JPH09251584A - Dangerous voiced word recognition monitor reporting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To protect a reporting person and to provide a highly reliable device by recognizing a dangerous state corresponding to the collation between the phrase of a speaker and a previously stored dangerous phrase through a dangerous state deciding part and reporting such a state through a dangerous state reporting part. SOLUTION: At a dangerous state deciding part 19, the phrase of the speaker converted by a voice analytic part 17 is compared with the dangerous phrase stored in a dangerous phrase table 18 and when these phrases are coincident, the dangerous state is decided. Thus, when a suspected person invades a shop or the like and the dangerous phrase is generated, the dangerous state can be decided, it is not necessary for the reporting staff such as a clerk to generate any sound, the reporting staff can be protected and a high-reliability dangerous voice word recognition monitor reporting device can be provided. Besides, when the dangerous state is decided, burglar prevention device is automatically operated, and the phrase corresponding to the voice signal of the suspected person can be preserved in a recognition data preserving part 22. Therefore, the crime can be prevented in advance and the search after the case is facilitated as well.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dangerous voice word recognition monitor reporting device used as crime prevention equipment.

[0002]

2. Description of the Related Art FIG. 8 is a block diagram showing a conventional voice recognition and notification system disclosed in, for example, Japanese Patent Laid-Open No. 6-131583, in which 1 is a voice input unit such as a microphone and 2 is its voice. A voice analysis unit that analyzes a voice signal input from the input unit 1 and extracts characteristic data. Reference numeral 3 is a storage unit for storing the characteristic data, 4 is a memory unit for storing the set characteristic data, and 5 is a predetermined value by comparing the characteristic data stored in the storage unit 3 with the characteristic data stored in the memory unit. This is a discriminating unit that outputs a signal when the matching rate is equal to or higher than.

Reference numeral 7 is a dial signal generating section for generating a dial signal based on the telephone number read from the telephone number storage section 6, and 9 is a voice synthesizing section for synthesizing voice based on the data from the message storage section 8. is there. Reference numeral 10 is an output unit for notifying the telephone line of the voice synthesized by the voice synthesizing unit 9 based on the dial signal of the dial signal generating unit 7, and 11 is a control unit for controlling the above respective units.

Next, the operation will be described. FIG. 9 is a flowchart showing the operation of the conventional voice recognition notification system. First, characteristic data such as "Is it Mr. Sasaki?" Is stored in the memory unit 4 in advance. This word is set as a keyword for security notification, and the characteristics of the voices of "Are you Mr. Sasaki?" Of multiple people who have been registered as notification staff at convenience stores in advance are extracted and stored in the memory unit 4. deep.

[0005] When an unbelievers intrudes during the operation of the store,
The notification staff inputs a voice from the voice input unit 1 "Is it Mr. Sasaki?" (Step ST1), analyzes the voice signal in the voice analysis unit 2 to extract characteristic data (step ST2), and the storage unit 3 In step ST3, the characteristic data is temporarily recorded (step ST3). Next, the characteristic data recorded in the storage unit 3 is compared with the characteristic data stored in advance in the memory unit 4 (step ST4), and if they match (step ST5), the notification device is activated (step ST6), The telephone number registered as the security notification destination is read from the telephone number storage unit 6 and the dial signal generation unit 7
Generates a dial signal (step ST7), and if there is an incoming call response (step ST8), the message storage unit 8
A voice obtained by synthesizing the above data by the voice synthesizing unit 9 is sent from the output unit 10 (step ST9). If they do not match in step ST5, it is confirmed whether all the registered data are compared (step ST10), and after confirming that they are not in the registered voice (step ST11), the notification is not required.

[0006]

Since the conventional voice recognition and notification system is constructed as described above, the characteristics of the voice of the notification staff when an unbelievers intrude and the previously stored voice of the notification staff. The characteristics of the above are compared, and when an unbelievers intrudes, it is necessary for the reporting staff to generate a voice. However, if a whistleblower makes a voice when an unbelievers intrude, it is expected that the danger of the whistleblower will increase further, and because of nervousness, the whistleblower cannot generate a predetermined voice. Therefore, there was a problem such as not being able to report when the reporting staff was silent.

The present invention has been made in order to solve the above problems, and an object thereof is to obtain a dangerous voice word recognition monitor reporting device which does not require the reporting staff to generate a voice.

[0008]

According to a first aspect of the present invention, there is provided a dangerous voice word recognition monitor / reporting device, which includes a phrase of a speaker converted by a voice analysis unit and a dangerous phrase stored in a dangerous phrase storage unit. And a dangerous situation reporting unit that reports a dangerous situation when it is determined to be a dangerous situation.

A dangerous voice word recognition monitor reporting device according to a second aspect of the present invention is provided with a crime prevention equipment operation section for activating a crime prevention system when a danger situation determination signal is input.

According to another aspect of the present invention, there is provided a dangerous voice word recognition monitor reporting device, wherein the notification content storage unit stores the telephone number stored in the telephone number storage unit when the dangerous situation determination signal is input. It is equipped with a telephone reporting unit that reports the content of the report.

According to another aspect of the present invention, there is provided a dangerous voice word recognition monitor reporting device which wirelessly reports a notification content stored in the notification content storage section to a predetermined notification destination when a dangerous situation determination signal is input. It is equipped with a reporting unit.

A dangerous voice word recognition monitor reporting device according to a fifth aspect of the present invention comprises a photographing unit for photographing the surroundings of a speaker when a dangerous situation determination signal is input.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below. Embodiment 1. 1 is a block diagram showing a dangerous voice word recognition monitor / report device according to Embodiment 1 of the present invention. In the figure, 12 is a voice word recognition monitor unit, and 13 is a voice word recognition monitor unit.
Is a dangerous situation reporting unit, and 14 is a signal line. Further, in the voice word recognition monitor unit 15, 15 is a voice input unit such as a microphone, 16 is a primary storage unit that temporarily stores the voice signal input from the voice input unit 15, and 17 is stored in the primary storage unit 16. A voice analysis unit that converts a voice signal into a phrase. Reference numeral 18 is a dangerous word / phrase table (dangerous word / phrase storage unit) in which dangerous words / phrases are stored in advance. Reference numeral 19 is a speaker's word / phrase converted by the voice analysis unit 17 and the dangerous word / phrase table 18.
It is a dangerous situation determination unit that determines a dangerous situation in accordance with a match with a dangerous phrase stored in.

Reference numeral 21 denotes a dangerous situation resetting unit for resetting the dangerous situation judging unit 19 according to the operation of the switch 20, and 22.
Is a recognition data storage unit that stores a word or phrase according to the voice signal of the speaker input from the voice input unit 15 when the dangerous situation determination unit 19 determines that the dangerous situation is present. It is a dangerous situation data output unit that outputs a dangerous situation determination signal when it is determined to be a dangerous situation. Further, in the dangerous situation reporting unit 13, 24 is a dangerous situation data input unit (dangerous situation input unit) for inputting a dangerous situation determination signal from the dangerous situation data output unit 23, and 25 is a dangerous situation determination unit at the dangerous situation data input unit 24. When a signal is input, a crime prevention equipment abnormality notification signal generation unit (crime prevention equipment operation unit) that generates an abnormality notification signal to the crime prevention equipment, 26
Is an output terminal section for outputting an abnormality notification signal to the crime prevention equipment.

Next, the operation will be described. FIG. 2 is a flow chart showing the operation of the dangerous voice word recognition monitor notification device according to the first embodiment of the present invention, and the operation will be described with reference to the drawing. First, a dangerous word / phrase table 18 is stored in advance with dangerous words / phrases, such as “put your hand up”, “hold up”, “pay money”, and “kill”. Next, in a state where the dangerous situation determination unit 19 is reset by the operation of the switch 20, the voice signal of the speaker input from the voice input unit 15 is temporarily stored in the primary storage unit 16,
The voice analysis unit 17 converts the voice signal into a phrase (step ST1). Next, in the dangerous situation judging unit 19, the phrase of the speaker converted by the voice analysis unit 17 is compared with the dangerous phrase stored in the dangerous phrase table 18,
If they match, it is judged as a dangerous situation (step ST2).

If the voice analysis unit 17 determines that the situation is dangerous, the recognition data storage unit 22 stores the word or phrase corresponding to the voice signal of the speaker input from the voice input unit 15 (step ST3). . Further, the dangerous situation data output unit 23 outputs a dangerous situation determination signal (step ST
4). Next, in the dangerous situation notification unit 13, the dangerous situation data input unit 24 inputs the dangerous situation determination signal from the dangerous situation data output unit 23 via the signal line 14, and the crime prevention equipment abnormality notification signal generation unit 25 receives the dangerous situation. When the judgment signal is input, an abnormality notification signal is generated (step ST5) and the crime prevention equipment is notified via the output terminal section 26.

As described above, according to the first embodiment, in the dangerous situation determination unit 19, the phrase of the speaker converted by the voice analysis unit 17 and the dangerous phrase stored in the dangerous phrase table 18 are stored. If they compare and match, it is judged as a dangerous situation, so an unbelievers intrude into a store or the like and generate a dangerous phrase so that the dangerous situation can be judged, and a notification staff such as a store clerk needs to generate a voice. Therefore, it is possible to protect the reporting personnel and obtain a highly reliable dangerous voice word recognition monitor reporting device. Further, if a dangerous situation is determined, the crime prevention equipment is automatically activated, and the recognition data storage unit 22 can store a word or phrase corresponding to the voice signal of the unbelievers, so that the crime can be stopped in an attempted manner. Not only can it be done, but it also has the effect of facilitating investigations after the incident.

Embodiment 2. 3 is a block configuration diagram showing a dangerous voice word recognition monitor reporting device according to a second embodiment of the present invention. In the figure, 27 is a telephone number registration table (telephone number storage section) in which a telephone number of a report destination is stored in advance. , 28 is a voice message recording section (report content storage section) in which the content of the notification to the report destination is stored in advance, and 29 is stored in the telephone number registration table 27 when the dangerous situation determination signal is inputted in the dangerous situation data input section 24. The automatic telephone notification control unit (telephone reporting unit) reports the content of the notification stored in the voice message recording unit 28 to the call destination of the telephone number. Reference numeral 30 is a telephone output jack, and 31 is a speaker 3 which outputs the voice of an unbelievers input from the voice input unit 15 when a danger situation determination signal is input to the danger situation data input unit 24.
2 is a voice echo back unit that echoes back to 2. The other configuration is the same as that of the first embodiment, and thus the description is omitted.

Next, the operation will be described. When the dangerous situation data input unit 24 inputs a dangerous situation determination signal, the automatic telephone notification control unit 29 dials the telephone number stored in the telephone number registration table 27, and the voice message recording unit 28 is notified to the report destination. The content of the notification stored in is reproduced and a notification is made via the telephone output jack 30. Also,
The voice echo back unit 31 echoes back the voice of the disbelief input from the voice input unit 15 to the speaker 32, and, for example, when the disbelief says, "Give me money,""I want to give me money in the store." Is output continuously in a loud voice,
Make unbelievers recognize that the security device is operating.

As described above, according to the second embodiment, when it is determined that there is a dangerous situation, the automatic telephone notification control unit 29 can automatically notify the destination of the notification, and the reliability can be improved. There is an effect that a highly dangerous dangerous word recognition monitor reporting device can be obtained. Also, if a dangerous situation is determined,
The voice echo back unit 31 has the effect of echoing back the voice of an unbelievers in the store, and of being able to stop crimes.

Embodiment 3 FIG. 4 is a block diagram showing a dangerous voice word recognition monitor reporting device according to a third embodiment of the present invention. In the figure, reference numeral 33 wirelessly reports when a dangerous situation determination signal is input to the dangerous situation data input section 24. This is a business wireless transmission control unit (wireless notification unit) that reports the content of the notification previously stored in the voice message recording unit 28 via the business wireless connection terminal 33a. The other configurations are the same as those in the first and second embodiments, and the description thereof will be omitted.

Next, the operation will be described. When the dangerous situation data input unit 24 inputs a dangerous situation determination signal, the commercial wireless transmission control unit 33 reproduces the content of the notification stored in the voice message recording unit 28 and notifies the business wireless connection terminal 33a. To report via.

As described above, according to the third embodiment, when it is determined that there is a dangerous situation, the business radio transmission control unit 33 can automatically inform the report destination of the report by radio. For example, there is an effect that a dangerous voice word recognition monitor notification device that can be installed even in a place where it cannot be reported to the outside unless it is wireless, such as a taxi.

Embodiment 4 FIG. 5 is a block configuration diagram showing a dangerous voice word recognition monitor reporting device according to a fourth embodiment of the present invention. In the figure, 15a to 15n are voice input units arranged for each floor such as a store building.
The other configurations are the same as those in the first to third embodiments, and the description thereof will be omitted. As described above, according to the fourth embodiment, by arranging the voice input units 15a to 15n for each floor, it is possible to obtain a dangerous voice word recognition monitor reporting device capable of collectively managing a wide range. .

Embodiment 5 FIG. 6 is a block configuration diagram showing a dangerous voice word recognition monitor reporting device according to a fifth embodiment of the present invention. In the figure, reference numeral 34 denotes a dangerous situation determination unit 19 which determines that a dangerous situation has occurred, and the recognition data storage unit 22 is shown. When a plurality of phrases corresponding to the voice signals of the speakers input from the voice input unit 15 are stored in, a dangerous voice occurrence story for recognizing the number of speakers from the sound quality of the phrases corresponding to the plurality of voice signals. It is a person counter section. Reference numeral 35 is a dangerous situation data output unit that outputs a dangerous situation determination signal including the number of speakers, and 36 is a dangerous situation data input unit (dangerous situation input unit) that inputs a dangerous situation determination signal including the number of speakers. .
Reference numeral 37 is a voice message recording unit that stores the content of the notification to the destination including the number of speakers, and 38 is the content of the notification according to the number of speakers when a dangerous situation determination signal including the number of speakers is input. Is a voice message editing control unit for extracting from the voice message recording unit 37 and outputting it to the commercial wireless transmission control unit 33. The other configurations are the same as those in the first to fourth embodiments, and the description thereof will be omitted.

Next, the operation will be described. The dangerous voice generating speaker counter unit 34 recognizes the dangerous situation each time the dangerous situation judging unit 19 judges that the dangerous situation occurs.
When there are a plurality of stored phrases, the number of speakers is recognized from the sound quality of the phrases corresponding to the plurality of voice signals. Dangerous situation data output unit 35 and dangerous situation data input unit 36
When a dangerous situation determination signal including the number of speakers is input to the voice message edit control unit 38 via the, the message content according to the number of speakers is extracted from the voice message recording unit 37, and the commercial radio is used. Output to the transmission control unit 33.

As described above, according to the fifth embodiment, when it is determined that there is a dangerous situation, the business radio transmission control unit 33 automatically informs the callee of the number of speakers by radio. The content can be reported, and there is an effect that it can be applied to, for example, an aircraft or the like where it is important to recognize whether a single offender or multiple offenders are important.

Embodiment 6 FIG. FIG. 7 is a block configuration diagram showing a dangerous voice word recognition monitor reporting device according to a sixth embodiment of the present invention. In the figure, 39 is a speaker when a dangerous situation determination signal is inputted to the dangerous situation data input section 24. It is a camera unit (imaging unit) that photographs the surroundings. The other configurations are the same as those in the first to fifth embodiments, and the description thereof will be omitted. As described above, according to the sixth embodiment, if a dangerous situation is determined, the crime prevention equipment is automatically activated, and the camera unit 39 can take a picture around the speaker. It has the effect of facilitating subsequent investigations.

[0029]

As described above, according to the first aspect of the present invention, the dangerous situation judging section judges the dangerous situation according to the collation of the phrase of the speaker and the dangerous phrase stored in advance, and the dangerous situation is judged. Since it is configured to report by the situation reporting unit, it is possible to automatically report even if the reporting staff does not generate voice, it is possible to protect the reporting staff and obtain a highly reliable device There is.

According to the second aspect of the present invention, the dangerous situation judging section judges the dangerous situation according to the collation of the speaker's word and the dangerous word stored in advance, and the crime prevention system operating section activates the crime prevention system. Since it is configured to operate, there is an effect that the existing crime prevention equipment can be automatically operated when a dangerous situation is determined.

According to the third aspect of the present invention, when the dangerous situation determination signal is input, the contents of the notification stored in the contents storage unit for the notification destination of the telephone number stored in the telephone number storage unit by the telephone notification unit. Since it is configured to report, the content of the report can be automatically reported to the report destination when a dangerous situation is determined, and a highly reliable device can be obtained.

According to the invention described in claim 4, when the dangerous situation determination signal is input, the wireless notification unit wirelessly notifies the predetermined notification destination of the notification content stored in the notification content storage unit. Therefore, if it is determined that there is a dangerous situation, the content of the report can be automatically reported to the report destination by wireless, and it can be installed even in places where it can not be reported to the outside unless it is wireless, such as taxis and aircrafts. is there.

According to the fifth aspect of the present invention, when the dangerous situation determination signal is input, the image capturing unit captures an image around the speaker. Therefore, when the dangerous situation is determined, the image is automatically captured. It is possible to take a picture of the area around the speaker, which has the effect of facilitating an investigation after the incident.

[Brief description of drawings]

FIG. 1 is a block configuration diagram showing a dangerous voice word recognition monitor notification device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing the operation of the dangerous voice word recognition monitor reporting device according to the first embodiment of the present invention.

FIG. 3 is a block diagram showing a dangerous voice word recognition monitor reporting device according to a second embodiment of the present invention.

FIG. 4 is a block diagram showing a dangerous voice word recognition monitor reporting device according to a third embodiment of the present invention.

FIG. 5 is a block diagram showing a dangerous voice word recognition monitor reporting device according to a fourth embodiment of the present invention.

FIG. 6 is a block configuration diagram showing a dangerous voice word recognition monitor notification device according to a fifth embodiment of the present invention.

FIG. 7 is a block configuration diagram showing a dangerous voice word recognition monitor reporting device according to a sixth embodiment of the present invention.

FIG. 8 is a block diagram showing a conventional voice recognition notification system.

FIG. 9 is a flowchart showing the operation of a conventional voice recognition notification system.

[Explanation of symbols]

13 dangerous situation reporting unit, 15 voice input unit, 17 voice analysis unit, 18 dangerous phrase table (dangerous phrase storage unit),
19 Dangerous situation determination part, 24, 36 Dangerous situation data input part (dangerous situation input part), 25 Crime prevention equipment abnormality notification signal generation part (crime prevention equipment operation part), 27 Phone number registration table (phone number storage part), 28 Voice Message recording section (report content storage section), 29 automatic telephone notification control section (telephone reporting section), 33 commercial wireless transmission control section (wireless reporting section), 39 camera section (imaging section).

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G10L 3/00 551 G10L 3/00 551A 551Z

Claims (5)

[Claims] 1. A voice analysis unit for converting a voice of a speaker input from a voice input unit into a phrase, a dangerous phrase storage unit for storing a dangerous phrase in advance, and a speaker for the speaker converted by the voice analysis unit. A dangerous situation determination unit that determines a dangerous situation in accordance with a match between a phrase and a dangerous phrase stored in the dangerous phrase storage unit, and a dangerous situation when the dangerous situation determination unit determines that the dangerous situation is a dangerous situation. A dangerous voice word recognition monitor reporting device having a dangerous situation reporting section for reporting the fact. 2. The dangerous situation reporting unit comprises a dangerous situation input unit for inputting a dangerous situation determination signal from the dangerous situation determination unit when the dangerous situation determination unit determines that the dangerous situation occurs, and the dangerous situation input unit. The dangerous voice word recognition monitor reporting device according to claim 1, further comprising: a crime prevention equipment operating unit that activates the crime prevention system when a danger situation determination signal is input. 3. The dangerous situation notification unit stores a dangerous situation input unit for inputting a dangerous situation determination signal from the dangerous situation determination unit when the dangerous situation determination unit determines that the dangerous situation is present, and a telephone number stored in advance. The telephone number storage unit, the notification content storage unit in which the content of the notification is stored in advance, and the notification destination of the telephone number stored in the telephone number storage unit when the dangerous situation determination signal is input in the dangerous situation input unit The dangerous voice word recognition monitor reporting apparatus according to claim 1, further comprising a telephone reporting section for reporting the content of the notification stored in the content storage section. 4. The dangerous situation reporting unit stores a dangerous situation input unit that inputs a dangerous situation determination signal from the dangerous situation determination unit when the dangerous situation determination unit determines that the dangerous situation is present, and the content of the notification is stored in advance. A notification content storage unit and a wireless notification unit for wirelessly reporting the notification content stored in the notification content storage unit to a predetermined report destination when a dangerous situation determination signal is input to the dangerous situation input unit The dangerous voice word recognition monitor notification device according to claim 1. 5. The dangerous situation notification unit includes a dangerous situation input unit for inputting a dangerous situation judgment signal from the dangerous situation judgment unit when the dangerous situation judgment unit judges that the dangerous situation is dangerous, and the dangerous situation input unit. The dangerous voice word recognition monitor reporting device according to claim 1, further comprising: a photographing unit that photographs the surroundings of the speaker when a dangerous situation determination signal is input.