Intelligent voice recognition system based on artificial intelligence

Abstract

The invention discloses an intelligent voice recognition system based on artificial intelligence, and relates to the technical field of intelligent voice recognition, and the system comprises an input library module, a data analysis module, and an adaptation adjustment module, the input library module is used for obtaining user information and voice instruction information of a corresponding user, and establishing a user library; the input library module comprises a primary input module, a secondary input module, a scene setting module and a data acquisition module, the data analysis module is used for performing corresponding use authorization and condition analysis according to identified user information, and the data analysis module comprises a condition comparison module, a filtering judgment module and a data transmission module. The adaptive adjustment module is used for instruction matching and execution of corresponding instructions, the adaptive adjustment module comprises a privacy authorization module, an execution adjustment module and a data interaction module, and the system improves the use universality and accuracy of the intelligent voice recognition system.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent speech recognition, and in particular to an intelligent speech recognition system based on artificial intelligence. Background Art

[0002] Artificial intelligence is a multidisciplinary research field whose goal is to develop systems and applications that can imitate human intelligence. With the development of computer technology and the continuous improvement of speech signal processing methods, speech recognition technology has also become increasingly sophisticated. It is also a key technology for realizing intelligent applications such as human interaction. It is mainly accomplished through technologies such as signal processing and pattern recognition.

[0003] With the continuous development of science and technology, more and more smart home devices are widely used, bringing great convenience and comfort to people's lives. Among them, the voice control switch light function is an indispensable and important function in the smart home system.

[0004] The traditional way of turning lights on and off requires manual operation, that is, plugging or rotating the switch in a specific position, which is relatively cumbersome; voice control realizes remote control of turning lights on and off. Users only need to shout commands to easily turn lights on or off. The principle is to convert sound signals into electrical signals through the connection of voice recognition technology and smart home devices, and then the smart home devices respond and execute corresponding commands.

[0005] Voice control of switching lights on and off allows people to use lights more conveniently and improves the quality of life to a certain extent. It is foreseeable that in the future technological development, smart home systems will become more and more popular, and the voice control of switching lights on and off will also usher in more extensive applications and development. When we walk around the living room at night, we no longer have to turn the lights on and off manually, we only need to issue simple voice commands to complete the operation. Similarly, if there are elderly people or people with physical disabilities in life, they can also easily operate the lights by voice control of switching lights on and off.

[0006] However, considering the speech defects of the special group of people such as the deaf and mute, the current intelligent voice recognition still has shortcomings and cannot further improve the comfort experience in these aspects. Therefore, how the intelligent voice recognition system can be linked with other smart home devices to achieve more extensive and precise control has become an urgent problem to be solved by people in this field. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent speech recognition system based on artificial intelligence to solve the problems raised in the above background technology.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: an intelligent speech recognition system based on artificial intelligence, comprising a database entry module, a data analysis module, and an adaptation and adjustment module. The database entry module is used to obtain user information and voice command information of the corresponding user, and establish a user database. The database entry module includes a primary entry module, a secondary entry module, a scene setting module and a data acquisition module. The data analysis module is used to perform corresponding usage authorization and situation analysis based on the identified user information. The data analysis module includes a situation comparison module, a filtering and judgment module, and a data transmission module. The adaptation and adjustment module is used for instruction matching and executing corresponding instructions. The adaptation and adjustment module includes a privacy authorization module, a data interaction module, and an execution adjustment module.

[0009] The present invention further illustrates that the primary entry module is used to establish a first user database, and obtain and enter the first user information, wherein the first user is set as the household owner, and has the use authorization function and the first user change function; The secondary entry module is used to establish the second and above user databases, and does not have the use authorization function or the first user modification function; The scene setting module is used to pre-set the transmission execution path corresponding to the sound command in the night scene; The data collection module is used in conjunction with several sound collectors that are set up separately to collect voice information. The sound collector is set at a place where sound collection is required, and the location is set according to user needs. The sound collector is connected to the smart light signal set up nearby.

[0010] The present invention further illustrates that the voice commands in the scene setting module include but are not limited to turning on the light, turning off the light, automatic turning on the light, light color switching X, brightness switching Y, as well as ordinary people's usage mode M1 and special people's usage mode M2.

[0011] The present invention further illustrates that the data collection module is used to collect the walking sound elements of indoor users in advance and in real time, and the walking sound elements include the walking sound frequency and the walking sound loudness.

[0012] The present invention further illustrates that the situation comparison module compares and verifies the voice or walking sound elements received by the data acquisition module with the data set in the primary input module, the secondary input module, and the scene setting module to determine the execution instruction; The data transmission module is used to transmit instructions.

[0013] The present invention further illustrates that the privacy authorization module is connected to the primary input module signal, and is used to assist the first user in inputting other user information, including ordinary people inputting information and special groups inputting information, and the input methods used are also different; When executing instructions, the data interaction module adjusts the instructions according to the real-time detection data, and also plays a role in energy saving.

[0014] The present invention further states that the light color switching X represents the light color at different color temperatures, X is an integer between 1 and x, x is the total number of light colors, and the larger the x, the higher the color temperature; The brightness switch Y represents the brightness under different light colors, Y is an integer between 1 and y, y is the total number of brightness levels, the larger the y is, the higher the brightness is, and the higher the color temperature and the higher the brightness, the more power is consumed; The automatic light-on control is used to automatically adjust the on and off of the smart lights in the corresponding area according to the light-on time and the sound of walking. After completing the sound recording in sequence, it becomes the ordinary person user library data, corresponding to the ordinary person usage mode M1.

[0015] The present invention further illustrates that the operation method of the intelligent speech recognition system based on artificial intelligence comprises the following steps: S1: obtaining first user information and voice command information of the first user, and establishing a first user database, wherein the first user database includes a plurality of voice command entries of the first user; S2: Under the authorization condition of the first user, obtain the second and above user information, and establish a corresponding user library and a transmission execution path for executing corresponding instructions; S3: When used at night, real-time sound information is collected and sound data is analyzed, and compared with the recorded sound data to determine the command; S4: After the command is confirmed, the intelligent light is controlled and adjusted according to the sound location to ensure that the brightness is sufficient to achieve energy saving effect.

[0016] The present invention further illustrates that S2 specifically includes: S21: When the user is an ordinary person, determine that the ordinary person usage mode M1 is a voice sound recognition mode, and establish a corresponding ordinary person user database, which includes a number of voice command inputs of the ordinary person second user and above user information; S22: When the user is a special group, the special group usage mode M2 ​​is determined to be a walking sound recognition mode, and a corresponding special group user database is established, which includes a number of walking sound command entries of the second user and above of the special group; The walking sound loudness is divided into a first sound loudness and a second sound loudness according to daily decibels. The first sound loudness is applicable to the sound loudness generated by daily walking, and the second sound loudness is applicable to recording and starting the walking sound recognition mode within a set time period; The walking sound frequency refers to the number of walking steps within a certain period of time, which is divided into a first sound frequency and a second sound frequency. The first sound frequency is within the daily walking frequency range, and the second sound frequency is suitable for recording and starting the walking sound recognition mode, which is much higher than the daily walking frequency range.

[0017] The present invention further illustrates that the special group uses mode M2 ​​to sequentially switch the instructions of turning on the light, turning off the light, turning on the light automatically, switching the light color X, and switching the brightness Y into walking sound elements.

[0018] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention combines the scene setting module with the sound collector to accurately identify the different needs of users, while facilitating the use of the system by special groups, thereby improving the wide application and accuracy of the intelligent voice recognition system; By using the execution adjustment module, the intelligent voice recognition system will automatically adjust the turning on and off of the smart lights in the corresponding area according to the lighting time and the sound of walking. The sound collector will automatically modify the light brightness according to the sound frequency within the area to ensure the user experience and safety of the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 is a schematic diagram of the overall system of Embodiment 1 of the present invention; Figure 2 It is a schematic diagram of a user library module of the present invention; Figure 3 is a schematic diagram of the overall system of Embodiment 2 of the present invention; Figure 4 It is a schematic diagram of the distribution of the sound collectors of the present invention; Figure 5 It is a schematic diagram of the equipment connection used in the present invention. DETAILED DESCRIPTION

[0020] The following is a further non-limiting detailed description of the technical solution of the present invention in conjunction with the preferred embodiments and the accompanying drawings. Obviously, the described embodiments are only 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.

[0021] For example, see Figure 1-3 ,The present invention provides a technical solution: an intelligent speech recognition system based on artificial intelligence, suitable for night scenes, including a database entry module, a data analysis module, and an adaptation adjustment module; The database entry module is used to obtain user information and the corresponding user's voice command information, and to establish a user database. The data analysis module is used to perform corresponding usage authorization and situation analysis based on the identified user information. The adaptation and adjustment module is used for command matching and execution of corresponding commands, thereby improving the wide range and accuracy of the intelligent voice recognition system.

[0022] The library entry module includes: The primary input module is used to establish a first user database, and obtain and input the first user information, wherein the first user is set as the household owner, and has the use authorization function and the first user change function; The secondary entry module is used to establish the second and above user databases, specifically, including the second user database, the third user database, ..., the nth user database, where n is the maximum number of users set, all of which are set as household members. The authority of household members is lower than that of household heads, and they do not have the use authorization function and the function of changing the first user; The scene setting module is used to pre-set the transmission execution path of the corresponding sound commands in the night scene, such as turning on the light, turning off the light, automatic turning on the light, light color switching X, brightness switching Y, as well as the ordinary people's use mode M1 and the special people's use mode M2; The data collection module is used to cooperate with several sound collectors set up separately to collect voice information. It can also collect the frequency and loudness of the walking sounds of indoor users in advance and in real time, which are recorded as walking sound elements. It should be noted that the sound collector is set at the place where sound collection is required, and the location is set according to user needs, and the sound collector is connected to the smart light signal set near it.

[0023] Data analysis modules include: The situation comparison module compares and verifies the voice or walking sound elements received by the data collection module with the data set in the primary input module, the secondary input module, and the scene setting module to determine the execution instruction; The data transmission module is used to transmit instructions.

[0024] The adaptation module includes: The privacy authorization module is connected to the primary input module signal, and is used to assist the first user in inputting other users' information, including ordinary people's input and special groups' input information, and the input methods used are also different; The data interaction module adjusts the instructions according to the real-time detection data when executing the instructions, and also plays the role of energy saving.

[0025] Specifically, the steps of operating a method of an intelligent speech recognition system based on artificial intelligence are as follows: S1: obtaining first user information and voice command information of the first user, and establishing a first user database, wherein the first user database includes a plurality of voice command entries of the first user; S2: Under the authorization condition of the first user, obtain the second and above user information, and establish a corresponding user library and a transmission execution path for executing corresponding instructions; S21: When the user is an ordinary person, determine that the ordinary person usage mode M1 is a voice sound recognition mode, and establish a corresponding ordinary person user database, which includes a number of voice command inputs of the ordinary person second user and above user information; S22: When the user is a special group, the special group usage mode M2 ​​is determined to be a walking sound recognition mode, and a corresponding special group user database is established, which includes a number of walking sound command entries of the second user and above of the special group; S3: When used at night, real-time sound information is collected and sound data is analyzed, and compared with the recorded sound data to determine the command; S4: After the command is confirmed, the intelligent light is controlled and adjusted according to the sound location to ensure that the brightness is sufficient to achieve energy saving effect.

[0026] Specifically, several voice commands in S1 and S21 include but are not limited to turning on the lights, turning off the lights, automatic light control, light color switching X, and brightness switching Y, wherein light color switching X represents the light color under different color temperatures, X is an integer from 1 to x, x is the total number of light colors, and the larger x is, the higher the color temperature is. Brightness switching Y represents the brightness under different light colors, Y is an integer from 1 to y, y is the total number of brightness levels, and the larger y is, the higher the brightness is. The higher the color temperature and the higher the brightness, the more power is consumed. Automatic light control is used to automatically adjust the smart lights in the corresponding area to turn on and off according to the light-on time and the sound of walking. After completing the sound recording in sequence, it becomes the user library data of ordinary people, corresponding to the ordinary people's usage mode M1.

[0027] In S22, data entry is performed into a user database for special groups. Under the authorization of the first user, the walking sound elements are pre-entered with the help of a sound collector at one of the locations. The walking sound loudness is divided into a first sound loudness and a second sound loudness according to the daily decibel classification. The distinction between the two is between 35 and 50 decibels. The first sound loudness is applicable to the sound loudness generated by daily walking, and the second sound loudness is applicable to recording and starting the walking sound recognition mode within a set time period, including but not limited to special groups. The special group usage mode M2 ​​in this solution is also applicable to ordinary people. The walking sound frequency refers to the number of walking steps within a certain period of time, which is divided into a first sound frequency and a second sound frequency. The first sound frequency is within the daily walking frequency range, and the second sound frequency is applicable to recording and starting the walking sound recognition mode, which is much higher than the daily walking frequency range. The special group use mode M2 ​​switches the instructions of turning on the light, turning off the light, turning on the light automatically, switching the light color to X, and switching the brightness to Y to the walking sound elements in sequence, among which: Turning on the light corresponds to the loudness of the second sound and the frequency of the first sound at the beginning of walking; Turning off the light corresponds to the loudness of the second sound and the frequency of the first sound at the end of walking; The automatic control of turning on the light corresponds to the loudness of the second sound and the frequency of the second sound at the beginning of walking; The light color switching X corresponds to the Xth second sound loudness during walking; Brightness switching Y corresponds to the Y second sound frequency during walking; when X and Y are out of range, the instructions are executed again from low level to high level.

[0028] refer to Figure 3 ,The data analysis module also includes a filtering and judgment module, which is used to filter the noise in the data collection process to ensure the accuracy of the voice and the sound of walking.

[0029] In the second embodiment, the adaptive adjustment module further includes an execution adjustment module, which is connected to the data transmission module by signal and is used for intelligent adjustment of execution instructions.

[0030] Specifically, in S4, the sound position is determined according to the detection range of the sound collector. The detection range of the sound collector is the area of ​​a circle with a certain length as the radius and the position of the sound collector as the center. The sound collector is set in a space-connected environment. If it encounters a wall blocking and forms an independent space, refer to Figure 4 In the H area, the detection range of the corresponding sound collector will cut off the intersecting space, and the remaining space is the detection range of the corresponding sound collector, wherein the detection range of the set sound collector is laid out according to user needs.

[0031] When the sound collector collects the light-on signal, the sound collector turns on the light according to the sound within the detection range, and the light color and brightness remain consistent. When non-light-on automatic control is used, the light-off process is executed when the light-off information is received; When automatic lighting control is adopted, the intelligent voice recognition system will automatically adjust the turning on and off of the smart lights in the corresponding area according to the lighting time and the sound of walking. The end sound collector is set in the scene setting module in advance, which is usually the sound collector closest to the bedroom. The direction of the walking sound can be used to determine whether the user has returned to the room. After the end sound collector collects the return information, the light will be kept on for a time. The remaining sound collectors determine the lights at the corresponding positions according to the location of the walking sound. Usually, the standard for lighting the lights is that under the first sound loudness and the first sound frequency, the lights maintain the first light color and the first brightness, and maintain a certain energy-saving mode. When the sound collector detects the second sound frequency range, it is presumed that the user does not belong to the daily walking frequency range. The sound collector will modify the light brightness according to the sound frequency within the area to ensure the user's experience and safety.

[0032] refer to Figure 5 , a device used in an intelligent speech recognition system based on artificial intelligence, comprising: A sound collector is used to collect voice and movement sounds, and then obtain user information and the corresponding user's voice command information; The data processor outputs corresponding execution instructions according to the identified user information; The linkage interaction terminal is used to connect the data processor with the signal for data transmission; A data recorder, used to record the sound data and the transmission execution path of the corresponding instructions in advance; Authorization interaction terminal, used for first user authorization and function change; Data display terminal, used to display execution data.

[0033] The execution of the system is accomplished through hardware associated with program instructions, and the program can be stored in a computer-readable storage medium.

[0034] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by terms such as "up", "down", "front", "back", "left", and "right" are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0035] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents, and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent speech recognition system based on artificial intelligence, characterized in that: include: The database entry module is used to obtain user information and voice command information of the corresponding user and establish a user database. The database entry module includes a primary entry module, a secondary entry module, a scene setting module and a data collection module; A data analysis module, used to perform corresponding use authorization and situation analysis according to the identified user information, the data analysis module includes a situation comparison module, a filtering and judgment module, and a data transmission module; An adaptation and adjustment module, used for instruction matching and executing corresponding instructions, the adaptation and adjustment module includes a privacy authorization module, a data interaction module, and an execution adjustment module; The primary input module is used to establish a first user database, and obtain and input the first user information, wherein the first user is set as the household owner, and has the use authorization function and the first user change function; The secondary entry module is used to establish the second and above user databases, and does not have the use authorization function or the first user modification function; The scene setting module is used to pre-set the transmission execution path corresponding to the sound command in the night scene; The data collection module is used in conjunction with several sound collectors that are set up separately to collect voice information, wherein the sound collector is set at a place where sound collection is required, and the place is set according to user needs, and the sound collector is connected to the smart light signal set up near it; The voice commands in the scene setting module include but are not limited to turning on the light, turning off the light, automatic turning on the light, light color switching X, brightness switching Y, and ordinary people use mode M1 and special people use mode M2; The light color switch X represents the light color under different color temperatures, X is an integer between 1 and x, x is the total number of light colors, and the larger the x, the higher the color temperature; The brightness switch Y represents the brightness under different light colors, Y is an integer between 1 and y, y is the total number of brightness levels, the larger the y is, the higher the brightness is, and the higher the color temperature and the higher the brightness, the more power is consumed; The light-on automatic control is used to automatically adjust the smart lights in the corresponding area to turn on and off according to the light-on time and the walking sound. After the sound recording is completed in sequence, it becomes the user library data of ordinary people, corresponding to the ordinary people's usage mode M1; The operation method of the intelligent speech recognition system comprises the following steps: S1: obtaining first user information and voice command information of the first user, and establishing a first user database, wherein the first user database includes a plurality of voice command entries of the first user; S2: Under the authorization condition of the first user, obtain the second and above user information, and establish a corresponding user library and a transmission execution path for executing corresponding instructions; S3: When used at night, real-time sound information is collected and sound data is analyzed, and compared with the recorded sound data to determine the command; S4: After the command is determined, the intelligent light is controlled and adjusted according to the sound location to ensure that the brightness is sufficient to achieve the effect of energy saving; The S2 specifically includes: S21: When the user is an ordinary person, determine that the ordinary person usage mode M1 is a voice sound recognition mode, and establish a corresponding ordinary person user database, which includes a number of voice command inputs of the ordinary person second user and above user information; S22: When the user is a special group, the special group usage mode M2 ​​is determined to be a walking sound recognition mode, and a corresponding special group user database is established, which includes a number of walking sound command entries of the second user and above of the special group; The walking sound loudness is divided into a first sound loudness and a second sound loudness according to the daily decibel level. The first sound loudness is applicable to the sound loudness generated by daily walking, and the second sound loudness is applicable to recording and starting the walking sound recognition mode within a set time period; The walking sound frequency refers to the number of steps in a certain period of time, which is divided into a first sound frequency and a second sound frequency. The first sound frequency is within the daily walking frequency range, and the second sound frequency is suitable for recording and starting the walking sound recognition mode, which is much higher than the daily walking frequency range; The special group usage mode M2 ​​switches the instructions of turning on the light, turning off the light, turning on the light automatically, switching the light color X, and switching the brightness Y into walking sound elements in sequence.

Images