by the recording means, the presence / absence of an NG word is determined together with the voice recognition, and the result is stored, so that data necessary for accident recovery training is stored. Thus, it is possible to reduce the burden on the hardware when reviewing the training content. For example, it is possible to open a way to review such training content with a low-performance client personal computer.

  According to the fifth aspect of the present invention, when an NG word is included in the result of the voice recognition performed when the voice is recorded by the recording means, a warning is displayed on the display when the voice is recorded. Workers can be made aware of the use of NG words from the stage, and more effective training can be performed.

The block diagram which shows the hardware constitutions of an apparatus as 1st embodiment. The schematic diagram which shows the file structure of a telephone number file. The schematic diagram which shows the file structure of a NG word file. The flowchart which shows the flow of the process at the time of the accident recovery training which a control part performs. The schematic diagram which illustrates the simulation event information displayed on a training screen. The flowchart which shows the flow of the process after the record specification which a control part performs. The schematic diagram of the review screen which shows a mode that the warning popup is displayed by pop-up at the time of NG word determination. The schematic diagram of the review screen which shows a mode that the warning history was displayed on the warning popup. The flowchart which shows the flow of the process at the time of the accident recovery training which a control part performs as 2nd embodiment. The schematic diagram which illustrates the event information displayed on a training screen. The flowchart which shows the flow of the process after the record specification which a control part performs.

  A first embodiment will be described with reference to FIGS.

  This embodiment is an example applied to the accident recovery training apparatus 11 having a simulated operation panel 101 (see FIG. 1) that is installed in a power plant as an example and imitates an operation panel installed in the central control room. Such an accident recovery training device 11 displays simulated event information including accident information over time on a display 102 (see FIG. 1) prepared in advance on the simulated operation panel 101. Then, the operation history performed by the worker according to the event information is displayed on the display 102 or recorded. The accident recovery training apparatus 11 can reproduce and display the operation history recorded on the display 102. The worker or the manager thereof can view such reproduction display later and use it as a reflection material or an evaluation material.

  The accident recovery training apparatus 11 according to the present embodiment also prepares a simulated telephone 103 (see FIG. 1) so that it is possible to practice contact with related places that are indispensable for accident recovery. Voices spoken by workers can be recorded. The contact time and destination when the simulated phone 103 is used are displayed on the display 102 and can be recorded. In addition to this, the excellent point of the accident recovery training apparatus 11 is that it recognizes the voice input to the simulated telephone 103, determines whether or not NG word is included, and can display the determination result on the display 102. It is the point which is doing. NG words include “unsuitable words and terms”, “words and terms that are difficult to understand or communicate”, and the like. By doing so, it becomes obvious at the time of reflection when playing back the recorded voice whether or not the NG word was given when contacting the related places using the simulated telephone 103, and to the related places that were used for the recovery from the accident. It becomes easy to grasp whether the contact is from the standpoint of the other party.

  Hereinafter, the accident recovery training apparatus 11 according to the present embodiment will be described in detail.

  FIG. 1 is a block diagram showing a hardware configuration of the accident recovery training apparatus 11. In the accident recovery training apparatus 11, the simulated operation panel 101 including the display 102 and the simulated telephone 103 are connected to the control unit 104 and controlled by the control unit 104. The control unit 104 is also connected with a voice capturing circuit 105, a voice recognition circuit 106, a voice synthesis output circuit 107, and a storage 108.

  The control unit 104 has a built-in processor (not shown) that executes a predetermined process. This processor may be configured by hardware including an integrated circuit, or may be configured by installing a program in a microcomputer. What is important is that the control unit 104 is configured to execute various processes necessary for executing the above-described processing in the accident recovery training apparatus 11.

  The voice capturing circuit 105 is a circuit that captures a voice signal from the simulated telephone 103. As an example, the simulated telephone 103 outputs a voice signal as an analog signal, and the voice capturing circuit 105 converts the analog signal into a digital signal. The audio capturing circuit 105 converts the audio signal converted into a digital signal into audio data in a format that is easy to handle such as a file format and outputs the audio data. The control unit 104 stores the audio data output from the audio capturing circuit 105 in the storage 108. The voice capturing circuit 105 may be provided separately from the simulated telephone 103 or may be incorporated in the simulated telephone 103.

  The simulated telephone 103 is connected to the control unit 104 through a route different from the route through which the voice capturing circuit 105 is interposed. This connection line is a line for transmitting operation information on the simulated telephone 103 to the control unit 104. The simulated telephone 103 generates an on-hook signal when it is on-hook, an off-hook signal when it is off-hook, and a dial signal when the telephone is dialed, and notifies the control unit 104 of these signals.

  The speech recognition circuit 106 is a circuit that recognizes the speech signal captured by the speech capture circuit 105. As an example, the voice recognition circuit 106 includes a voice recognition chip in which digital circuits for voice recognition are integrated. As such a voice recognition chip, various chips are currently put into practical use, and can be easily selected and adopted as appropriate. The voice recognition circuit 106 outputs a voice recognition result as a data string of digital data.

  The voice synthesis output circuit 107 is a circuit that converts voice data, which is digital data, into an analog signal that can be played back, amplifies it, and outputs the voice from the speaker 109. As an example, the speech synthesis output circuit 107 includes a speech synthesis chip in which digital circuits for speech synthesis are integrated. As with the speech recognition chip, such a speech synthesis chip is now in practical use as various chips and can be easily selected and adopted.

  For the storage 108, various devices such as HDD and SSD can be used.

  FIG. 2 is a schematic diagram showing the file structure of the telephone number file 151. The storage 108 stores a telephone number file 151. The telephone number file 151 stores information on the other party in association with the telephone number. The control unit 104 can recognize the telephone number made by the simulated telephone 103 based on the telephone number signal received from the simulated telephone 103, and by searching the telephone number file 151, the other party can be identified from the recognized telephone number. You can get the information.

  FIG. 3 is a schematic diagram showing the file structure of the NG word file 152. The storage 108 stores an NG word file 152. The NG word file 152 stores a message in association with the NG word. The NG word is stored in the form of a data string that can be collated with the data string output from the speech recognition circuit 106. The message is stored in a form that can be displayed on the display 102, for example, in the form of text data. The control unit 104 collates the data string that is the voice recognition result output from the voice recognition circuit 106 with the data string of the NG word stored in the NG word file 152, and determines whether the two data strings are similar or not. judge.

  Here, the NG words stored in the NG word file 152 are “inappropriate words and terms”, “words and terms that are difficult to understand or communicate”, and the like. Examples of the “inappropriate word or term” include a word or term that is likely to be mistaken for a certain word or term. Examples of “a word or term that is difficult to understand or difficult to communicate” include a word or term that is too technical, a site term, and the like. The reason why NG words are used as words or terms that are too technical is that workers in plants such as power plants are highly technical and tend to use technical terms. Because it can be difficult to understand or, in some cases, cannot be understood at all.

  FIG. 4 is a flowchart showing a flow of processing during accident recovery training executed by the control unit 104. This flowchart is a process of displaying simulated event information including accident information on the display 102 over time among the processes executed by the processor of the control unit 104, and the voice from the simulated telephone 103 executed at this time Only the import process is described. For example, the process of taking in data input by an operation on the simulated operation panel 101 is executed by a process different from the process shown in the flowchart of FIG. 4, but the description thereof is omitted here.

  FIG. 5 is a schematic view illustrating simulated event information SII displayed on the training screen TI. The processing contents of the flowchart shown in FIG. 4 will be described with reference to the training screen TI shown in FIG.

  First, the control unit 104 displays simulated event information SII, which is simulated event information, on the training screen TI (see FIG. 5) displayed on the display 102 (step S101). Here, the function of the display means is executed. As shown in FIG. 5, the simulated event information SII includes items of date, time, electric station name, facility name, device name, operation, and event. An important point of the simulated event information SII is that accident information is intentionally included because it is information for accident recovery training.

  While the training screen TI is displayed, the control unit 104 waits for determination of the presence or absence of an off-hook signal notified from the simulated telephone 103 (step S102). When the control unit 104 determines the notification of the off-hook signal (Y in step S102), the control unit 104 waits for the notification of the dial signal (step S103). If there is a notification of the dial signal (Y in step S103), the control unit 104 searches the telephone number file 151. Then, the other party's information is acquired from the telephone number specified from the received dial signal (step S104). Then, the control unit 104 acquires the current time by its own time counting function, sets the acquired current time as the contact time CT, sets the information of the other party acquired in step S104 as the destination CD, and sets the contact time CT and the destination. The CD is displayed on the display 102 (step S105). Here, the function of the recovery display means is executed.

  Here, as is clear from the training screen TI illustrated in FIG. 5, the simulated event information SII is displayed over time along the time flow, and the contact time CT when the worker makes the simulated phone 103 and The destination CD is also displayed over time. For example, in the training screen TI illustrated in FIG. 5, the simulated event information SII starts at 13:32:24 on June 12, and the operator It is shown that the mid-salary pay at 40:20, the power transmission and substation section at 13:41:00, and the simulated telephone 103 is placed at the sales office at 13:41:00. As shown in FIG. 5, these simulated events and operator actions are all arranged over time according to the flow of time. In FIG. 5, simulations of “OFF (manual)” of 13:41:20 and “ON (manual)” of 13:41:24 and 13:43:25 shown in the simulated event information SII column. The event is a simulated event displayed on the display 102 in accordance with an operation by the operator on the simulated operation panel 101.

  Next, the control unit 104 performs a process of capturing sound that is generated following the dial signal determined in step S103 (step S106). This process is performed by activating the voice capturing circuit 105 and is executed until an on-hook signal is received from the simulated telephone 103 (step S107).

  When the control unit 104 receives an off-hook signal from the simulated telephone 103 (Y in step S107), the control unit 104 returns to the process in step S101 until it is determined in step S108 that the training has ended. When the control unit 104 determines the end of the training (Y in step S108), the control unit 104 outputs the contact time CT and the data of the other party CD to the simulated event information SII that has been executed up to that time, and is output from the simulated telephone 103 at that time. The voice data captured by the input circuit 105 and the input data by the operator on the simulated operation panel 101 are associated with each other, and these data are stored in the storage 108 (step S109). By doing so, the control unit 104 records the voice input to the simulated telephone 103 connected to the simulated destination in association with the record of the contact time CT and the destination CD, Here, the function of the recording means is executed. The control unit 104 ends the process after the process of step S109.

  Thus, the accident recovery training is completed. Therefore, the trained operator reviews the training alone or together with the manager of the operator. Hereinafter, the processing content at the time of such reflection will be described.

  FIG. 6 is a flowchart showing the flow of processing after record designation executed by the control unit 104. When reviewing the training, the control unit 104 calls the data stored by the process of step S109 in the flowchart of FIG. 4 from the storage 108 and displays the data on the display 102. Thereby, an operator who reflects on the training can view the review screen RI (see FIGS. 7 and 8) that displays the same display as the training screen TI illustrated in FIG. However, this review screen RI is different from the training screen TI in that the contact time CT and the destination CD can be selected and designated. The selection designation in this case is performed by a pointing device (not shown) provided on the simulated operation panel 101 as an example. When such a contact time CT and destination CD are selected, the control unit 104 recognizes the designation of a record associated with these. Here, the function of the designation means is executed.

  When the control unit 104 determines the selection of the desired record recorded in the contact time CT and the destination CD (Y in Step S201), the control unit 104 opens the audio data file associated with the selected record (Step S201). S202). Then, the control unit 104 activates an audio playback player constructed by a process by a built-in processor, and starts playback of the audio data whose file has been opened (step S203). At this time, the control unit 104 activates the voice synthesis output circuit 107, generates synthesized voice based on the voice data, and drives the speaker 109 to realize the reproduction. Here, the function of the reproducing means is executed.

In this way, the driver who is reconsidering the training, the manager of the person, etc., displays the simulated event information SII, the corresponding contact time CT and the counterpart CD on the review screen RI displayed on the display 102. It can be confirmed by the notation over time along the flow of time. As a result, according to the accident information intentionally included in the simulated event information SII, it is obvious at what timing and where the driver has contacted, and effective reflection can be performed. Therefore, among the items to be noted regarding the communication to the relevant places to be carried out at the time of accident recovery,
(1) Presence / absence of communication (2) Appropriate reflection and evaluation can be made as to whether or not the communication was timely.

Moreover, according to the accident recovery training apparatus 11 of the present embodiment, it is possible to select and specify the desired record recorded in the contact time CT and the partner CD on the review screen RI displayed on the display 102. In response to this selection designation, the words of the workers who were spoken at the time of training can be reproduced as they are and released to the space by the speaker 109. As a result, among the items to be noted regarding the communication to the relevant places to be carried out at the time of accident recovery,
(3) Appropriate reflection and evaluation are possible for the appropriateness of the content of communication.

  What is important in the present embodiment is that the event that occurred during the training, that is, the handling operation using the simulated operation panel 101 for the accident information included in the simulated event information SII, and the related places using the simulated telephone 103 This means that the contact is not only displayed as text but also played as audio. In other words, what you should remember is the state during training. At the time of training, the worker visually recognizes the simulated event information SII displayed on the display 102, and when the occurrence of a simulated accident is recognized, the simulated operation panel 101 performs necessary operations. In the present embodiment, such an event reproduction is realized by display on the display 102. On the other hand, at the time of training, when an operator recognizes the occurrence of a simulated accident, he / she communicates with related places by voice using the simulated telephone 103. In the present embodiment, such event reproduction is realized by sound reproduction. For this reason, when performing a reflection or evaluation of the training, the worker or the manager of the worker or the like can relive the training in a state very close to the training actually performed. Therefore, in this Embodiment, the opportunity of the appropriate reflection and evaluation about said (1)-(3) can be provided very effectively.

Here, as items to keep in mind about the communication to the various places that should be carried out at the time of accident recovery,
(4) I mentioned above whether or not we are communicating from the standpoint of the other party. The accident recovery training apparatus 11 of the present embodiment makes it possible to check this item very easily. This is explained by the processing contents of steps S204 to S208 in the flowchart shown in FIG. The content of these processes is to determine whether or not the NG word is included in the voice data stored in the storage 108 at the time of training, and if the NG word is included, this is displayed on the display 102 as a warning. Is. Hereinafter, it demonstrates along the flowchart of FIG.
The control unit 104, after selecting and specifying the desired record determined in step S201, separates the voice recognition circuit from the voice reproduction process using the voice synthesis output circuit 107 of the corresponding voice data stored in the storage 108. Voice recognition of the voice data using 106 is executed (step S204). Here, the function of the voice recognition means is executed.

  Then, the control unit 104 determines whether the NG word defined in the NG word file 152 (see FIG. 3) stored in the storage 108 is not included in the voice recognized in Step S204 (Step S204). S205). Here, the function of the NG recognition means is executed.

  As a result of the determination in step S205, if an NG word is not included (N in step S205), the control unit 104 determines from the voice recognition in step S204 to step S208 until it determines the end of reproduction (Y in step S208). The process up to the end of playback is continued. On the other hand, when it is determined that an NG word is included (Y in step S205), the control unit 104 searches the NG word file 152 for a message defined in association with the NG word ( Step S206). Then, a warning pop-up WPU including the searched message is pop-up displayed on the review screen RI displayed on the display 102 (step S207), and a reproduction end determination process is executed (step S208). Here, the function of the warning means is executed.

  FIG. 7 is a schematic diagram of the review screen RI showing a state where a warning pop-up WPU is pop-up displayed at the time of NG word determination. In the warning pop-up WPU, a warning message “XXX is a difficult word to communicate” is displayed. Here, the NG words stored in the NG word file 152 are “inappropriate words and terms”, “words and terms that are difficult to understand or difficult to communicate”, and the like, as described above. Therefore, when the training voice data stored in the storage 108 includes such an NG word, in other words, when the operator utters such an NG word during training, the warning popup WPU pops up. Will be displayed. As a result, when an operator utters a word or term that is easy to make a mistake about a certain word or term, a word or term that is too professional, a site term, etc. Can be seen at a glance. As a result, according to the present embodiment, it is possible to easily determine whether or not the communication to the related places performed during the accident recovery training is a communication from the standpoint of the other party during the subsequent reflection.

  FIG. 8 is a schematic diagram of a training screen showing a warning history displayed on the warning popup WPU. As shown in FIG. 7, the warning pop-up WPU includes an OK button A and a history button B. When the OK button A is selected and specified by the simulated operation panel 101, the warning pop-up WPU is not displayed on the review screen RI by interpreting that the worker or the administrator of the person understands the warning content. At this time, the control unit 104 stores the warning display history in the storage 108 in the background. Therefore, when the history button B is selected and specified by the simulated operation panel 101, the control unit 104 uses the warning display history saved in the storage 108 as the warning history WH as shown in FIG. To display. This warning history WH includes the time when the worker uttered the NG word and the counterpart at the time of training, and a message about such an NG word. Such a warning history WH effectively provides the worker or the manager of the worker with the reflection material or evaluation material of the training previously performed. Then, the warning pop-up WPU including the warning history WH illustrated in FIG. 8 returns to the initial warning pop-up WPU illustrated in FIG. 7 by selecting and specifying the return button C included in the screen.

  A second embodiment will be described with reference to FIGS.

  FIG. 9 is a flowchart showing the flow of processing during accident recovery training executed by the control unit 104 as the second embodiment. The process at the time of accident recovery training of this embodiment is different from the process of the first embodiment shown in FIG. 4 only in the process of steps S106-1 to S106-4. That is, in the first embodiment, at the time of accident recovery training, the voice from the simulated telephone 103 is captured and stored in the storage 108 in the form of voice data (steps S106 and S109 in FIG. 4). At this time, in the first embodiment, voice recognition using the voice recognition circuit 106 is not executed. On the other hand, in the present embodiment, after the voice is captured (step S106-1), the control unit 104 recognizes voice data that has been captured and converted into data using the voice recognition circuit 106 (step S106-2). . Then, it is determined whether or not the recognized voice includes the NG word defined by the NG word file 152 (see FIG. 3) (step S106-3). Thereafter, if the NG word is not included, the control unit 104 proceeds to determine whether there is an on-hook (N in Step S106-3, Step S107). If the NG word is included (Y in Step S106-3). ), The presence of an NG word is displayed on the training screen TI displayed on the display 102 (step S106-4), and then the process proceeds to the presence / absence determination of on-hook (step S107). At this time, the control unit 104 stores the determination result of the presence / absence of the NG word in the storage 108 in the background.

  FIG. 10 is a schematic diagram illustrating event information displayed on the training screen TI. The training screen TI of the present embodiment basically has display contents equivalent to the training screen TI of the first embodiment. The difference is the display with NG word displayed in step S106-4 in the flowchart shown in FIG. For this purpose, the training screen TI of this embodiment is provided with a display field called NG word WI, which displays the number of utterances of the NG word. The control unit 104 counts the number of NG word utterances in the background of the NG word determination process in step S106-3 in the flowchart shown in FIG. 9, and sets the count in the column of the NG word WI on the training screen TI. Update display. Here, the function of the second warning means is executed.

  FIG. 11 is a flowchart showing the flow of processing after record designation executed by the control unit 104. The process after record designation in this embodiment is basically the same as the process in the first embodiment. The difference is that the voice recognition process (step S204) in the first embodiment shown in FIG. 6 is not executed. In the present embodiment, since speech recognition processing has already been executed at the time of training (step S106-2 in FIG. 9) and the result has been saved in the storage 108 (step S106-3 in FIG. 9), it is determined whether or not there is an NG word. This is because the result of the presence / absence of NG words stored in the storage 108 can be used as it is for the determination (step S205 in FIG. 11).

  As described above, in the present embodiment, at the time of accident recovery training, a warning can be given in the form of the number of utterances whether or not NG words are included in the utterances to the other party on the simulated telephone 103. Thereby, the worker can be aware of the use of the NG word from the training stage, and can perform more effective training.

102 Display 101 Simulated operation panel 103 Simulated telephone Step S101 Display means Step S104, Step S105 Recovery display means Step S106, Step S109 Recording means Step S201 Designating means Step S202, Step S203 Reproducing means Step S204, Step S106-2 Voice recognition means Step S205, Step S106-3 NG recognition means Step S206, Step S207 Warning means Step S106-4 Second warning means

Claims (5)

A simulated operation panel including a display;
A mock phone that performs a mock phone call,
Display means for displaying simulated event information including accident information on the display over time;
When the simulated telephone makes a simulated off-hook connection to a predetermined destination, the contact time, which is the current time at that time, and the destination are displayed together with the event information on the display over time. Display means;
Recording means for recording the voice input to the mock telephone connected to the mock destination in association with the contact time and the destination record;
A designation means for designating a desired contact time and a record of the other party by an operation on the simulated operation panel;
Playback means for playing back the voice recorded in association with the designated contact time and the record of the other party;
Voice recognition means for voice recognition of the recorded voice;
NG recognizing means for determining whether or not a pre-registered NG word is included in the speech recognition result;
A warning means for displaying a warning on the display when the voice is reproduced when an NG word is included in the result of the voice recognition;
An accident recovery training device comprising:   The accident recovery training apparatus according to claim 1, wherein the recording unit records voice as digital data.   The accident recovery training apparatus according to claim 1, wherein the voice recognition unit performs voice recognition when the reproduction unit reproduces the voice. The voice recognition means performs voice recognition when recording voice by the recording means, and the NG recognition means determines the presence or absence of an NG word together with the voice recognition and stores the result.
The accident recovery training device according to claim 2.   The accident recovery training apparatus according to claim 4, further comprising second warning means for displaying a warning on the display when recording the voice when an NG word is included in the voice recognition result.