CN117672200B - Control method, equipment and system of Internet of things equipment - Google Patents
Control method, equipment and system of Internet of things equipment Download PDFInfo
- Publication number
- CN117672200B CN117672200B CN202410147789.7A CN202410147789A CN117672200B CN 117672200 B CN117672200 B CN 117672200B CN 202410147789 A CN202410147789 A CN 202410147789A CN 117672200 B CN117672200 B CN 117672200B
- Authority
- CN
- China
- Prior art keywords
- voice
- voice interaction
- record
- internet
- control instruction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000003993 interaction Effects 0.000 claims abstract description 246
- 230000000875 corresponding effect Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 10
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/10—Speech classification or search using distance or distortion measures between unknown speech and reference templates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/194—Calculation of difference between files
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y40/00—IoT characterised by the purpose of the information processing
- G16Y40/30—Control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
- H04L67/125—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks involving control of end-device applications over a network
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Telephonic Communication Services (AREA)
Abstract
The method, the device and the system can determine the confidence coefficient of the voice interaction information for each expected control instruction based on the analysis of the voice interaction information and the voice interaction record big data, can realize the control of the Internet of things equipment in a fuzzy voice control mode, and are reasonable in the confidence coefficient of the determined expected control instruction, so that the control accuracy of the fuzzy voice control of the Internet of things equipment is improved.
Description
Technical Field
The application relates to the technical field of the internet of things, in particular to a control method, equipment and a system of internet of things equipment.
Background
How to realize the control of the internet of things equipment is always one of the important components of the internet of things technology. Along with the development of voice recognition technology, in order to facilitate the control of the internet of things equipment, a voice control technology is also introduced into the internet of things equipment. However, most of the voice control instructions in the existing voice control technology require explicit target keywords and instruction keywords so as to determine which control instruction is sent to which internet of things device, and when the number of the internet of things devices applying the voice control technology increases in the same application scene, a large number of differentiated target keywords and instruction keywords are required to realize the correct control of each internet of things device, that is, it is inconvenient to control the internet of things device by using the existing voice control technology.
In order to control the internet of things equipment in a more convenient manner, a fuzzy voice control technology is also proposed in the related art, namely a voice control technology which does not need to define target keywords and instruction keywords, however, although the fuzzy voice control technology is more convenient, the accuracy is relatively poor.
Disclosure of Invention
The application provides a control method, equipment and system of Internet of things equipment, which are beneficial to improving the accuracy of fuzzy voice control of the Internet of things equipment.
In a first aspect, the present application provides a control method for an internet of things device. The method is applied to a controller, and the controller is connected with a plurality of Internet of things devices.
The method comprises the following steps:
acquiring voice interaction record big data and voice interaction information, wherein the voice interaction information comprises voice text content and voice voiceprint identification, and the voice interaction record in the voice interaction record big data comprises the voice text content, the voice voiceprint identification, terminal identification and control instruction data;
retrieving the voice interaction record big data based on the terminal identifiers of the plurality of internet of things devices to obtain a first voice interaction data set, wherein the terminal identifiers of the voice interaction records in the first voice interaction data set and one of the terminal identifiers of the plurality of internet of things devices belong to the same terminal category;
searching the voice interaction record big data based on the voice voiceprint identifier to obtain a second voice interaction data set, wherein the terminal identifier of the voice interaction record in the second voice interaction data set and the voice interaction information carry the same voice voiceprint identifier;
retrieving the voice interaction record big data based on the terminal identifiers of the plurality of internet of things devices to obtain a third voice interaction data set, wherein the terminal identifiers of the voice interaction records in the third voice interaction data set are the same as one of the terminal identifiers of the plurality of internet of things devices;
and determining an expected control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction records in the first voice interaction data set, the second voice interaction data set and the second voice interaction data set, wherein the expected control instruction carries the confidence degree.
By adopting the technical scheme, the confidence coefficient of the voice interaction information for each expected control instruction can be determined based on the analysis of the voice interaction information and the voice interaction record big data, the control of the Internet of things equipment can be realized in a fuzzy voice control mode, and the determined confidence coefficient of the expected control instruction is also reasonable, so that the control accuracy of the fuzzy voice control of the Internet of things equipment is improved.
Further, determining the expected control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set, the second voice interaction data set and the second voice interaction data set includes:
determining a plurality of candidate control instructions and basic confidence scores of each candidate control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set;
based on the candidate control instruction, determining a first confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the second voice interaction data set;
based on the candidate control instruction, determining a second confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the third voice interaction data set;
and calculating the confidence coefficient of the candidate control instruction according to the basic confidence coefficient score, the first confidence coefficient weight and the second confidence coefficient weight.
Further, determining a plurality of candidate control commands and a basic confidence score of each candidate control command according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set comprises:
in the voice interaction records of the first voice interaction data set, determining the voice interaction record with the voice text content similarity higher than a similarity threshold value of the voice interaction information as a first similar interaction record;
classifying the first similar interaction records based on the terminal category;
sorting the control instruction data in each type of first similar interaction records according to the record number, and taking the preset number of control instruction data with the largest record number as candidate control instructions of the terminal type;
and determining the basic confidence scores of the candidate control instructions according to the proportion of the record quantity of each candidate control instruction to the record quantity of all the candidate control instructions.
Further, based on the candidate control instruction, determining the first confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the second voice interaction data set includes:
in the voice interaction records of the second voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a second similar interaction record;
based on the candidate control instructions, searching in the second similar interaction records, and determining a second record number of each candidate control instruction;
and determining a first confidence weight of the candidate control instruction according to the proportion of the second record quantity of the candidate control instruction to the sum of all the second record quantities.
Further, based on the candidate control instruction, determining the second confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the third voice interaction data set includes:
in the voice interaction records of the third voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a third voice interaction record;
searching a third similar interaction record based on the candidate control instructions, and determining a third record number of each candidate control instruction;
and determining a second confidence weight of the candidate control instruction according to the proportion of the third record quantity of the candidate control instruction to the sum of all the third record quantities.
Further, the calculating the confidence coefficient of the candidate control instruction according to the basic confidence coefficient score, the first confidence coefficient weight and the second confidence coefficient weight comprises:
the confidence level is equal to the product of the basic confidence level score and the first confidence level weight and the second confidence level weight.
Further, the method for determining the similarity of the voice text content comprises the following steps:
calculating cosine similarity between voice text contents;
respectively extracting terminal category keywords and control command keywords of the voice text content based on a pre-constructed terminal category keyword library and a control command keyword library;
calculating cosine similarity of terminal category keywords and cosine similarity of control command keywords;
the similarity of the two voice text contents is equal to the sum of the cosine similarity of the terminal category keywords multiplied by a first preset weight, the cosine similarity of the control command keywords multiplied by a second weight and the cosine similarity between the two voice text contents multiplied by a third preset weight.
Further, the method further comprises the following steps: determining a processing strategy of the expected control instruction based on the confidence;
when the confidence coefficient of the expected control instruction is larger than a first confidence coefficient threshold value, the processing strategy comprises executing the expected control instruction on the internet of things equipment of the terminal class designated by the plurality of internet of things equipment;
when the confidence level of the expected control instruction is not greater than the first confidence level threshold value and is greater than the second confidence level threshold value, the processing strategy comprises inquiring whether the Internet of things equipment of the specified terminal class of the plurality of Internet of things equipment executes the expected control instruction;
when the confidence of the expected control instruction is not greater than a second confidence threshold, the processing strategy includes not executing the expected control instruction.
In a second aspect, the present application provides a control device of an internet of things device. The control device is connected to a plurality of internet of things devices, the control device being configured for performing any of the methods as described in the first aspect above.
In a third aspect, the present application provides a control system for an internet of things device. The system comprises a control device as described in the second aspect above.
In summary, the present application at least comprises the following beneficial effects:
the control method, the control device and the control system for the Internet of things equipment are convenient to control in a fuzzy voice control mode, and the expected control instruction and the corresponding confidence level of voice interaction information are determined based on voice interaction record big data analysis, so that the control of the Internet of things equipment is more accurately achieved.
It should be understood that the description in this summary is not intended to limit key or critical features of embodiments of the present application, nor is it intended to be used to limit the scope of the present application. Other features of the present application will become apparent from the description that follows.
Drawings
The above and other features, advantages and aspects of embodiments of the present application will become more apparent by reference to the following detailed description when taken in conjunction with the accompanying drawings. In the drawings, wherein like or similar reference numerals denote like or similar elements, in which:
FIG. 1 illustrates a schematic diagram of an exemplary operating environment in which embodiments of the present application can be implemented;
fig. 2 shows a flowchart of a control method of an internet of things device in an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
In addition, the term "and/or" herein is merely an association relationship describing an association object, and means that three relationships may exist, for example, a and/or B may mean: a exists alone, A and B exist together, and B exists alone. In addition, the character "/" herein generally indicates that the front and rear associated objects are an "or" relationship.
The application provides a control method, equipment and a system of Internet of things equipment, which are used for analyzing voice interaction information by combining voice interaction record big data and determining an expected control instruction and confidence level of the voice interaction information, so that more accurate fuzzy voice control of the Internet of things equipment is facilitated.
In a first aspect, the present application provides a control method for an internet of things device.
FIG. 1 illustrates a schematic diagram of an exemplary operating environment in which embodiments of the present application can be implemented.
Referring to fig. 1, an operating environment includes a controller and a plurality of internet of things devices. The controller is used for sending a control instruction to the internet of things equipment so that the internet of things equipment executes corresponding actions, and the controller can be a local computer. The plurality of the internet of things devices are configured in the same application scene, the internet of things devices are intelligent terminal facilities, such as intelligent televisions, intelligent projection, intelligent air conditioners, intelligent lighting systems, intelligent fresh air systems and the like, voice acquisition modules can be configured for the internet of things devices, and a plurality of voice acquisition modules can be distributed in the actual application scene so as to realize voice control of the internet of things devices.
In the actual application process, a user sends out voice interaction information in an application scene, and the controller analyzes the voice interaction information to determine target equipment and target instructions of the voice interaction information, so that control over the equipment of the Internet of things is realized.
Fig. 2 shows a flowchart of a control method of an internet of things device in an embodiment of the present application. The method may be performed by the controller of fig. 1.
Referring to fig. 2, the method specifically includes the following steps.
S210: and acquiring the voice interaction record big data and voice interaction information.
The voice interaction information comprises voice text content and voice voiceprint identification. The voice interaction record is acquired by the voice acquisition module and is sent to the controller, the voice interaction information is used for controlling the Internet of things equipment, and it is understood that the voice interaction information is generally fuzzy voice information and does not contain explicit target equipment and target instructions, so that the target equipment and the target instructions contained in the voice interaction record can be determined after the Internet of things equipment is analyzed, and accurate control of the Internet of things equipment is realized.
The big data of the voice interaction record are deployed at the cloud, and the big data comprise the voice interaction records of a plurality of Internet of things devices in the application scene and the voice interaction records of the Internet of things devices in other application scenes. The voice interaction record in the voice interaction record big data comprises voice text content, voice voiceprint identification, terminal identification and control instruction data.
The terminal identification is the unique identification of the internet of things equipment, the terminal type of the internet of things equipment can be determined based on the terminal identification, the whole set of the terminal type is prestored in the cloud, the terminal type essentially reflects the type of intelligent terminal facilities, such as an intelligent television, intelligent projection, an intelligent air conditioner, an intelligent lighting system, an intelligent fresh air system and the like, and when the terminal identification is determined, the terminal type can be determined according to the terminal identification.
S220: and searching the voice interaction record big data based on the terminal identifiers of the plurality of internet of things devices to obtain a first voice interaction data set.
The plurality of internet of things devices in the embodiment of the application refer to a plurality of internet of things devices connected by a controller and belonging to an application scene in the embodiment of the application, each internet of things device comprises a terminal identifier, and a terminal category can be determined according to each terminal identifier, so that a set of terminal categories can be determined according to the terminal identifiers of the plurality of internet of things devices, and each internet of things device in the application scene in the embodiment of the application belongs to one terminal category in the set of terminal categories.
After the terminal category set is determined, searching can be performed in the voice interaction data set based on each terminal category in the set, and the terminal identification belonging to the terminal category is searched, so that the voice interaction record carrying the corresponding terminal identification is obtained. After the search of the terminal identifiers subordinate to all the terminal categories in the set is completed, the obtained set of all the voice interaction records is the first voice interaction data set. Namely, the terminal identification of the voice interaction record in the first voice interaction data set and one of the terminal identifications of the plurality of internet of things devices belong to the same terminal category.
S230: and searching the voice interaction record big data based on the voice voiceprint mark to obtain a second voice interaction data set.
The voice voiceprint identification reflects the identity of the user and the voice voiceprint identification of the voice interaction information reflects the identity of the user from which the voice interaction information originated. Because the voice interaction record in the voice interaction record big data also carries the voice voiceprint mark, the voice interaction record carrying the voice interaction information can be obtained by searching in the voice interaction record big data based on the voice voiceprint mark, the voice interaction records form a second voice interaction data set, and the voice interaction record in the second voice interaction data set is also sent by the user sending the voice interaction information. The terminal identification of the voice interaction record in the second voice interaction data set and the voice interaction information carry the same voice voiceprint identification.
S240: and searching the voice interaction record big data based on the terminal identifiers of the plurality of internet of things devices to obtain a third voice interaction data set.
The plurality of internet of things devices are in an application scene in the embodiment of the application and are connected with the controller in the embodiment of the application, the terminal identification of each internet of things device is unique and determined, and the voice interaction records can be searched in the voice interaction record big data based on the terminal identifications, so that the voice interaction records of the plurality of internet of things devices are obtained, and the voice interaction records form a third voice interaction data set. And the terminal identification of the voice interaction record in the third voice interaction data set is the same as one of the terminal identifications of the plurality of internet of things devices.
S250: and determining an expected control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set, the second voice interaction data set and the second voice interaction data set.
The confidence of the expected control instruction is carried, and the higher the confidence of the expected control instruction is, the more likely the voice interaction instruction is used for executing the expected control instruction.
In order to reasonably determine the expected control instructions and the confidence levels thereof, the method of the step comprises the following steps: determining a plurality of candidate control instructions and basic confidence scores of each candidate control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set; based on the candidate control instruction, determining a first confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the second voice interaction data set; based on the candidate control instruction, determining a second confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the third voice interaction data set; and calculating the confidence coefficient of the candidate control instruction according to the basic confidence coefficient score, the first confidence coefficient weight and the second confidence coefficient weight.
Specifically, the determining the plurality of candidate control instructions and the basic confidence score of each candidate control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set includes: in the voice interaction records of the first voice interaction data set, determining the voice interaction record with the voice text content similarity higher than a similarity threshold value of the voice interaction information as a first similar interaction record; classifying the first similar interaction records based on the terminal category; sorting the control instruction data in each type of first similar interaction records according to the record number, and taking the preset number of control instruction data with the largest record number as candidate control instructions of the terminal type; and determining the basic confidence scores of the candidate control instructions according to the proportion of the record quantity of each candidate control instruction to the record quantity of all the candidate control instructions.
In one example, the first similar records belonging to the same terminal category are divided into one category, the number of the control instruction data in each category of the first similar interaction records is counted, the control instruction data are ordered from high to low, and the preset number of the control instruction data in front is determined to be candidate control instructions. The preset number is typically 3-5.
Setting the number of the j candidate control instructions of the i terminal category in the first similar interaction record asThe terminal category is m, the preset number is n, and the basic confidence score of the candidate control instruction is +.>The calculation method of (1) is as follows:
based on the candidate control instruction, determining a first confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the second voice interaction data set comprises: in the voice interaction records of the second voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a second similar interaction record; based on the candidate control instructions, searching in the second similar interaction records, and determining a second record number of each candidate control instruction; and determining a first confidence weight of the candidate control instruction according to the proportion of the second record quantity of the candidate control instruction to the sum of all the second record quantities.
In one example, a second similar interaction record is providedThe number of the j candidate control instructions of the i terminal class isFirst confidence weight of candidate control instruction +.>The calculation method of (1) is as follows:
based on the candidate control instruction, determining a second confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the third voice interaction data set includes: in the voice interaction records of the third voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a third voice interaction record; searching a third similar interaction record based on the candidate control instructions, and determining a third record number of each candidate control instruction; and determining a second confidence weight of the candidate control instruction according to the proportion of the third record quantity of the candidate control instruction to the sum of all the third record quantities.
In one example, the number of the j candidate control commands of the i terminal class in the third interaction record is set to beSecond confidence weight of candidate control instruction +.>The calculation method of (1) is as follows:
the calculating the confidence coefficient of the candidate control instruction according to the basic confidence coefficient score, the first confidence coefficient weight and the second confidence coefficient weight comprises the following steps: the confidence level is equal to the product of the basic confidence level score and the first confidence level weight and the second confidence level weight.
In one example, let the number of j candidate control commands of the i terminal class beConfidence of candidate control instruction +.>The calculation method of (1) is as follows:
in the above, the method for determining the similarity of the voice text content comprises the following steps: calculating cosine similarity between voice text contents; respectively extracting terminal category keywords and control command keywords of the voice text content based on a pre-constructed terminal category keyword library and a control command keyword library; calculating cosine similarity of terminal category keywords and cosine similarity of control command keywords; the similarity of the two voice text contents is equal to the sum of the cosine similarity of the terminal category keywords multiplied by a first preset weight, the cosine similarity of the control command keywords multiplied by a second weight and the cosine similarity between the two voice text contents multiplied by a third preset weight. Here, the terminal category keywords and the control command keywords are both preset in the controller.
In the above, it is advantageous to makeThe method falls within a proper range, and the large probability falls between 0 and 1, so that the method is favorable for judging the confidence coefficient of the candidate control instruction, and is considered in combination with the terminal category keywords and the control command keywords when the similarity of the voice text content is considered, so that the method is favorable for ensuring that the judgment of the similarity is more reasonable, and avoiding the influence of invalid words as far as possible.
After determining the desired control instruction and the confidence level, the method further includes determining a processing strategy for the desired control instruction based on the confidence level. Specifically, when the confidence level of the expected control instruction is greater than a first confidence threshold, the processing policy includes executing the expected control instruction on the internet of things device of the specified terminal class of the plurality of internet of things devices; when the confidence level of the expected control instruction is not greater than the first confidence level threshold value and is greater than the second confidence level threshold value, the processing strategy comprises inquiring whether the Internet of things equipment of the specified terminal class of the plurality of Internet of things equipment executes the expected control instruction; when the confidence of the expected control instruction is not greater than a second confidence threshold, the processing strategy includes not executing the expected control instruction.
The first confidence coefficient threshold value is larger than the second confidence coefficient threshold value, and the first confidence coefficient threshold value and the second confidence coefficient threshold value can be set according to practical considerations. The query may be a voice query, such as "whether to execute XX instructions for XX device".
In summary, the method can reasonably infer the control instruction contained in the voice interaction information, and is beneficial to improving the accuracy of voice control under the condition of fuzzy voice control.
It should be noted that, for simplicity of description, the foregoing method embodiments are all expressed as a series of action combinations, but it should be understood by those skilled in the art that the present application is not limited by the described order of action, as some steps may be performed in other order or simultaneously according to the embodiments of the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all alternative embodiments, and that the acts and modules referred to are not necessarily required in the present application.
In a second aspect, the present application provides a control device of an internet of things device. The control device is connected to a plurality of internet of things devices, the control device being configured for performing any of the methods as described in the first aspect above.
In a third aspect, the present application provides a control system for an internet of things device. The system comprises a control device as described in the second aspect above.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working processes of the control apparatus and system described above may refer to corresponding processes in the foregoing method embodiments, which are not described herein again.
In summary, the present application at least comprises the following beneficial effects:
the control method, the control device and the control system for the Internet of things equipment are convenient to control in a fuzzy voice control mode, and the expected control instruction and the corresponding confidence level of voice interaction information are determined based on voice interaction record big data analysis, so that the control of the Internet of things equipment is more accurately achieved.
The foregoing description is only of the preferred embodiments of the present application and is presented as a description of the principles of the technology being utilized. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this application is not limited to the specific combinations of features described above, but it is intended to cover other embodiments in which any combination of features described above or equivalents thereof is possible without departing from the spirit of the disclosure. Such as the above-described features and technical features having similar functions (but not limited to) disclosed in the present application are replaced with each other.
Claims (10)
1. The control method of the Internet of things equipment is characterized by being applied to a controller, wherein the controller is connected with a plurality of Internet of things equipment, the controller is used for sending control instructions to the Internet of things equipment so that the Internet of things equipment executes corresponding actions, the plurality of Internet of things equipment is configured in the same application scene, and the Internet of things equipment is configured with a voice acquisition module or a plurality of voice acquisition modules distributed in the application scene so as to realize voice control of the Internet of things equipment;
the method comprises the following steps:
acquiring voice interaction record big data and voice interaction information, wherein the voice interaction information comprises voice text content and voice voiceprint identification, and the voice interaction record in the voice interaction record big data comprises the voice text content, the voice voiceprint identification, terminal identification and control instruction data;
retrieving the large voice interaction record data based on the terminal identifications of the plurality of internet of things devices to obtain a first voice interaction data set, wherein the terminal identifications of the voice interaction records in the first voice interaction data set and one of the terminal identifications of the plurality of internet of things devices belong to the same terminal category, and the voice interaction records in the first voice interaction data set are obtained by determining the terminal category of the terminal identifications of the plurality of internet of things devices and retrieving the voice interaction records in the voice interaction data set based on each terminal category; the voice interaction record big data is deployed at the cloud and comprises voice interaction records of a plurality of Internet of things devices in a plurality of application scenes;
searching the voice interaction record big data based on the voice voiceprint identifier to obtain a second voice interaction data set, wherein the terminal identifier of the voice interaction record in the second voice interaction data set and the voice interaction information carry the same voice voiceprint identifier;
retrieving the voice interaction record big data based on the terminal identifiers of the plurality of internet of things devices to obtain a third voice interaction data set, wherein the terminal identifiers of the voice interaction records in the third voice interaction data set are the same as one of the terminal identifiers of the plurality of internet of things devices;
and determining an expected control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction records in the first voice interaction data set, the second voice interaction data set and the second voice interaction data set, wherein the expected control instruction carries the confidence degree.
2. The method of claim 1, wherein determining the desired control instruction based on a degree of matching of the voice text content of the voice interaction information with the voice text content of the voice interaction records in the first voice interaction data set, the second voice interaction data set, and the second voice interaction data set comprises:
determining a plurality of candidate control instructions and basic confidence scores of each candidate control instruction according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the first voice interaction data set;
based on the candidate control instruction, determining a first confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the second voice interaction data set;
based on the candidate control instruction, determining a second confidence weight according to the matching degree of the voice text content of the voice interaction information and the voice text content of the voice interaction record in the third voice interaction data set;
and calculating the confidence coefficient of the candidate control instruction according to the basic confidence coefficient score, the first confidence coefficient weight and the second confidence coefficient weight.
3. The method of claim 2, wherein determining a plurality of candidate control commands and a base confidence score for each candidate control command based on a degree of matching of the voice text content of the voice interaction information to the voice text content of the voice interaction record in the first voice interaction dataset comprises:
in the voice interaction records of the first voice interaction data set, determining the voice interaction record with the voice text content similarity higher than a similarity threshold value of the voice interaction information as a first similar interaction record;
classifying the first similar interaction records based on the terminal category;
sorting the control instruction data in each type of first similar interaction records according to the record number, and taking the preset number of control instruction data with the largest record number as candidate control instructions of the terminal type;
and determining the basic confidence scores of the candidate control instructions according to the proportion of the record quantity of each candidate control instruction to the record quantity of all the candidate control instructions.
4. The method of claim 2, wherein determining the first confidence weight based on the candidate control instructions based on a degree of matching of the voice text content of the voice interaction information with the voice text content of the voice interaction record in the second voice interaction dataset comprises:
in the voice interaction records of the second voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a second similar interaction record;
based on the candidate control instructions, searching in the second similar interaction records, and determining a second record number of each candidate control instruction;
and determining a first confidence weight of the candidate control instruction according to the proportion of the second record quantity of the candidate control instruction to the sum of all the second record quantities.
5. The method of claim 2, wherein determining the second confidence weight based on the candidate control instructions based on a degree of matching of the voice text content of the voice interaction information with the voice text content of the voice interaction record in the third voice interaction dataset comprises:
in the voice interaction records of the third voice interaction data set, determining the voice interaction record with the voice text content similarity higher than the similarity threshold value of the voice interaction information as a third voice interaction record;
searching a third similar interaction record based on the candidate control instructions, and determining a third record number of each candidate control instruction;
and determining a second confidence weight of the candidate control instruction according to the proportion of the third record quantity of the candidate control instruction to the sum of all the third record quantities.
6. The method of claim 2, wherein calculating the confidence level of the candidate control command based on the base confidence score and the first and second confidence weights comprises:
the confidence level is equal to the product of the basic confidence level score and the first confidence level weight and the second confidence level weight.
7. The method of any of claims 3-6, wherein determining similarity of phonetic text content comprises:
calculating cosine similarity between voice text contents;
respectively extracting terminal category keywords and control command keywords of the voice text content based on a pre-constructed terminal category keyword library and a control command keyword library;
calculating cosine similarity of terminal category keywords and cosine similarity of control command keywords;
the similarity of the two voice text contents is equal to the sum of the cosine similarity of the terminal category keywords multiplied by a first preset weight, the cosine similarity of the control command keywords multiplied by a second weight and the cosine similarity between the two voice text contents multiplied by a third preset weight.
8. The method according to any one of claims 1-6, further comprising: determining a processing strategy of the expected control instruction based on the confidence;
when the confidence coefficient of the expected control instruction is larger than a first confidence coefficient threshold value, the processing strategy comprises executing the expected control instruction on the internet of things equipment of the terminal class designated by the plurality of internet of things equipment;
when the confidence level of the expected control instruction is not greater than the first confidence level threshold value and is greater than the second confidence level threshold value, the processing strategy comprises inquiring whether the Internet of things equipment of the specified terminal class of the plurality of Internet of things equipment executes the expected control instruction;
when the confidence of the expected control instruction is not greater than a second confidence threshold, the processing strategy includes not executing the expected control instruction.
9. The control equipment of the Internet of things equipment is characterized by being connected with a plurality of Internet of things equipment; the control device is configured to: for performing the method of any of claims 1-8.
10. A control system of an internet of things device, comprising the control device of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410147789.7A CN117672200B (en) | 2024-02-02 | 2024-02-02 | Control method, equipment and system of Internet of things equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410147789.7A CN117672200B (en) | 2024-02-02 | 2024-02-02 | Control method, equipment and system of Internet of things equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117672200A CN117672200A (en) | 2024-03-08 |
| CN117672200B true CN117672200B (en) | 2024-04-16 |
Family
ID=90068413
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410147789.7A Active CN117672200B (en) | 2024-02-02 | 2024-02-02 | Control method, equipment and system of Internet of things equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117672200B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009140884A1 (en) * | 2008-05-23 | 2009-11-26 | 深圳市北科瑞声科技有限公司 | A vehicle speech interactive system |
| KR20100066917A (en) * | 2008-12-10 | 2010-06-18 | 한국전자통신연구원 | Method for providing speech recognition in vehicle navigation system |
| CN104842869A (en) * | 2014-07-29 | 2015-08-19 | 北汽福田汽车股份有限公司 | Voice warning method and system of electric automobile and electric automobile |
| WO2019015435A1 (en) * | 2017-07-19 | 2019-01-24 | 腾讯科技（深圳）有限公司 | Speech recognition method and apparatus, and storage medium |
| CN115841821A (en) * | 2022-11-15 | 2023-03-24 | 浙江大学杭州国际科创中心 | Voice interference noise design method based on human voice structure |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7043438B2 (en) * | 2002-08-27 | 2006-05-09 | Kabushiki Kaisha Micronet | Moral standard protecting system in voice and/or text communication and interactive game system using therewith |
| CN111524521B (en) * | 2020-04-22 | 2023-08-08 | 北京小米松果电子有限公司 | Voiceprint extraction model training method, voiceprint recognition method, voiceprint extraction model training device and voiceprint recognition device |
| CN112836039B (en) * | 2021-01-27 | 2023-04-21 | 成都网安科技发展有限公司 | Voice data processing method and device based on deep learning |
-
2024
- 2024-02-02 CN CN202410147789.7A patent/CN117672200B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009140884A1 (en) * | 2008-05-23 | 2009-11-26 | 深圳市北科瑞声科技有限公司 | A vehicle speech interactive system |
| KR20100066917A (en) * | 2008-12-10 | 2010-06-18 | 한국전자통신연구원 | Method for providing speech recognition in vehicle navigation system |
| CN104842869A (en) * | 2014-07-29 | 2015-08-19 | 北汽福田汽车股份有限公司 | Voice warning method and system of electric automobile and electric automobile |
| WO2019015435A1 (en) * | 2017-07-19 | 2019-01-24 | 腾讯科技（深圳）有限公司 | Speech recognition method and apparatus, and storage medium |
| CN115841821A (en) * | 2022-11-15 | 2023-03-24 | 浙江大学杭州国际科创中心 | Voice interference noise design method based on human voice structure |
Non-Patent Citations (1)
| Title |
|---|
| 基于物联网的变电设备故障诊断研究;张超;孟梦;;自动化与仪器仪表;20180525(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117672200A (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106547744B (en) | Image retrieval method and system | |
| US8315430B2 (en) | Object recognition and database population for video indexing | |
| US6651057B1 (en) | Method and apparatus for score normalization for information retrieval applications | |
| CN112328891B (en) | Method for training search model, method for searching target object and device thereof | |
| CN109241319B (en) | Picture retrieval method, device, server and storage medium | |
| CN109977253B (en) | Semantic and content-based rapid image retrieval method and device | |
| CN112528022A (en) | Method for extracting characteristic words corresponding to theme categories and identifying text theme categories | |
| CN115455249A (en) | Double-engine driven multi-modal data retrieval method, equipment and system | |
| JP7341962B2 (en) | Learning data collection device, learning device, learning data collection method and program | |
| CN113704623A (en) | Data recommendation method, device, equipment and storage medium | |
| CN109977397B (en) | News hotspot extracting method, system and storage medium based on part-of-speech combination | |
| CN117672200B (en) | Control method, equipment and system of Internet of things equipment | |
| CN118013020B (en) | Patent query method and system for generating joint training based on retrieval | |
| CN114461783A (en) | Keyword generating method, apparatus, computer equipment, storage medium and product | |
| CN110209895B (en) | Vector retrieval method, device and equipment | |
| CN108170665B (en) | Keyword expansion method and device based on comprehensive similarity | |
| CN113792709B (en) | Rapid large-scale face recognition method and system | |
| CN110633430B (en) | Event discovery method, apparatus, device, and computer-readable storage medium | |
| CN117095426B (en) | A computational framework for cross-border person re-identification | |
| CN114187321A (en) | Target tracking method and device | |
| CN118897825B (en) | An intelligent retrieval method and system based on water affairs government data | |
| CN114756705B (en) | Image theme color retrieval method and device based on ES | |
| CN113793191B (en) | Commodity matching method and device and electronic equipment | |
| CN114860227B (en) | Facet-based component description and retrieval method, device and medium | |
| CN112214631B (en) | Method and system for re-identifying, retrieving and reordering vehicles guided by direction information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |