Voice recognition alarm method and system, electronic equipment and medium

Abstract

The invention discloses a sound recognition alarm method and system, electronic equipment and a medium, and relates to the technical field of sound recognition, a plurality of preset time periods are set, corresponding sound acquisition sensitivities are configured corresponding to the preset time periods and sound types respectively, and the sound acquisition sensitivities are marked in a sound existence model; sound information in the environment is collected in real time according to the sound existence model, the sound information comprises audio information, sound collection time and sound decibel, and the sound information is constructed in the sound existence model; and analyzing the audio information to obtain a sound type, and implementing an alarm mode for the interference sound corresponding to the sound decibel which does not meet the preset condition. According to the method, the interference sound feature set and the normal sound feature set are stored to obtain the sound type library, a large number of powerful databases used for follow-up environment sound recognition are provided, square dance music can be quickly recognized and determined in follow-up actual life, and response can be efficiently carried out for alarm processing.

Description

Technical Field

[0001] The present invention relates to the technical field of sound recognition, and in particular to a sound recognition alarm method, system, electronic equipment and medium. Background Art

[0002] Square dancing, as a mass fitness activity that is deeply loved by middle-aged and elderly people, not only enriches the elderly’s later life and promotes their physical and mental health, but also strengthens communication and interaction among community members and creates a harmonious and friendly community atmosphere.

[0003] However, the popularity of square dancing is also accompanied by a series of problems, the most prominent of which is noise pollution. Due to the improper selection of some square dance venues, such as close to residential areas, schools, hospitals and other places that require a quiet environment, high-decibel music and cheers often interfere with the study, work and rest of nearby residents. At present, it is difficult to identify square dancing in a complex sound environment in the regulation of square dancing, and it is difficult to issue an alarm. Summary of the invention

[0004] The purpose of the present invention is to provide a sound recognition alarm method, system, electronic device and medium to solve the shortcomings of the background technology.

[0005] In order to achieve the above object, the present invention provides the following technical solution: a sound recognition alarm method, comprising the following steps: Setting multiple preset time periods, configuring corresponding sound collection sensitivities for the multiple preset time periods and sound types, and marking them in the sound existence model, wherein the sound types include normal sound and interference sound; According to the sound existence model, the sound information in the environment is collected in real time, wherein the sound information includes audio information, the time when the sound is collected, and the sound decibel, and the sound information is constructed in the sound existence model; The audio information is analyzed to obtain the sound type, and an alarm is implemented for the interference sound corresponding to the sound decibel that does not meet the preset conditions.

[0006] In a preferred embodiment, the step of setting a plurality of preset time periods, configuring corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and marking them in the sound existence model includes: Determine the jurisdiction of sound recognition, obtain the geographical environment within the jurisdiction, and construct a three-dimensional sound existence model based on the geographical environment and the jurisdiction; Marking multiple sound collection points in the sound existence model, setting multiple wandering position ends in the sound existence model, wherein analog channels are constructed between the wandering position ends and the multiple sound collection points for bidirectional communication, and the wandering position ends are ports with storage areas; Set multiple preset time periods, and configure corresponding sound collection sensitivities corresponding to the multiple preset time periods; Constructing identification bars and formulating a sound type library, wherein the sound type library stores an interference sound feature set and a normal sound feature set; The interference sound feature set and the normal sound feature set in the sound type library are respectively used as interference sound and normal sound; Disturbing sounds and normal sounds were considered as sound types; The preset time period and the corresponding sound collection sensitivity are marked in the sound presence model.

[0007] In a preferred embodiment, the steps of constructing the identification bar and formulating the sound type library include: An audio storage bar is set, and a plurality of recognition nodes are set on the audio storage bar, wherein the plurality of recognition nodes on the audio storage bar are all configured to connect to a cloud server; The audio storage strip is structurally replicated to obtain multiple type library audio storage strips, and multiple identification nodes on the multiple type library audio storage strips are all connected to the cloud server corresponding to the audio storage strips; Binding the type library audio storage strips to the audio of the interfering sound and the audio of the normal sound one-to-one, respectively, placing the audio of the interfering sound and the audio of the normal sound in the corresponding bound type library audio storage strips to obtain an interfering sound feature set and a normal sound feature set; The interference sound feature set and the normal sound feature set are stored to obtain a sound type library.

[0008] In a preferred embodiment, the step of placing the audio of the interfering sound and the audio of the normal sound in the corresponding bound type library audio storage strips to obtain the interfering sound feature set and the normal sound feature set comprises: Place the audio of the interfering sound and the audio of the normal sound in the audio storage bar of the corresponding bound type library respectively; Collecting feature information of the audio of the interfering sound or the normal sound through the identification nodes in the audio storage strip of the type library, wherein the feature information includes the sound frequency change information between adjacent identification nodes and the difference information between the tone peaks between adjacent identification nodes; The feature information is stored in the corresponding cloud server through adjacent recognition nodes to obtain an interference sound feature set and a normal sound feature set in each type library audio storage strip.

[0009] In a preferred embodiment, the step of constructing the sound information into the sound existence model includes: Collecting audio information in the environment within the jurisdiction corresponding to the sound existence model, the time of collecting the sound, the audio time, and the sound decibel as the sound information; Extracting the timbre of the audio information, and dividing the audio information according to the timbre and the audio time to obtain multiple independent audio information; When multiple independent audio information is obtained, a corresponding number of roaming position ends are enabled to roam in the sound existence model. The roaming position end and the multiple sound collection points are always connected. The multiple sound collection points transmit the sound information corresponding to the collected independent audio information back to the roaming position end through the analog channel. The multiple sound collection points and the roaming position end can jointly determine the location where the sound information is generated as the generation location corresponding to the independent audio information, and mark the multiple independent audio information in the sound existence model according to the generation location.

[0010] In a preferred embodiment, the step of analyzing the audio information to obtain the sound type and implementing an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions includes: Processing the independent audio information according to the audio storage strip to obtain the sound type; When the sound type is an interference sound and the sound decibel does not meet the preset conditions, an alarm method is implemented, wherein the alarm method includes emitting a sound wave with a phase opposite to the interference sound for noise reduction, actively reducing the noise of normal sound, and alarming.

[0011] In a preferred embodiment, the processing of the independent audio information according to the audio storage strip to obtain the sound type includes: The audio storage strip is copied according to the number of independent audio information to obtain a plurality of identification audio storage strips; The independent audio information is placed one-to-one in the identification audio storage strip, and the identification nodes on the identification audio storage strip extract the identification feature information of the independent audio information, wherein the identification feature information includes the sound frequency change information between adjacent identification nodes and the difference information between the tone peaks between adjacent identification nodes; The identification feature information is stored in the corresponding cloud server through adjacent identification nodes and matched with the interference sound feature set and the normal sound feature set in the sound type library to obtain the sound types corresponding to the multiple independent audio information.

[0012] The present invention also provides a sound recognition alarm system, comprising: A setting module, used to set a plurality of preset time periods, respectively configure corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and mark them in the sound existence model, wherein the sound types include normal sound and interference sound; An analysis module connected to the setting module is used to collect sound information in the environment in real time according to the sound existence model, wherein the sound information includes audio information, the time of sound collection and the sound decibel, and construct the sound information in the sound existence model; The alarm module is connected to the analysis module and is used to analyze the audio information to obtain the sound type and implement an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention can extract square dance music from ambient sound, store the feature information in the corresponding cloud server through adjacent recognition nodes, obtain an interference sound feature set and a normal sound feature set in each type library audio storage strip, realize the collection and storage of features of various audios under big data, store the interference sound feature set and the normal sound feature set to obtain a sound type library, have a large number of powerful databases for subsequent ambient sound recognition, can help to quickly identify and determine square dance music in subsequent real life, and can efficiently respond to alarm processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0015] Figure 1 The figure is a flow chart of the method of the present invention.

[0016] Figure 2 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1, please refer to Figure 1 As shown, the sound recognition alarm method described in this embodiment includes the following steps: S1. Set a plurality of preset time periods, configure corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and mark them in the sound existence model, wherein the sound types include normal sound and interference sound; S2. According to the sound existence model, the sound information in the environment is collected in real time, wherein the sound information includes audio information, the time of collecting the sound, and the sound decibel, and the sound information is constructed in the sound existence model (represented by the sound information being marked in the sound existence model, for example, the set parameter of the sound information is A, and A is marked in the sound existence model at the location where the sound information is generated in the sound existence model); S3, analyzing the audio information to obtain the sound type, and implementing an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions; As described in the above steps S1-S3, the popularity of square dance is also accompanied by a series of problems, the most prominent of which is noise pollution. Due to improper selection of some square dance venues, such as places that need a quiet environment such as adjacent to residential areas, schools, hospitals, etc., high-decibel music and cheers often interfere with the study, work and rest of surrounding residents. It is difficult to identify square dance in a complex sound environment in the current control of square dance, and then it is difficult to perform alarm processing. In this application, square dance music can be extracted from ambient sound, and feature information is stored in the corresponding cloud server through adjacent recognition nodes to obtain interference sound feature set and normal sound feature set in each type library audio storage strip, which can realize the collection and storage of various audio features under big data, and the interference sound feature set and normal sound feature set are stored to obtain a sound type library, with a large number of powerful databases for subsequent ambient sound recognition, which can help the rapid identification of square dance music in subsequent real life to determine, and can efficiently react to alarm processing.

[0019] In one embodiment, the step S1 of setting a plurality of preset time periods, configuring corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and marking them in the sound existence model includes: S11, determining the jurisdiction of sound recognition, obtaining the geographical environment within the jurisdiction, and performing three-dimensional construction based on the geographical environment and the jurisdiction to obtain a sound existence model; S12, marking a plurality of sound collection points in the sound existence model, and setting a plurality of wandering position terminals in the sound existence model, wherein analog channels are constructed between the wandering position terminals and the plurality of sound collection points for bidirectional communication, and the wandering position terminals are ports provided with storage areas; S13, setting a plurality of preset time periods, and configuring corresponding sound collection sensitivities corresponding to the plurality of preset time periods respectively; S14, constructing an identification bar and formulating a sound type library, wherein the sound type library stores an interference sound feature set and a normal sound feature set; S15, using the interference sound feature set and the normal sound feature set in the sound type library as interference sound and normal sound respectively; S16, treating the interfering sound and the normal sound as the sound type; S17, marking the preset time period and the corresponding sound collection sensitivity in the sound existence model; As described in the above steps S11-S17, a sound recognition system has a certain management scope. Sound is recognized within the management scope. First, the jurisdiction of the sound recognition is determined, and the geographical environment within the jurisdiction is obtained. A sound existence model is constructed in three dimensions based on the geographical environment and the jurisdiction. The sound existence model is a scene model constructed in three dimensions within the space of the jurisdiction. Here, as the sound existence model, for example, in a leisure garden in a residential area in the city, the leisure garden is used as the jurisdiction, so the buildings and geographical environment in the leisure garden are constructed in three dimensions to obtain a three-dimensional scene model showing the leisure garden. , here is used as a sound existence model. In the actual sound collection process, it is necessary to mark multiple sound collection points in the sound existence model. The sound collection point is a device used for subsequent sound information collection. The device is installed on site to collect sound information within the jurisdiction. Multiple wandering position terminals are set in the sound existence model. Among them, analog channels are constructed between the wandering position terminals and multiple sound collection points for two-way communication. The wandering position terminal is a port with a storage area. When there is no sound, the multiple wandering position terminals are not enabled when there is no need to simulate the location where the sound is generated. When there is a location where the sound is generated and needs to be simulated and determined, The number of sounds is determined according to the need to simulate and enable the corresponding wandering position end. In actual use, for example, if there is sound information in a certain area within the jurisdiction, the wandering position end is enabled to wander in the sound existence model. The wandering position end and multiple sound collection points are always connected. Therefore, the sound information corresponding to the independent audio information collected by multiple sound collection points is transmitted back to the wandering position end through the analog channel. The analog channel simulates the attenuation characteristics and directional characteristics of the actual sound transmission in the environment. Multiple sound collection points and the wandering position end can jointly determine the location where the sound information is generated. The storage area of ​​the wandering position end is used to store the sound information that meets the requirements of the sound in the environment. The position of the wandering position end of the transmission attenuation feature and the directional feature in the sound existence model is the position where the sound information is generated, which can determine the position where the independent audio information is generated. Therefore, it is necessary to pre-formulate the sound recognition alarm rules for the management scope. The first purpose is to identify the square dance music in the environment and the alarm that does not meet the subsequent preset conditions. Therefore, the alarm rules should be set according to the conditions of the square dance music in the actual life of the environment. First of all, the time when square dance starts is mostly in the afternoon or morning and evening. For example, 4 pm to 2 am and 6 am to 10 am are time periods where square dance is prone to exist;There is basically no square dancing between 2 a.m. and 6 a.m. and between 10 a.m. and 4 p.m., and multiple time periods are preset according to the actual living environment, and corresponding sound collection sensitivities are configured for the preset time periods. The sound collection sensitivity is adjusted by changing the working number of the sensitivity microphone receiving module to adjust the sensitivity of the sound decibel detection, and the working number of the sensitivity microphone receiving module can be reasonably used according to the sensitivity required for actual collection. In addition, this application is to identify the alarm that square dance music affects the environmental noise, but in real life there will be other normal life sounds, such as car horns, hawking and other environmental sounds, so it is necessary to distinguish the environmental sounds as sound types, and the sound of square dance music is used here as interference sound. Sound, other life sounds are regarded as normal sounds. In order to better identify the interference sounds, an identification bar is constructed for the subsequent type recognition of the collected sounds. In addition, it is necessary to pre-formulate an interference sound feature set and a normal sound feature set for storage to obtain a sound type library. The interference sound feature set here is the feature set corresponding to the big data music library. The interference sound feature set and the normal sound feature set in the sound type library are respectively regarded as interference sounds and normal sounds. The interference sounds and normal sounds are regarded as sound types. The preset time period and the corresponding sound collection sensitivity are marked in the sound existence model. The sound existence model can be used to display the recognition alarm management of the sound, which has faster recognition efficiency and can improve recognition accuracy. In one embodiment, the step S13 of constructing the identification bar and formulating the sound type library includes: S131, setting an audio storage bar, setting multiple identification nodes on the audio storage bar, and each of the multiple identification nodes on the audio storage bar is configured to connect to a cloud server; S132, structurally replicating the audio storage bar to obtain multiple type library audio storage bars, wherein multiple identification nodes on the multiple type library audio storage bars are all connected to the cloud server corresponding to the audio storage bar; S133, binding the audio storage bar of the type library to the audio of the interference sound and the audio of the normal sound one-to-one, respectively, placing the audio of the interference sound and the audio of the normal sound in the corresponding bound audio storage bar of the type library to obtain an interference sound feature set and a normal sound feature set; S134, storing the interference sound feature set and the normal sound feature set to obtain a sound type library.

[0020] As described in the above steps S131-S134, firstly, an audio storage bar is set, which can store and carry a large amount of audio of interference sound and audio of normal sound. In addition, multiple identification nodes are set on the audio storage bar, and the multiple identification nodes on the audio storage bar are configured to connect to a cloud server. The type library audio storage bar is bound one-to-one with the audio of interference sound and the audio of normal sound respectively, and the audio of interference sound and the audio of normal sound are placed in the corresponding bound type library audio storage bar to obtain an interference sound feature set and a normal sound feature set. The identification node is used to identify the feature information in the audio, and then the feature information is stored in the cloud server. The feature information here is the sound frequency change information between adjacent identification nodes, the sound frequency change information between adjacent identification nodes, and the sound frequency change information between adjacent identification nodes. The difference information between the peak values ​​is mainly used to identify square dance music. Square dance music is basically regular audio. Therefore, after collection, it can be used for rapid identification and determination of square dance music in subsequent actual situations. Square dance music can be extracted from ambient sound, and feature information can be stored in the corresponding cloud server through adjacent recognition nodes to obtain interference sound feature sets and normal sound feature sets in each type library audio storage strip. It can realize the collection and storage of features of various audios under big data, store interference sound feature sets and normal sound feature sets to obtain a sound type library, and has a large number of powerful databases for subsequent ambient sound recognition, which can help to quickly identify and determine square dance music in subsequent actual life, and can efficiently respond to alarm processing; In one embodiment, the step S133 of placing the audio of the interfering sound and the audio of the normal sound in the corresponding bound type library audio storage strips to obtain the interfering sound feature set and the normal sound feature set includes: S1331, placing the audio of the interfering sound and the audio of the normal sound into the corresponding bound type library audio storage bar respectively; S1332, collecting feature information of the audio of the interfering sound or the normal sound through the identification nodes in the audio storage strip of the type library, wherein the feature information includes the sound frequency change information between adjacent identification nodes and the difference information between the tone peaks between adjacent identification nodes; S1333. Store the feature information in the corresponding cloud server through adjacent recognition nodes to obtain an interference sound feature set and a normal sound feature set in each type library audio storage strip.

[0021] In one embodiment, the step S2 of constructing the sound information into the sound existence model includes: S21, collecting audio information in the environment within the jurisdiction corresponding to the sound existence model, the time of collecting the sound, the audio time, and the sound decibel as the sound information; S22, extracting the timbre of the audio information, and dividing the audio information according to the timbre and the audio time to obtain a plurality of independent audio information; S23, when a plurality of independent audio information is obtained, a corresponding number of wandering position terminals are enabled to wander in the sound existence model, the wandering position terminal and the plurality of sound collection points are always in a connected state, the plurality of sound collection points transmit the sound information corresponding to the collected independent audio information to the wandering position terminal through an analog channel in reverse, the plurality of sound collection points and the wandering position terminal can jointly determine the position where the sound information is generated as the generation position corresponding to the independent audio information, and the plurality of independent audio information are marked in the sound existence model according to the generation position; As described in the above steps S21-S23, in actual collection, the audio information in the collection environment within the jurisdiction corresponding to the sound existence model, the time of collecting the sound, the audio time and the sound decibel are used as the sound information. Since there are multiple sounds in the environment, the collected audio is also a mixture of multiple sounds. Therefore, it is necessary to extract the timbre of the audio information, and divide the audio information according to the timbre and audio time to obtain multiple independent audio information. Audio time: The duration of a song is generally 3-6 minutes, and the longest is no more than 7-8 minutes, but there is rarely music less than 90 seconds, and there is basically no music less than 20 seconds. Therefore, the range of the preset time period is set to 20s-90 seconds, which basically covers all the music content. Therefore, it is possible to accurately identify whether the sound is square dance music. In addition, it can also be set to 30s, 45s, etc., so that part of the sound can be screened out according to the audio time to obtain multiple independent audio information. After obtaining the multiple independent audio information, the corresponding generation positions of the multiple independent audio information are obtained, and the multiple independent audio information are marked in the sound existence model according to the generation position. The position where the audio is generated on the sound existence model can be used to better help personnel understand whether the sound is interference information, and in subsequent alarms, it can assist the police to maintain the living noise environment.

[0022] In one embodiment, the step S3 of analyzing the audio information to obtain the sound type and implementing an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions includes: S31, processing the independent audio information according to the audio storage bar to obtain the sound type; S32. When the sound type is an interference sound and the sound decibel does not meet the preset conditions, an alarm method is implemented, wherein the alarm method includes emitting a sound wave with a phase opposite to the interference sound to reduce noise, actively reducing noise for normal sounds, and alarming.

[0023] In one embodiment, the independent audio information is processed according to the audio storage strip to obtain the sound type S31, including: S311, duplicating the audio storage strip according to the number of independent audio information to obtain a plurality of identified audio storage strips; S312, placing the independent audio information one-to-one in the identification audio storage strip, and extracting identification feature information of the independent audio information from the identification nodes on the identification audio storage strip, wherein the identification feature information includes sound frequency change information between adjacent identification nodes and difference information between tone peaks between adjacent identification nodes; S313, storing the recognition feature information in the corresponding cloud server through adjacent recognition nodes, and matching it with the interference sound feature set and the normal sound feature set in the sound type library to obtain the sound types corresponding to the multiple independent audio information.

[0024] As described in the above steps S311-S313, after collecting multiple independent audio information, the audio storage strips are copied corresponding to the number of independent audio information to obtain multiple identification audio storage strips, and the identification audio storage strips are used to extract features of the independent audio information to obtain identification feature information, wherein the identification feature information includes sound frequency change information between adjacent identification nodes and gap information between tone peaks between adjacent identification nodes. The identification feature information is stored in the corresponding cloud server through adjacent identification nodes and matched with the interference sound feature set and the normal sound feature set in the sound type library to obtain multiple independent audio information. The corresponding sound type, and then it can distinguish between normal sound and interference sound when multiple independent audio information are distinguished, and can be quickly analyzed and identified, greatly improving the subsequent alarm response ability. In addition, when the sound type is interference sound and the sound decibel does not meet the preset conditions, an alarm method is implemented, wherein the alarm method includes emitting a sound wave with a phase opposite to the interference sound for noise reduction, actively reducing noise for normal sound, and alarming, emitting a warning sound and light, and / or sending a prompt message to relevant personnel; actively reducing noise for the normal sound: emitting a sound wave with a phase opposite to the interference sound; actively reducing noise for sound includes: emitting a sound wave with a phase opposite to the interference sound. By emitting a warning sound and light, the crowd dancing in the square can be prompted to avoid causing adverse interference to the people around them. In addition, active noise reduction uses a sound wave with a phase opposite to the sound, including identifying the sound wave of the current sound, generating an opposite sound wave, and in addition, the sound wave position of the current sound can be detected, and the opposite sound wave can be sent in the direction of the sound wave position of the current sound.

[0025] Example 2, please refer to Figure 2 As shown, the sound recognition alarm system described in this embodiment includes: A setting module, used to set a plurality of preset time periods, respectively configure corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and mark them in the sound existence model, wherein the sound types include normal sound and interference sound; An analysis module connected to the setting module is used to collect sound information in the environment in real time according to the sound existence model, wherein the sound information includes audio information, the time of sound collection and the sound decibel, and construct the sound information in the sound existence model; The alarm module is connected to the analysis module and is used to analyze the audio information to obtain the sound type and implement an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions.

[0026] It should be noted that the interference sound feature set and the normal sound feature set in the sound type library are respectively used as interference sound and normal sound, and the interference sound and normal sound are used as sound types. The preset time period and the corresponding sound collection sensitivity are marked in the sound existence model, and the recognition alarm management of the display sound can be performed through the sound existence model, which has a faster recognition efficiency and can improve the recognition accuracy; the alarm method includes emitting a sound wave with a phase opposite to the interference sound for noise reduction, actively reducing noise for normal sound, and alarming, emitting a warning sound and light, and / or sending a prompt message to relevant personnel; actively reducing noise for the normal sound: the active noise reduction of the sound includes: emitting a sound wave with a phase opposite to the interference sound. By emitting a warning sound and light, the crowd dancing in the square can be prompted to avoid causing adverse interference to the people around. In addition, active noise reduction uses a sound wave with a phase opposite to the sound, including identifying the sound wave of the current sound, generating an opposite sound wave, and in addition, the sound wave position of the current sound can be detected, and the opposite sound wave can be sent in the direction of the sound wave position of the current sound.

[0027] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A sound recognition alarm method, characterized in that: The following steps are involved: Setting multiple preset time periods, configuring corresponding sound collection sensitivities for the multiple preset time periods and sound types, and marking them in the sound existence model, wherein the sound types include normal sound and interference sound; According to the sound existence model, the sound information in the environment is collected in real time, wherein the sound information includes audio information, the time when the sound is collected, and the sound decibel, and the sound information is constructed in the sound existence model; The audio information is analyzed to obtain the sound type, and an alarm is implemented for the interference sound corresponding to the sound decibel that does not meet the preset conditions.

2. A sound recognition alarm method according to claim 1, characterized in that: The step of setting a plurality of preset time periods, configuring corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and marking them in the sound existence model includes: Determine the jurisdiction of sound recognition, obtain the geographical environment within the jurisdiction, and construct a three-dimensional sound existence model based on the geographical environment and the jurisdiction; Marking multiple sound collection points in the sound existence model, setting multiple wandering position ends in the sound existence model, wherein analog channels are constructed between the wandering position ends and the multiple sound collection points for bidirectional communication, and the wandering position ends are ports with storage areas; Set multiple preset time periods, and configure corresponding sound collection sensitivities corresponding to the multiple preset time periods; Constructing identification bars and formulating a sound type library, wherein the sound type library stores an interference sound feature set and a normal sound feature set; The interference sound feature set and the normal sound feature set in the sound type library are respectively used as interference sound and normal sound; Disturbing sounds and normal sounds were considered as sound types; The preset time period and the corresponding sound collection sensitivity are marked in the sound presence model.

3. A sound recognition alarm method according to claim 2, characterized in that: The steps of constructing the identification bar and formulating the sound type library include: An audio storage bar is set, and a plurality of recognition nodes are set on the audio storage bar, wherein the plurality of recognition nodes on the audio storage bar are all configured to connect to a cloud server; The audio storage strip is structurally replicated to obtain multiple type library audio storage strips, and multiple identification nodes on the multiple type library audio storage strips are all connected to the cloud server corresponding to the audio storage strips; Binding the type library audio storage strips to the audio of the interfering sound and the audio of the normal sound one-to-one, respectively, placing the audio of the interfering sound and the audio of the normal sound in the corresponding bound type library audio storage strips to obtain an interfering sound feature set and a normal sound feature set; The interference sound feature set and the normal sound feature set are stored to obtain a sound type library.

4. A sound recognition alarm method according to claim 3, characterized in that: The step of placing the audio of the interfering sound and the audio of the normal sound in the corresponding bound type library audio storage strips to obtain the interfering sound feature set and the normal sound feature set comprises: Place the audio of the interfering sound and the audio of the normal sound in the audio storage bar of the corresponding bound type library respectively; Collecting feature information of the audio of the interfering sound or the normal sound through the identification nodes in the audio storage strip of the type library, wherein the feature information includes the sound frequency change information between adjacent identification nodes and the difference information between the tone peaks between adjacent identification nodes; The feature information is stored in the corresponding cloud server through adjacent recognition nodes to obtain an interference sound feature set and a normal sound feature set in each type library audio storage strip.

5. A sound recognition alarm method according to claim 1, characterized in that: The step of constructing the sound information into the sound existence model comprises: Collecting audio information in the environment within the jurisdiction corresponding to the sound existence model, the time of collecting the sound, the audio time, and the sound decibel as the sound information; Extracting the timbre of the audio information, and dividing the audio information according to the timbre and the audio time to obtain multiple independent audio information; When multiple independent audio information is obtained, a corresponding number of roaming position ends are enabled to roam in the sound existence model. The roaming position end and the multiple sound collection points are always connected. The multiple sound collection points transmit the sound information corresponding to the collected independent audio information back to the roaming position end through the analog channel. The multiple sound collection points and the roaming position end can jointly determine the location where the sound information is generated as the generation location corresponding to the independent audio information, and mark the multiple independent audio information in the sound existence model according to the generation location.

6. A sound recognition alarm method according to claim 1, characterized in that: The step of analyzing the audio information to obtain the sound type and implementing an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions includes: Processing the independent audio information according to the audio storage strip to obtain the sound type; When the sound type is an interference sound and the sound decibel does not meet the preset conditions, an alarm method is implemented, wherein the alarm method includes emitting a sound wave with a phase opposite to the interference sound for noise reduction, actively reducing the noise of normal sound, and alarming.

7. A sound recognition alarm method according to claim 6, characterized in that: The step of processing the independent audio information according to the audio storage strip to obtain the sound type includes: The audio storage strip is copied according to the number of independent audio information to obtain a plurality of identification audio storage strips; The independent audio information is placed one-to-one in the identification audio storage strip, and the identification nodes on the identification audio storage strip extract the identification feature information of the independent audio information, wherein the identification feature information includes the sound frequency change information between adjacent identification nodes and the difference information between the tone peaks between adjacent identification nodes; The identification feature information is stored in the corresponding cloud server through adjacent identification nodes and matched with the interference sound feature set and the normal sound feature set in the sound type library to obtain the sound types corresponding to the multiple independent audio information.

8. A sound recognition alarm system, used to implement a sound recognition alarm method according to any one of claims 1 to 7, characterized in that: include: A setting module, used to set a plurality of preset time periods, respectively configure corresponding sound collection sensitivities for the plurality of preset time periods and sound types, and mark them in the sound existence model, wherein the sound types include normal sound and interference sound; An analysis module connected to the setting module is used to collect sound information in the environment in real time according to the sound existence model, wherein the sound information includes audio information, the time of sound collection and the sound decibel, and construct the sound information in the sound existence model; The alarm module is connected to the analysis module and is used to analyze the audio information to obtain the sound type and implement an alarm method for the interference sound corresponding to the sound decibel that does not meet the preset conditions.

9. An electronic device, characterized in that: Electronic equipment includes: A processor and a memory, wherein the memory is communicatively connected to the processor; The memory is used to store at least one executable instruction executed by the processor, and the processor is used to execute the executable instruction to implement a sound recognition alarm method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a sound recognition alarm method according to any one of claims 1 to 7 is implemented.

Images