JP2021179639A - Voice recognition device, management system, management program, and voice recognition method - Google Patents

Abstract

To provide a voice recognition device, a management system, a management program, and a management method that enable an administrator to manage utterances of an object person without confirming a voice of the object person in detail.SOLUTION: A voice recognition device 20 according to the present invention has: an utterance data generation part 211 which makes voice data received from a microphone device 10 capable of converting a voice of an object person into voice data, into a text to generate utterance data of the object person; and a totalizing part 212 which compares the utterance data with an inhibited word list 221 stored in a storage part 22 in advance to totalize a frequency of coincidence or similarity; and a display part 24 which displays the results of totalizing by the totalizing part 212 to the administrator.SELECTED DRAWING: Figure 1

Description

The present invention relates to a voice recognition device, a management system, a management program and a voice recognition method. More specifically, the present invention relates to a device, a system, a program and a method capable of preventing and grasping troubles caused by conversation by converting voice data into text to generate speech data and comparing it with predetermined prohibited word data.

Conversations can cause trouble at work sites, especially in direct contact with consumers. There are often troubles in which inappropriate remarks become harassment acts, and troubles in which the presence or absence of remarks regarding the contents of contracts and the details of the contents are disputed later. Among the troubles caused by conversation, there are serious cases such as intimidation by words and actions, suggestion of crime, and incitement.

Trouble caused by conversation is an important problem not only in the workplace but also in daily life. In such troubles, it is important to collect and analyze the conversation, which is useful for trouble prevention and post-processing. Traditionally, the cockpit of an aircraft has been equipped with a voice recorder that picks up and records crew conversations. In the unlikely event of an incident or accident, the voice recorded on this voice recorder is played back and analyzed, which is useful for investigating the cause of the incident and accident. In call center operators, a management method is often adopted in which the administrator reproduces and analyzes the voice data recorded by the microphone device picking up the response to the operator's customer to help troubleshoot.

On the other hand, a voice recorder that is portable by combining a small microphone device and a recording device has been devised and used (Patent Document 1). According to the invention described in Patent Document 1, since the conversation during work by the employee is picked up and recorded in the voice recorder, the administrator is recorded in the voice recorder for post-processing. You can play and analyze the conversation.

Japanese Unexamined Patent Publication No. 2007-325392

However, in the management method described in Patent Document 1, the recording time on the voice recorder becomes enormous. Then, this recording time increases as the number of subjects increases. In addition, this recording time increases as the time to be managed increases. Therefore, it takes a lot of labor and time for the administrator to search the recorded voice data for the purpose of indicating the occurrence of the trouble. In addition, the administrator collects the recorded voice data, searches for the voice data corresponding to the trouble that has occurred, and further reproduces the found voice data when dealing with the trouble using the recorded conversation. It is necessary to take the procedure of inspecting the contents of the voice data. Therefore, the administrator takes time from the occurrence of the trouble to the response. Therefore, the management of recording the conversation of the target person by such a voice recorder or the like is insufficient in terms of the administrator's early response to the trouble caused by the conversation, and in terms of the time and labor required for the analysis. There was a point.

In order for the administrator to take an early action against the trouble caused by the conversation of the target person, it is important to find out from the conversation the remarks that directly indicate the occurrence of the trouble such as "pay money" and "kill". Is. Furthermore, it is important for the administrator to understand the frequency of statements that are suspected of having trouble. For example, the statement "mother" does not indicate the occurrence of trouble by itself. However, if the administrator can grasp the frequency of saying "mother", the administrator is threatened to repeat phrases such as "I will tell my mother" and "Is it okay for my mother to know?" I can grasp it. Therefore, the administrator can take an early action.

In management using the conversation of the subject who has picked up and recorded the sound, the administrator grasps the frequency of remarks through the work of playing back most of the recorded voice data and confirming the content of the voice data. It is very difficult for the manager to perform such confirmation work for a large number of subjects in charge, both in terms of labor and time. Therefore, in the management method of collecting and analyzing the conversation of the target person, there is room for further improvement in the conventional method of recording the collected conversation and confirming the content by the administrator.

The present invention has been made based on such an idea, and the voice data received from the microphone device is converted into text, compared with a predetermined prohibited word data, and the matching or similar frequency is aggregated and displayed. It is an object of the present invention to provide a voice recognition device, a management system, a management program, and a voice recognition method capable of managing a subject's remarks without the administrator confirming the subject's voice one by one.

As a result of diligent studies to solve the above problems, the present inventors have converted the voice data received from the microphone device into text, compared it with predetermined prohibited word data, and aggregated and displayed the matching or similar frequency. By doing so, it was found that the above object can be achieved, and the present invention has been completed. Specifically, the present invention provides the following.

A feature according to the first invention is a voice recognition device that is connected to a microphone device capable of converting a subject's voice into voice data and can convert the voice data received from the microphone device into text. The voice data received from the voice data is converted into text, and the voice data generation unit that generates the voice data of the target person is compared with the voice data and a predetermined prohibited word data, and the matching or similar frequency is totaled. It is a voice recognition device having a unit and a display unit that displays the result of aggregation by the aggregation unit to the administrator.

According to the invention according to the first feature, the frequency of prohibited words included in the voice of the subject is automatically totaled and displayed on the display unit. For example, by setting a phrase indicating the occurrence of trouble such as "I'll kill you" in the prohibited word, the frequency at which the target person made a statement including such a phrase is totaled and displayed to the administrator through the display unit. .. The administrator grasps the frequency of remarks including prohibited words by the target person through the display of the display unit. The administrator grasps the occurrence of troubles around the target person by using the frequency of remarks including the displayed prohibited words, and manages warnings to the target person and reports to the police. The administrator can manage this series of remarks of the target person without checking the voice of the target person one by one.

Therefore, according to the invention according to the first feature, it is possible to provide a voice recognition device capable of managing the remarks of the target person without the administrator confirming the voice of the target person one by one.

The invention according to the second feature is an invention according to the invention of the first feature, and as a result of aggregation by the aggregation unit, the frequency at which the speech data and the prohibited word data match or resemble is equal to or higher than a predetermined frequency. A voice recognition device is provided which further has a warning unit for warning the administrator when the above is the case.

According to the invention according to the second feature, when the frequency at which the speech data and the prohibited word data match or are similar is equal to or higher than a predetermined frequency, the warning unit warns the administrator. Since the warning unit gives a warning, the administrator can respond to the trouble as soon as possible. Further, by issuing a warning by the warning unit when the frequency is higher than a predetermined frequency, it is possible to prevent mischievous warnings and erroneous warnings caused by counting words with similar pronunciations as matching or similar. Therefore, the speech management device can promptly deliver the required warning to the administrator without being confused with the false warning. The administrator can more appropriately and quickly grasp and manage the remarks indicating the occurrence of trouble.

The invention according to the third feature is the invention according to the invention of the second feature, in which the voice data is compared with the voice print data stored in advance, and the source of the voice data is the voice source of the voice print data. Further having a source determination unit for determining whether or not the data is, the warning unit provides a voice recognition device that warns the administrator when the source of the voice data is the voice source of the voice print data. do.

According to the invention according to the third feature, the source determination unit determines the origin of the voice data, and the warning unit informs the administrator only when the voice data matches the voiceprint data stored in advance. A warning action is taken. For example, the administrator warns the administrator that there was an unjust solicitation by the target person by registering the voiceprint of the target person and a prohibited word indicating an unjust solicitation such as "absolutely profitable". Will be done. As a result, prohibited words contained in the voices of speakers other than the target person, especially the voices of third parties who are present around the target person who has not participated in the conversation with the target person, are aggregated and warned the administrator. You can avoid false warnings.

The invention according to the fourth feature is an invention according to any one of the first to third features, and the speech data generated by the speech data generation unit is classified and stored for each source of the voice data. Provided is a voice recognition device further having a speech data storage unit for storing the speech data.

According to the invention according to the fourth feature, the generated speech data is classified, accumulated and stored for each source. As a result, the administrator can read the stored remark data and scrutinize its contents. In cases such as when cheating is suspected, there are often discrepancies between the parties regarding the presence or absence of remarks and the explanation of the content of the remarks. In a situation where the presence or absence of a statement and the explanation of the content of the statement differ between the parties, the administrator can objectively grasp the situation by examining the evidence of the stored statement data. In addition, in a labor trial in which the subject is a party, the manager can present the stored statement data as evidence.

Further, according to the invention according to the fourth feature, by storing the remark data, the administrator can analyze and use the stored remark data as big data. By such an analysis, the administrator can analyze the speech tendency of the target person, find words that are likely to appear before the trouble occurs, and add them to the prohibited words.

The invention according to the fifth feature is the invention according to the fourth feature, and the display unit extracts speech data of a specific sender from a plurality of speech data stored in the speech data storage section. To provide a voice recognition device that can be displayed.

According to the invention according to the fifth feature, the display unit can extract and display the speech data of a specific sender. As a result, the manager can analyze and use the remark tendency of the sender, which has favorable characteristics for the manager such as good sales performance. Alternatively, the manager can analyze and use the remark tendency of the sender having characteristics that are unfavorable to the manager, such as poor sales performance and many complaints.

The invention according to the sixth feature is an invention according to any one of the first to fifth features, wherein the aggregation unit is predetermined and is frequently used in the attribute to which the source of the voice data belongs. Provided is a voice recognition device that compares voice data with the voice data and performs the aggregation for the voice data different from the specific voice data.

According to the invention according to the sixth feature, the manager defines a loud sound that often occurs in the attribute to which the subject belongs as specific voice data. As a result, the loud sound is not processed by the tabulation unit. For example, when the target person is a waterworks worker and the sound of the tool is frequently heard in the vicinity, the administrator registers the sound of the tool as specific voice data. As a result, the frequently generated tool sounds are not processed by the tabulation unit. Therefore, the load on the tabulation unit is reduced.

The invention according to the seventh feature includes a microphone device capable of converting a subject's voice into voice data and the above-mentioned voice recognition device, and the microphone device measures the volume of sound included in the voice data. Provided is a management system including a volume measuring unit and a transmitting unit that transmits voice data having a volume equal to or higher than a predetermined amount to the voice recognition device.

According to the invention according to the seventh feature, the volume measuring unit measures the volume of the sound included in the voice data, so that the voice data whose volume is equal to or higher than a predetermined amount is transmitted to the voice recognition device. As a result, it is possible to prevent the speech data generation unit and the aggregation unit from processing the voice data that is low in volume and therefore does not include the speech that is converted into text, compared, and aggregated. Therefore, the amount of processing in the speech data generation unit and the aggregation unit is reduced, and the load on the management system is reduced.

The invention according to the eighth feature includes a microphone device capable of converting a subject's voice into voice data and the above-mentioned voice recognition device, and the microphone device provides a positioning unit for positioning the current position of the microphone device. The prohibited word data is predetermined according to the existing position of the microphone device, and the aggregation unit includes the speech data and the prohibited word data based on the current position received from the microphone device. Provides a management system that compares and aggregates matches or similar frequencies.

According to the invention according to the eighth feature, since the tabulation unit refers to the prohibited word data according to the existence position of the microphone device, the frequency of inappropriate remarks according to the facility including the existence position is determined. Can be aggregated. Inappropriate remarks according to this facility include, for example, remarking confidential information in a public place with many people such as a train station. In addition, the tabulation unit can also tabulate the frequency with which appropriate remarks are made according to the facility including its location. For example, the administrator registers a word related to the work of the target person as a prohibited word. Then, when the target person is away from the place where the business is performed, the management system aggregates and displays the frequency of making a statement including a word related to the business. As a result, the administrator can say whether the target person has left the place where the business is performed as part of the business, or simply abandoned the business and left the place where the business is performed, including words related to the business. It can be judged by using the amount of.

The inventions according to the ninth and tenth features are different categories of the inventions according to the first feature.

According to the present invention, the voice data received from the microphone device is converted into text, compared with a predetermined prohibited word data, and the matching or similar frequency is aggregated and displayed, so that the administrator can display the voice of the target person. It is possible to provide a voice recognition device that can manage the remarks of the target person without checking each time.

FIG. 1 is a block diagram schematically showing a hardware configuration and a software configuration of a management system 1 using the voice recognition device 20 in the present embodiment. FIG. 2 is a flowchart showing a flow of management using the management system 1 in the present embodiment. FIG. 3 is a diagram showing an example of the prohibited word list 221 in the present embodiment. FIG. 4 is a diagram showing an example of the tabulation list 222 in the present embodiment. FIG. 5 is a diagram showing an example of the remark storage list 223 in the present embodiment. FIG. 6 is a diagram showing an example when a management screen showing an aggregation result or the like is displayed on the display unit 24 using the management system 1 in the present embodiment. FIG. 7 is a diagram showing an example of the specific voice list 225 in the modified example 2. FIG. 8 is a conceptual diagram showing an example of a process in which the specific voice list 225 is referred to in the modified example 2 and the voice data of the loud sound that is often generated is removed from the voice data collected by the microphone device 10 and aggregated. ..

Hereinafter, an example of a suitable embodiment for carrying out the present invention will be described with reference to the drawings. It should be noted that this is only an example, and the technical scope of the present invention is not limited to this.

<Management system 1>
FIG. 1 is a block diagram schematically showing a hardware configuration and a software configuration of a management system 1 using the voice recognition device 20 in the present embodiment.

The management system 1 is connected to a microphone device 10 capable of converting the voice of the target person into voice data, and is connected to the microphone device 10, and is used for data management, data processing, screen display, and warning to the administrator in order to operate the management system 1. It is configured to include a voice recognition device 20 for performing the above and the like.

[Microphone device 10]
The microphone device 10 communicates with a control unit 11 that controls the operation of the microphone device 10, a storage unit 12 that stores a control program and the like executed by the microcomputer of the control unit 11, and a voice recognition device 20 and other devices. A communication unit 13 for performing the operation and a positioning unit 15 for positioning the current position of the microphone device 10 are provided.

Although not essential, it is preferable that the microphone device 10 has unique identification information and this identification information is transmitted to the voice recognition device 20. When the unique identification information is transmitted, the voice recognition device 20 receives and uses the identification information, so that the microphone device 10 and the source can be easily associated with each other. Thereby, the voice recognition device 20 can quickly identify the source. As this unique identification information, for example, the IPv6 address of the microphone device 10, the MAC address of the network card included in the microphone device 10, and the like can be used.

The control unit 11 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like.

The storage unit 12 is a device for storing data and files, and has a data storage unit such as a hard disk, a semiconductor memory, a recording medium, and a memory card. The storage unit 12 stores a control program or the like executed by the microcomputer of the control unit 11.

The communication unit 13 has a device for enabling communication with the microphone device 10 and other devices, for example, a network card compatible with the Ethernet (registered trademark) standard, a wireless device compatible with a mobile phone network, and the like.

The configuration of the positioning unit 15 is not particularly limited. Examples of the positioning unit 15 include a positioning system using a GPS receiver for positioning by receiving a signal from a GPS satellite, a system for positioning using information from a base station of a mobile phone, and the like.

[Voice recognition device 20]
The voice recognition device 20 is a control unit 21 that controls the operation of the voice recognition device 20, a control program executed by the microcomputer of the control unit 21, various data used for processing speech data, and speech data. It includes a storage unit 22 for storing data and the like generated by processing, a communication unit 23 for communicating with the microphone device 10, and a display unit 24 for displaying an aggregation result and the like. The data used for processing the speech data includes prohibited word data and the like. The data generated by processing the remark data includes the result of totaling the remarks that match or are similar to the prohibited words.

The type of the display unit 24 is not particularly limited. Examples of the display unit 24 include a monitor, a touch panel, and the like.

The control unit 21 reads a predetermined program and cooperates with the storage unit 22 and / or the communication unit 23 as necessary, so that the remark data generation unit 211 and the aggregation unit 212, which are elements of the software configuration in the management system 1, are used. , Warning unit 213, speech data storage unit 214, etc. are realized.

The hardware configurations of the control unit 21, the storage unit 22, and the communication unit 23 are the same as the hardware configurations of the control unit 11, the storage unit 12, and the communication unit 13, respectively. Further, the storage unit 22 stores a prohibited word list 221 that lists prohibited word data, a summary list 222 that aggregates the frequency of speech data that matches or is similar to the prohibited words, and the like. In addition, the storage unit 22 is configured to be able to set the voice data of the target person received from the microphone device 10.

<Management procedure using management system 1>
FIG. 2 is an example of a flowchart showing a procedure for managing the remarks of the target person by using the management system 1 using the voice recognition device 20. In the following, the preferred software configuration of the management system 1 will be described in more detail with reference to FIG.

[Step S10: Voice conversion]
First, the microphone device 10 picks up the voice of the subject. Then, the control unit 11 of the microphone device 10 cooperates with the storage unit 12 to convert the collected voice into voice data.

Well-known formats such as WAV format, MP3 format, and FLAC format are used as the format of the audio data. In order to reduce the consumption of the communication band, it is preferable to use a format using lossy compression such as the MP3 format.

[Step S11: Volume measurement]
Subsequently, the control unit 11 cooperates with the storage unit 12 to execute the volume measuring unit 111 and measures the volume of the voice data.

[Step S12: Determining whether the volume is equal to or higher than a predetermined amount]
If the volume measured in step S11 is equal to or higher than a predetermined amount, the process proceeds to step S13. If the volume is less than the specified amount, the process ends and the system returns to the standby state waiting for voice input.

According to the management system 1 according to the present embodiment, by going through the steps related to steps S11 and S12, the voice data can be transmitted to the voice recognition device 20 only when the volume is equal to or higher than a predetermined amount.

[Step S13: Positioning of the current position]
The control unit 11 determines the current position of the microphone device 10 in cooperation with the positioning unit 15, and stores it in the storage unit 12 as the current position data.

Although it is not an essential aspect, the current position determined by the positioning unit 15 at a predetermined timing is stored in the storage unit 12 as the current position data, and the current position data is read out to determine the current position in this step. It is also preferable to replace it with. As a result, it is possible to determine the current position each time the voice is converted and avoid waiting for the completion of the positioning process.

[Step S14: Voice data transmission]
The control unit 11 executes the transmission unit 112 in cooperation with the storage unit 12 and the communication unit 13, and transmits the voice data and the current position data to the voice recognition device 20.

According to the management system 1 according to the present embodiment, the voice data is transmitted to the voice recognition device 20 by going through the steps from step S10 to step S14. Further, according to the management system 1 according to the present embodiment, the current position can be transmitted to the voice recognition device 20 by going through the steps from step S13 to step S14.

[Step S20: Voice data reception]
When the voice data and the current position data are transmitted in step S14, the communication unit 23 of the voice recognition device 20 receives them. The received voice data and the current position data are stored in the storage unit 22.

[Step S21: Remark data generation]
When the voice data is received in step S20, the control unit 21 executes the speech data generation unit 211 in cooperation with the storage unit 22, converts the received voice data into text, and generates speech data. The generated speech data is stored in the storage unit 22. Well-known speech recognition technology such as speech recognition using a statistical model is used for converting speech data into text.

[Step S23: Comparison with prohibited words]
When the speech data is generated in step S21, the control unit 21 executes the aggregation unit 212 in cooperation with the storage unit 22 and compares the speech data with each of the prohibited words stored in the prohibited word list 221. do. If the spoken data and the prohibited word match, the process proceeds to step S24. If the spoken data and the prohibited word do not match, the process proceeds to step S26.

FIG. 3 is a diagram showing an example of the prohibited word list 221 in the present embodiment. As shown by "source" and "prohibited word" in FIG. 3, in the prohibited word list 221, a set consisting of the source and the prohibited word associated with the source is listed together with the ID. There is. ID: 1 in FIG. 3 for example indicates that the phrase "mother" is set as a prohibited word for the sender "Taro Tanaka" who is the target person.

By associating the sender with the prohibited word, the administrator can set the prohibited word for each sender. As a result, it is possible to finely manage inappropriate remarks according to the sender. For example, the phrase "hair is beautiful" is a statement that can lead to sexual harassment, so it is set as a prohibited word and aggregated, but if the sender is a hairdresser, it usually accompanies the sender's work. It is a remark of, and it is possible to manage such things as not stipulating prohibited words and not counting.

Although not required, the prohibited word list 221 preferably includes area information associated with the prohibited word, as shown by "area" in FIG. When the prohibited word list 221 includes an area, the aggregation unit 212 can compare this area with the current position data and refer to the prohibited word according to the current position. This makes it possible to aggregate prohibited words in more detail according to the current position.

The prohibited words according to the current position will be described using the IDs 1 to 3 and 101 to 104 of FIG. 3 as an example. In this example, seven sets of prohibited words and areas are defined for "Taro Tanaka" who is engaged in waterworks. In ID: 1 and 101, the phrase "mother" is defined as a prohibited word for the two areas of "station" and "residential area" for the sender "Taro Tanaka" who is the target person. The phrase "mother" is defined in both areas to indicate troubles such as intimidation, regardless of the area where the statement was made. In ID: 2, the phrase "water charges" is defined as a prohibited word for the "station" area. In Taro Tanaka, who is engaged in waterworks, the station is not a place for normal business. However, since "water charges" is a phrase related to business, if the remarks including "water charges" are repeated at the station, the manager can determine that Taro Tanaka visited the station to perform business. IDs: 102 to 104 define phrases that refer to appearance such as "beautiful woman", "baby face", and "beautiful hair" as prohibited words in the area of "residential area". Mentioning appearance in a residential area where Taro Tanaka operates indicates the possibility of sexual harassment by Taro Tanaka. The administrator can discover sexual harassment by Taro Tanaka from the frequency with which Taro Tanaka made statements including these prohibited words in the residential area.

Although not essential, the prohibited word list 221 preferably includes attribute information that bundles a plurality of sources, as shown by "attributes" in FIG. Since the prohibited word list 221 has attribute information that bundles a plurality of sources, the administrator can perform more efficient management such as changing the prohibited words of a plurality of sources at once by using the attributes. ..

[Step S24: Aggregation process]
Return to FIG. When step S23 detects a match between the prohibited word and the speech data, the control unit 21 cooperates with the storage unit 22 to increase the frequency corresponding to the prohibited word stored in the summary list 222.

FIG. 4 is a diagram showing an example of the tabulation list 222 in the present embodiment. As shown in FIG. 4, in the summary list 222, the source, the prohibited word, and the frequency are associated with each other and stored together with the ID. ID: 1 in FIG. 4 stores that the sender "Taro Tanaka" has not made a statement including the prohibited word "mother". ID: 2 in FIG. 4 stores that the sender "Taro Tanaka" made a statement including the prohibited word "water charges" 886 times.

By associating the sender, the prohibited word, and the frequency, the display unit 24 can display to the administrator the frequency of making a statement including the prohibited word for each sender. From this display, the administrator can make a detailed judgment of the occurrence of trouble and the precursor of trouble for each sender.

Although not required, it is also preferred that the aggregate list 222 stores an area containing current location data and / or current location data. By storing the current position data and the area in the summary list 222, the display unit 24 can display to the administrator the frequency of remarks including the prohibited words for each area including the saved current position. With this display, the administrator can manage the frequency with which the target person makes inappropriate remarks determined for each area.

[Step S26: Saving speech data]
Return to FIG. The control unit 21 executes the remark data storage unit 214 in cooperation with the storage unit 22, and stores the remark data in the remark storage list 223.

FIG. 5 is a diagram showing an example of the remark storage list 223 in the present embodiment. As shown in FIG. 5 by "source", "speech data", and "voice data", the source, the textualized speech data, and the voice data are associated with each other in the speech storage list 223. It is saved with the ID. In ID: 1 in Fig. 5, the sender "Taro Tanaka" made a statement in Chiyoda-ku, Tokyo, at exactly 12:00 on December 16, 2018, saying "This month's water bill is ...". That and the voice data [sound1] at that time are stored.

By associating the sender with the speech data, the administrator can manage the speech data for each sender. In addition, by associating the speech data with the voice data, the administrator can reproduce the voice data to obtain more detailed information about the speech, such as the tone and loudness of the voice of the target person.

Although it is not essential, it is also preferable that the position information indicating the current position of the target person at the time when the speech is made and the speech data are associated and stored in the speech storage list 223. By storing the location information, the administrator can grasp the place where the remark was made. This allows the administrator to conduct additional investigations such as listening around the position where the statement was made.

Although not essential, it is also preferable that the time information indicating the date and time when the remark was made is associated with the remark data and stored in the remark storage list 223. By storing the time information, the administrator can grasp the date and time when the remark was made. This allows the administrator to determine whether the trouble that occurred at a specific date and time and the remark data are related to each other. In addition, the administrator can search the remark data using this time information and efficiently search for the remark data related to the trouble that occurred at a specific date and time. In addition, the administrator can perform additional investigations, such as listening using the date and time when the statement was made.

[Step S27: Comparison with a predetermined frequency]
Return to FIG. The control unit 21 executes the warning unit 213 in cooperation with the storage unit 22 and compares the frequency of remarks including prohibited words with a predetermined frequency. If the frequency of speech is equal to or higher than the predetermined frequency, the process proceeds to step S28. If the frequency of speech is less than the predetermined frequency, the process proceeds to step S29.

[Step S28: Warning processing]
In step S27, if the frequency of remarks is equal to or higher than a predetermined frequency, the control unit 21 cooperates with the storage unit 22 to warn the administrator. It is preferable that this warning displays on the display unit 24 that the frequency of speech is equal to or higher than a predetermined frequency.

By being warned that the frequency of remarks has exceeded the predetermined frequency, the administrator will be aware of the occurrence of trouble or its precursors as soon as possible. As a result, the administrator can promptly give a warning to the target person or report to the police. In addition, by issuing a warning when the frequency is higher than a predetermined frequency, it is possible to prevent mischievous warnings and erroneous warnings caused by aggregating words with similar pronunciations as matching or similar.

Although not required, it is also preferred that this warning include the source. By including the source of the warning, the administrator can quickly determine which target person should be dealt with.

Although not required, this warning may also include remark data. By including the remark data in the warning, the administrator can determine what kind of trouble should be dealt with without searching for the corresponding remark data.

Although not an essential aspect, it is also preferable that the control unit 21 cooperates with the storage unit 22 and the communication unit 23 to send an e-mail and / or an instant message to the administrator to warn the administrator. By sending an email and / or an instant message to the administrator, the administrator can receive a warning even when the administrator is away from the voice recognition device 20. As a result, the administrator can quickly know that the frequency of prohibited words has exceeded the predetermined frequency and can deal with the trouble.

Further, although not essential, it is also preferable that the control unit 21 cooperates with the storage unit 22 and the communication unit 23 to give an additional warning using a voice such as a warning sound or a telephone call. By using a warning sound or a voice such as a telephone call, the administrator can receive the warning without constantly monitoring the display unit 24.

[Step S29: Display process]
The control unit 21 cooperates with the storage unit 22 to display the aggregation result on the display unit 24. The aggregated results displayed include the originator, the prohibited words, and the frequency of remarks including the prohibited words.

By displaying the sender, the prohibited word, and the frequency of the remark including the prohibited word on the display unit 24, the administrator grasps the frequency of the remark including the prohibited word by the sender. The administrator grasps the occurrence of troubles around the sender by using the frequency of remarks including the displayed prohibited words, and manages warnings to the target person and reports to the police. The administrator can manage this series of remarks of the sender without checking the voice of the sender one by one.

Although not required, a predetermined number of outlines of the latest speech data saved by the speech data storage unit 214, including the source, date and time, location, conversation data and / or access means to voice data. It is also possible to enumerate only the sets and display them on the display unit 24. By displaying the summary of the latest remark data, the administrator can grasp the target person who is currently remarking. Since the outline includes the means for accessing the conversation data and / or the voice data, the administrator can quickly confirm the speech data, the voice data, and the like of the target person who is currently speaking.

<Usage example of management system 1>
Subsequently, an example of using the management system 1 using the voice recognition device 20 in the present embodiment will be described.

First, the administrator connects the microphone device 10 and the voice recognition device 20. In this connection, for example, the administrator registers an IPv6 address indicating the voice recognition device 20 in the storage unit 12 of the microphone device 10, and further registers an IPv6 address indicating the microphone device 10 in the storage unit 22 of the voice recognition device 20. However, it is done by configuring each other so that they can communicate with each other via the Internet.

Subsequently, the administrator makes the subject carry the microphone device 10, installs the microphone device 10 on the subject's desk, and so on, so that the subject's voice can be converted into voice data. The volume of the converted voice data is measured, and when the volume is equal to or higher than a predetermined value, the microphone device 10 transmits the voice data to the voice recognition device 20.

The voice recognition device 20 receives voice data from the microphone device 10, converts the voice data into text, generates speech data, compares it with a predetermined prohibited word, and aggregates and displays the frequency of matching or similar speech. Perform a series of operations displayed on the unit 24. Further, the voice recognition device 20 can classify the generated speech data for each source and save it in the speech storage list 223.

FIG. 6 is a display example of the aggregated result displayed on the display unit 24. Under "Keyword detection number (Taro Tanaka)", for each prohibited word, a set of information consisting of the prohibited word and the frequency of remarks that match or are similar to the prohibited word is displayed in a predetermined number. .. “(Taro Tanaka)” in FIG. 6 indicates that the prohibited words and the frequency of remarks are displayed for “source: Taro Tanaka”. By designating the source, the administrator can display the prohibited words and the frequency of remarks in the same manner for any source. The administrator can display a set of more prohibited words and the frequency of remarks by operating "more".

As shown in FIG. 6, it is also possible to display a bar graph having a size corresponding to the frequency of speech. By displaying a bar graph sized according to the frequency of remarks, the administrator can more intuitively grasp the frequency of remarks that match or resemble the prohibited word.

As shown in FIG. 6 as the "latest usage status list" and the table below it, the source, date and time, place, speech data and / or voice data of the target person are accessed for the saved latest speech data. It is also possible to list and display a predetermined number of sets as an outline of one set including. By displaying the summary of the latest remark data, the administrator can grasp the target person who is currently remarking. By displaying the summary, the administrator can quickly confirm the speech data, voice data, etc. of the target person who is currently speaking. The administrator can access the saved speech data by manipulating the "details" in the "conversation content (text)" column. In addition, the administrator can access the saved voice data by operating the "details" in the "conversation content (voice)" column. The administrator can display more summaries by operating "More".

The administrator looks at the result of the aggregation displayed on the display unit 24, and determines whether or not a trouble has occurred or whether or not there is a sign of trouble from the frequency of remarks. Then, for the target person who is judged to have a trouble or a sign of trouble, management including trouble handling such as warning the target person and reporting to the police is performed. In addition, the administrator refers to the current position associated with the displayed voice data, heads to the site where the problem is occurring, or contacts the person in charge near the site where the problem is occurring. Manage including correspondence.

Further, the voice recognition device 20 can warn the administrator through the display unit 24 or the like when the frequency of remarks that match or resemble the prohibited words is equal to or higher than a predetermined frequency. The administrator who received the warning can perform management including troubleshooting as described above.

This series of operations of the management system 1 is automatically performed without the administrator checking the voice of the target person one by one. Therefore, the management system 1 in the present embodiment allows the administrator to manage the remarks of the target person without checking the voice of the target person one by one. As a result, the manager can manage the remarks of the target person with less effort. By being able to manage remarks with less effort, the administrator can simultaneously manage the remarks of a larger number of subjects without increasing the effort required to manage the remarks. In addition, by issuing a warning, the administrator can respond more quickly to troubles that occur around the target person.

<Can be provided as a management program>
So far, the present invention has been described as being provided as the management system 1, but the present invention is not limited thereto. The invention described in this embodiment can also be provided as a management program that can be executed by a computer. In the management program, the pronunciation data generation step in which the speech data generation unit 211 converts the voice data received from the microphone device 10 into text and generates the speech data of the target person, and the aggregation unit 212 are predetermined as speech data. The aggregation step of comparing with the prohibited word data and summarizing the matching or similar frequency and the display step of displaying the aggregation result by the aggregation unit 212 to the administrator are executed by the display unit 24.

<Modification example>
Hereinafter, various modifications of the invention described in the present embodiment will be illustrated.

[Modification 1] Origin determination The storage unit 22 includes a voiceprint data list 224 that stores voiceprint data in advance, and the control unit 21 compares the voiceprint data with the voiceprint data stored in the voiceprint data list 224. However, it is also preferable to include a source determination unit 215 for determining whether or not the source of the voice data is the source of the voiceprint data. At this time, it is preferable that the speech data is classified by the source and stored in the speech storage list 223. Further, the warning unit 213 may warn the administrator when the source of the voice data is the source of the voiceprint data.

The voiceprint data of the target person is stored in the voiceprint data list 224, and the source determination unit 215 determines the source, so that the voice recognition device 20 can aggregate only the statements of the target person and warn the administrator. That is, the prohibited words contained in the voices of third parties other than the target person, especially the voices of third parties who are present around the target person who did not participate in the conversation with the target person, are aggregated and sent to the administrator. You will not be warned. The administrator can obtain a more accurate aggregate result of the required frequency of prohibited words.

Further, by determining the originator, the speech data storage unit 214 can more appropriately associate and classify the speech data with the source, and store the speech data in the speech storage list 223. As a result, the administrator can specify the sender, read the stored speech data corresponding to the designated sender, and examine the content of the speech data by the sender. In a situation where there is a discrepancy in the presence or absence of a statement and the explanation of the content of the statement between the parties, the administrator can objectively grasp the situation by examining the evidence of the saved statement data from the sender. In addition, in a labor trial in which the subject is a party, the manager can present the stored statement data by the sender as evidence.

Further, since the speech data is more appropriately associated with the sender and stored, the administrator can more effectively analyze and use the stored speech data as big data. By such an analysis, the administrator can specify a source, analyze the speaking tendency of the source, find a word that is likely to appear before the trouble occurs, and add it to the prohibited word list 221. ..

[Modification 2] Reduction of load using specific voice data It is also preferable that the storage unit 22 includes a specific voice list 225 that stores specific voice data determined according to the attributes of the target person. By providing the specific voice list 225, the totaling unit 212 compares the specific voice data and the voice data according to the attributes of the target person stored in the specific voice list 225, and totals the voice data different from the specific voice data. Can be done.

FIG. 7 is an example of the specific voice list 225. In the specific voice list 225, the attribute of the sender and the specific voice data corresponding to the attribute are associated and stored. FIG. 7 shows that voice data [noise1] (a knocking sound of “ganging”) is defined as specific voice data for the attribute of water works. Although not essential, as shown as "remarks" in FIG. 7, the specific voice list 225 may include a description of the specific voice data associated with the specific voice data and the like as text. By including the explanation and the like as text in the specific voice list 225, the administrator can grasp and easily manage the specified specific voice data without reproducing the specific voice data one by one.

The aggregation using this specific voice data will be described by an example using the conceptual diagram of FIG. In this example, the subject is a worker engaged in water works, and a knocking sound of the water is frequently generated near the subject.

In the uppermost a) voice data, the example voice data transmitted from the microphone device 10 is a statement that can be the target of aggregation (indicated by "this water service", "wet", and "bad" in FIG. 8). In addition, it shows the inclusion of a tapping sound (indicated by "ganging" in FIG. 8).

The second stage b) specific voice data shows a knocking sound corresponding to the attribute of the waterworks work stored as the specific voice data in the specific voice list 225 in FIG. 7.

Return to FIG. In the third stage c) processing, the control unit 21 executes the aggregation unit 212 and compares the specific voice data shown in b) with the illustrated voice data. In c), the part where the background color of the voice data is changed is the part that matches the specific voice data as a result of this comparison.

As a result of this comparison, as shown in d) after processing at the bottom, the part corresponding to the specific voice data was removed from the illustrated voice data, and the subject indicated by "this water supply", "wet", and "bad". Only the content of the remarks made by is aggregated as remark data. In this way, by totaling only the portion different from the specific voice data, the processing load due to the execution of the totaling unit 212 is reduced.

Although the embodiments of the present invention have been described above, the present invention is not limited to these embodiments described above. Further, the effects described in the embodiments of the present invention are merely a list of the most suitable effects arising from the present invention, and the effects according to the present invention are limited to those described in the embodiments of the present invention. It's not a thing.

Further, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations. Further, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. ..

Moreover, each of the above-mentioned configurations, functions, and processing units may realize a part or all of them by hardware (for example, an integrated circuit). Further, each of the above configurations, functions, and processing units may be realized by software by the processor interpreting and executing a program that realizes each function. Information such as programs, tables, and files that realize each function can be placed in a memory, a hard disk, a recording device such as an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, or a DVD.

1 Management system 10 Microphone device 11 Control unit 111 Volume measurement unit 112 Transmission unit 12 Storage unit 13 Communication unit 15 Positioning unit 20 Voice recognition device 21 Control unit 211 Speech data generation unit 212 Aggregation unit 213 Warning unit 214 Speech data storage unit 215 Transmission Original discrimination unit 22 Storage unit 221 Prohibited word list 222 Aggregate list 223 Speech storage list 224 Voice print data list 225 Specific voice list 23 Communication unit 24 Display unit

Claims (10)

A voice recognition device that is connected to a microphone device capable of converting the voice of a target person into voice data and can convert the voice data received from the microphone device into text.
A speech data generation unit that converts the voice data received from the microphone device into text and generates speech data of the target person, and a speech data generation unit.
A tabulation unit that compares the remark data with predetermined prohibited word data and aggregates the frequency of matching or similarities.
A display unit that displays the result of aggregation by the aggregation unit to the administrator,
Has a voice recognition device. The first aspect of claim 1, further comprising a warning unit that warns the administrator when the frequency at which the speech data and the prohibited word data match or are similar as a result of aggregation by the aggregation unit is equal to or higher than a predetermined frequency. Voice recognition device. Further, it has a source determination unit that compares the voice data with the voiceprint data stored in advance and determines whether or not the source of the voice data is the voiceprint source of the voiceprint data.
The voice recognition device according to claim 2, wherein the warning unit warns the administrator when the source of the voice data is the source of the voiceprint data. The voice according to any one of claims 1 to 3, further comprising a voice data storage unit that classifies, stores, and stores the voice data generated by the voice data generation unit for each source of the voice data. Recognition device. The voice recognition device according to claim 4, wherein the display unit can extract and display speech data of a specific sender from a plurality of speech data stored in the speech data storage section. The aggregation unit compares the specific voice data, which is predetermined and frequently used in the attribute to which the source of the voice data belongs, with the voice data, and performs the totalization for the voice data different from the specific voice data. , The voice recognition device according to any one of claims 1 to 5. A microphone device capable of converting a subject's voice into voice data, and a voice recognition device according to any one of claims 1 to 6 connected to the microphone device.
The microphone device is
A volume measuring unit that measures the volume of sound included in the voice data,
A transmission unit that transmits voice data whose volume is equal to or higher than a predetermined amount to the voice recognition device, and
Has a management system. A microphone device capable of converting a subject's voice into voice data, and a voice recognition device according to any one of claims 1 to 6 connected to the microphone device.
The microphone device has a positioning unit for positioning the current position of the microphone device.
The prohibited word data is predetermined according to the current position of the microphone device.
The aggregation unit is a management system that compares the speech data with the prohibited word data based on the current position received from the microphone device, and aggregates the matching or similar frequency. It has a pronunciation data generation unit, an aggregation unit, and a display unit, and is connected to a microphone device capable of converting the voice of the target person into voice data, and the voice data received from the microphone device can be converted into text. It is a management program that is executed by the computer of the voice recognition device.
A pronunciation data generation step in which the speech data generation unit converts the voice data received from the microphone device into text and generates speech data of the target person.
An aggregation step in which the aggregation unit compares the remark data with predetermined prohibited word data and aggregates the matching or similar frequency.
A display step in which the display unit displays the result of aggregation by the aggregation unit to the administrator.
A management program that causes the computer to execute. A method in a voice recognition device that is connected to a microphone device capable of converting a subject's voice into voice data and can convert the voice data received from the microphone device into text.
A pronunciation data generation step in which the speech data generation unit converts the voice data received from the microphone device into text and generates speech data of the target person.
An aggregation step in which the aggregation unit compares the remark data with the predetermined prohibited word data and aggregates the matching or similar frequency.
A display step in which the display unit displays the result of aggregation by the aggregation unit to the administrator, and
The method.

Images