A campus anti-bullying alarm method and system based on voice recognition

Abstract

The application relates to a campus anti-bullying alarm method and system based on voice recognition, which comprises the following steps: collecting alarm voice information of students; recognizing help-seeking information in the alarm voice information; judging whether to trigger bullying alarm according to the help-seeking information, if yes, sending alarm information through a network, if no, not sending alarm information; determining the position of the student who alarms and generating alarm position information according to the alarm information; and sending the alarm position information. The application collects alarm information of students, recognizes help-seeking information in the alarm information, triggers alarm through the help-seeking information, so that students can call for help when bullied in the campus, the school can know the bullying event in the campus in the first time, and the bullying in the campus can be stopped or intervened to prevent the bullying from continuing.

Description

Technical Field

[0001] This invention relates to the field of campus security technology, specifically to a campus anti-bullying alarm method and system based on voice recognition. Background Technology

[0002] School bullying seriously affects students' healthy growth and learning. Currently, bullying is only discovered after the fact through student feedback. Without student feedback, schools are unlikely to know, potentially causing indelible psychological trauma to students. Existing school security systems use video surveillance to record student behavior in public places; however, to protect student privacy, cameras are not installed in all areas. Furthermore, even when cameras are installed, bullying cannot be detected immediately; the incident is only understood afterward by reviewing surveillance footage. Therefore, existing school security measures are insufficient to prevent bullying in a timely manner. Summary of the Invention

[0003] To address the technical problems in existing campus security systems, such as their inability to promptly prevent bullying, this invention provides a campus anti-bullying alarm method and system based on voice recognition.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A campus anti-bullying alarm method based on voice recognition includes the following steps: Collect students' alarm voice information; Identify the distress message in the alarm voice message; Based on the distress message, determine whether a bullying alarm has been triggered. If so, send an alarm message over the network; otherwise, do not send an alarm message. The location of the student who triggered the alarm is determined based on the alarm information, and alarm location information is generated; Send the alarm location information.

[0005] The beneficial effects of this invention are: by collecting students' alarm information and identifying the distress information in the alarm information, an alarm can be triggered through the distress information, thereby making it easier for students to call for help when they are bullied on campus, and making it easier for the school to know about the bullying incident at the first time, so as to stop or intervene and prevent the bullying from continuing.

[0006] Based on the above technical solution, the present invention can be further improved as follows.

[0007] Furthermore, the alarm voice information includes voice content, volume, and duration.

[0008] Furthermore, identifying the distress message in the alarm voice information includes the following steps: Convert the voice content information in the alarm voice information into text information; Identify distress signals in the text message; The number of distress words is recorded to obtain the number of distress texts; the distress information includes the number of distress texts, the volume, and the duration of the voice.

[0009] Furthermore, the process of determining whether a bullying alarm is triggered based on the distress message involves the following steps: determining whether a bullying alarm is triggered based on the number of distress messages, the volume, and the duration of the voice message.

[0010] Furthermore, determining whether to trigger a bullying alarm based on the number of distress texts, the volume, and the duration of the voice message includes the following steps: If the number of distress messages is greater than or equal to a preset threshold, a bullying alarm will be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold, and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that a bullying alarm should be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that the bullying alarm will not be triggered.

[0011] Furthermore, sending alarm information via the network includes the following steps: The alarm information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm information is sent to the teacher's mobile terminal via the Internet; The teacher's mobile terminal generates a confirmation command based on the alarm information; The system determines whether to report the alarm information based on the confirmation instruction. If yes, a reporting instruction is generated; otherwise, the reporting instruction is not generated. According to the reporting instruction, the alarm information is reported to the information receiving terminal of the education department and the police terminal of the local police department via the Internet.

[0012] Furthermore, there are multiple teacher mobile terminals, each held by a different teacher. Determining whether to report the alarm information based on the confirmation command includes the following steps: If all of the teacher's mobile terminals generate the confirmation command, it is determined that the alarm information needs to be reported. If at least one of the multiple teacher mobile terminals fails to generate the confirmation command, it is determined that the alarm information does not need to be reported.

[0013] Furthermore, the location of the student who triggered the alarm is determined based on the alarm information, and alarm location information is generated, including the following steps: Multiple audio acquisition modules are set up and each audio acquisition module is numbered; the multiple audio acquisition modules are located in different areas of the campus; The location of the audio acquisition module is determined by identifying the number of the audio acquisition module that collects the alarm voice information from the students, thereby determining the alarm location information.

[0014] Furthermore, sending the alarm location information includes the following steps: The alarm location information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm location information is sent to the teacher's mobile terminal via the Internet.

[0015] To address the aforementioned technical problems, this invention provides a campus anti-bullying alarm system based on voice recognition, the specific technical content of which is as follows: A campus anti-bullying alarm system based on voice recognition includes: The audio acquisition module is used to collect students' alarm voice information; An audio recognition module is used to recognize distress messages in the alarm voice messages; The judgment and calculation module is used to determine whether a bullying alarm is triggered based on the distress information. If so, an alarm message is sent through the network; otherwise, no alarm message is sent. The location determination module is used to determine the location of the alarming student based on the alarm information, generate alarm location information, and send the alarm location information. Attached Figure Description

[0016] Figure 1 This is a flowchart of a campus anti-bullying alarm method based on speech recognition in an embodiment of the present invention; Figure 2 This is a schematic diagram of a campus anti-bullying alarm system based on voice recognition, as described in an embodiment of the present invention. Detailed Implementation

[0017] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0018] like Figure 1 As shown in the figure, this embodiment provides a campus anti-bullying alarm method based on voice recognition, including the following steps: S1. Collect students' alarm voice information; the alarm voice information includes voice content information, volume and duration.

[0019] S2. Identify the distress message in the alarm voice message; Identifying the distress message in the alarm voice message includes the following steps: The audio content in the alarm voice message is converted into text; specifically, existing tools such as Yingyi, iFlytek Hearing, and Otter can be used for text conversion. Yingyi: Fast multilingual recognition, suitable for adding subtitles to video / audio. iFlytek Hearing: Strong dialect and multi-speaker recognition with high accuracy. Otter: Real-time transcription and meeting summarization, suitable for meeting recording.

[0020] The text information can be identified by recognizing words or phrases related to distress. Specifically, Natural Language Processing (NLP) techniques can be used to analyze the text content, enabling classification, sentiment analysis, or machine translation. This is suitable for handling complex text; therefore, NLP can be used to identify key characters such as "help" and "alarm." Alternatively, template matching can be used, employing pre-defined templates and regular expressions to locate key fields such as "help" and "alarm" to extract the words or phrases related to distress from the text.

[0021] The number of distress words is recorded to obtain the number of distress texts; the distress information includes the number of distress texts, the volume, and the duration of the voice.

[0022] For example, if the word "help" appears 10 times in the converted text information within a preset time, then the number of the distress words is 10.

[0023] By recording the number of SOS words, the accuracy of alarm confirmation can be improved, preventing some students from accidentally triggering alarms by uttering SOS words during chat.

[0024] S3. Determine whether a bullying alarm is triggered based on the distress message. If so, send an alarm message via the network; otherwise, do not send an alarm message. The specific steps for determining whether to trigger a bullying alarm based on the distress message are as follows: determine whether to trigger a bullying alarm based on the number of distress texts, the volume, and the duration of the voice message.

[0025] Determining whether to trigger a bullying alarm based on the number of distress texts, the volume, and the duration of the voice message includes the following steps: If the number of distress messages is greater than or equal to a preset threshold, a bullying alarm will be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold, and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that a bullying alarm should be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that the bullying alarm will not be triggered.

[0026] For example, if a preset threshold of 10 is set, and a student's alarm voice message contains 10 or more SOS messages, the student is deemed to have been threatened or bullied. When the number of SOS messages is less than 10, the volume and duration of the speech can be used to determine if the student is being bullied or threatened. If a student is emotionally agitated, their voice will be louder, and the verbal conflict will last longer, for example, A minutes (where A is between 5 and 10 minutes). Therefore, if the number of SOS messages is less than 10, the volume is greater than or equal to the preset volume threshold, and the speech duration is greater than or equal to A minutes, the student's alarm message can be considered genuine, thus triggering a bullying alarm.

[0027] Sending alarm information over the network includes the following steps: The alarm information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm information is sent to the teacher's mobile terminal via the Internet; The teacher's mobile terminal generates a confirmation command based on the alarm information; The system determines whether to report the alarm information based on the confirmation instruction. If yes, a reporting instruction is generated; otherwise, the reporting instruction is not generated. According to the reporting instruction, the alarm information is reported to the information receiving terminal of the education department and the police terminal of the local police department via the Internet.

[0028] There are multiple teacher mobile terminals, and each teacher mobile terminal is held by multiple teachers. Determining whether to report the alarm information based on the confirmation command includes the following steps: If all of the teacher's mobile terminals generate the confirmation command, it is determined that the alarm information needs to be reported. If at least one of the multiple teacher mobile terminals fails to generate the confirmation command, it is determined that the alarm information does not need to be reported.

[0029] For example, the teacher's mobile terminal can be integrated with the teacher's mobile phone, or an application can be installed on the teacher's mobile phone to receive alarm information. A confirmation dialog box will pop up through the application. If the teacher confirms the alarm information through the confirmation dialog box, it is determined that the corresponding teacher has confirmed that their alarm information needs to be reported, and the mobile phone will generate the confirmation command. When all teachers confirm that their alarm information needs to be reported, it indicates that the situation of student bullying is relatively serious and needs to be taken seriously by the education department and the police department.

[0030] S4. Determine the location of the student who triggered the alarm based on the alarm information, and generate alarm location information; The location of the student who triggered the alarm is determined based on the alarm information, and alarm location information is generated, including the following steps: Multiple audio acquisition modules are set up and each audio acquisition module is numbered; the multiple audio acquisition modules are located in different areas of the campus; The location of the audio acquisition module is determined by identifying the number of the audio acquisition module that collects the alarm voice information from the students, thereby determining the alarm location information.

[0031] For example, audio acquisition module 1 is located in classroom 1, and audio acquisition module 2 is located in the men's restroom of teaching building 2. If the alarm voice information is acquired from audio acquisition module 2, the student's location is determined to be in the men's restroom of teaching building 2. If the alarm voice information is acquired from audio acquisition module 1, the student's location is determined to be in classroom 1.

[0032] S5. Send the alarm location information.

[0033] Sending the alarm location information includes the following steps: The alarm location information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm location information is sent to the teacher's mobile terminal via the Internet.

[0034] This invention collects students' alarm information and identifies distress messages within the alarm information to trigger an alarm. This allows students to call for help when they are bullied on campus, enabling schools to be aware of bullying incidents immediately and take measures to stop or intervene, thus preventing the bullying from continuing.

[0035] like Figure 2 As shown, in some other embodiments, a campus anti-bullying alarm system based on voice recognition is also provided, including: The audio acquisition module is used to collect students' alarm voice information; An audio recognition module is used to recognize distress messages in the alarm voice messages; The judgment and calculation module is used to determine whether a bullying alarm is triggered based on the distress information. If so, an alarm message is sent through the network; otherwise, no alarm message is sent. The location determination module is used to determine the location of the alarming student based on the alarm information, generate alarm location information, and send the alarm location information.

[0036] The audio acquisition module can use audio acquisition devices such as microphones, which are wall-mounted and fixed to the walls or ceiling of classrooms, corridors, or rooftops. The audio recognition module, judgment and calculation module, and location determination module can all be integrated into a computer. Software corresponding to these modules is integrated into the computer to perform audio recognition, judgment and calculation, and location determination.

[0037] The alarm information is received through the alarm platform of the campus security room, teachers' mobile phones, information receiving terminals of the education department, and police terminals of the local police department.

[0038] If it is necessary to install audio collection modules in private areas such as student dormitories, the alarm voice information corresponding to the alarm information can be stored after the alarm is sent for subsequent evidence collection; voice information that does not trigger a bullying alarm should not be stored to protect students' privacy.

[0039] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the concept and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A campus anti-bullying alarm method based on speech recognition, characterized in that, Includes the following steps: Collect students' alarm voice information; Identify the distress message in the alarm voice message; Based on the distress message, determine whether a bullying alarm has been triggered. If so, send an alarm message over the network; otherwise, do not send an alarm message. The location of the student who triggered the alarm is determined based on the alarm information, and alarm location information is generated; Send the alarm location information.

2. The campus anti-bullying alarm method based on speech recognition according to claim 1, characterized in that, The alarm voice information includes voice content, volume, and duration.

3. The campus anti-bullying alarm method based on speech recognition according to claim 2, characterized in that, Identifying the distress message in the alarm voice message includes the following steps: Convert the voice content information in the alarm voice information into text information; Identify distress signals in the text message; The number of distress words is recorded to obtain the number of distress texts; the distress information includes the number of distress texts, the volume, and the duration of the voice.

4. The campus anti-bullying alarm method based on speech recognition according to claim 3, characterized in that, The specific steps for determining whether to trigger a bullying alarm based on the distress message are as follows: determine whether to trigger a bullying alarm based on the number of distress texts, the volume, and the duration of the voice message.

5. The campus anti-bullying alarm method based on speech recognition according to claim 4, characterized in that, Determining whether to trigger a bullying alarm based on the number of distress texts, the volume, and the duration of the voice message includes the following steps: If the number of distress messages is greater than or equal to a preset threshold, a bullying alarm will be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold, and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that a bullying alarm should be triggered. If the number of distress texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is greater than or equal to the preset volume threshold and the duration of the voice is less than the preset time threshold, then it is determined that the bullying alarm will not be triggered. If the number of SOS texts is greater than or equal to 1 and less than the preset number threshold, and the volume is less than the preset volume threshold and the duration of the voice is greater than or equal to the preset time threshold, then it is determined that the bullying alarm will not be triggered.

6. The campus anti-bullying alarm method based on speech recognition according to claim 1, characterized in that, Sending alarm information over the network includes the following steps: The alarm information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm information is sent to the teacher's mobile terminal via the Internet; The teacher's mobile terminal generates a confirmation command based on the alarm information; The system determines whether to report the alarm information based on the confirmation instruction. If yes, a reporting instruction is generated; otherwise, the reporting instruction is not generated. According to the reporting instruction, the alarm information is reported to the information receiving terminal of the education department and the police terminal of the local police department via the Internet.

7. The campus anti-bullying alarm method based on speech recognition according to claim 6, characterized in that, There are multiple teacher mobile terminals, and each teacher mobile terminal is held by multiple teachers. Determining whether to report the alarm information based on the confirmation command includes the following steps: If all of the teacher's mobile terminals generate the confirmation command, it is determined that the alarm information needs to be reported. If at least one of the multiple teacher mobile terminals fails to generate the confirmation command, it is determined that the alarm information does not need to be reported.

8. The campus anti-bullying alarm method based on speech recognition according to claim 6, characterized in that, The location of the student who triggered the alarm is determined based on the alarm information, and alarm location information is generated, including the following steps: Multiple audio acquisition modules are set up and each audio acquisition module is numbered; the multiple audio acquisition modules are located in different areas of the campus; The location of the audio acquisition module is determined by identifying the number of the audio acquisition module that collects the alarm voice information from the students, thereby determining the alarm location information.

9. The campus anti-bullying alarm method based on speech recognition according to claim 6, characterized in that, Sending the alarm location information includes the following steps: The alarm location information is sent to the alarm platform of the campus security room via the campus LAN or the Internet; The alarm location information is sent to the teacher's mobile terminal via the Internet.

10. A system employing the campus anti-bullying alarm method based on speech recognition as described in any one of claims 1 to 9, characterized in that, include: The audio acquisition module is used to collect students' alarm voice information; An audio recognition module is used to recognize distress messages in the alarm voice messages; The judgment and calculation module is used to determine whether a bullying alarm is triggered based on the distress information. If so, an alarm message is sent through the network; otherwise, no alarm message is sent. The location determination module is used to determine the location of the alarming student based on the alarm information, generate alarm location information, and send the alarm location information.

Images