Washing machine data analysis system and method based on voice recognition

Abstract

The invention relates to the technical field of washing machines, in particular to a washing machine data analysis system and method based on voice recognition, and the method comprises the steps: obtaining user voice data, to-be-washed clothes feature data and historical washing record data; analyzing a clothes cleaning period and cleaning program instruction preference of the user based on the historical cleaning record data, and matching a corresponding voice control reference instruction in combination with the feature data of the clothes to be cleaned; recognizing a user voice control instruction based on the user voice data, and correcting a recognition result of the user voice control instruction in combination with the voice control reference instruction; and according to the corrected user voice control instruction, matching washing parameters and executing washing operation. According to the invention, intelligent correction of the voice instruction recognition result is realized by matching the voice control reference instruction, and the accuracy and reliability of voice control instruction recognition are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of washing machines, and in particular to a washing machine data analysis system and method based on voice recognition. BACKGROUND

[0002] With the continuous improvement of the intelligent level of household appliances, the application of voice control technology in washing machines has gradually become popular. Users can select the washing mode, adjust the water level, set the washing time, etc. through voice input, thereby improving the convenience of use. However, the existing voice recognition technology of washing machines still has deficiencies, mainly in the following aspects: (1) The voice recognition process of the existing washing machine mostly relies on the recognition result of a single voice input, which is easily affected by factors such as user accent, ambiguous pronunciation, environmental noise, etc., resulting in deviation of the voice control instruction recognition result, and even execution of the wrong washing program, affecting the user experience; (2) The existing technology usually fails to combine the user's historical cleaning record information, the specific characteristics of the clothes to be cleaned, and the user's long-term accumulated usage habits to intelligently correct or assist in judging the user's voice instructions. When there is a deviation between the voice recognition result and the user's true intention, the system often cannot effectively correct it, lacking adaptive and personalized instruction correction ability.

[0003] Therefore, there is an urgent need for a washing machine data analysis method that can intelligently correct user voice instructions by combining user historical cleaning behavior, clothing characteristics, and voice keyword similarity characteristics, to improve the accuracy and personalization of voice control. SUMMARY

[0004] In order to overcome the defects and deficiencies of the prior art, the present application provides a washing machine data analysis system and method based on voice recognition, which intelligently corrects the voice instruction recognition result by matching voice control reference instructions, effectively improving the accuracy and reliability of voice control instruction recognition.

[0005] In order to achieve the above purpose, the present application adopts the following technical solutions: In a first aspect, the present application provides a washing machine data analysis method based on voice recognition, comprising the following steps: Obtaining user voice data, clothing feature data to be cleaned, and historical cleaning record data; Analyzing the user's clothing cleaning cycle and cleaning program instruction preference based on the historical cleaning record data and matching the corresponding voice control reference instruction in combination with the clothing feature data to be cleaned; Recognizing the user's voice control instruction based on the user's voice data and correcting the recognition result of the user's voice control instruction in combination with the voice control reference instruction; Match the washing parameters and perform the washing operation according to the corrected user voice control instructions.

[0006] Optionally, the matching corresponding voice control reference instruction includes: Acquire historical cleaning record data and characteristic data of the clothes to be cleaned, wherein the historical cleaning record data includes the characteristics of the clothes to be cleaned, historical cleaning program control instructions, and corresponding historical cleaning timestamps; Determine the similarity of clothing characteristics between the clothing to be washed and the clothing historically washed and the weight of the clothing washing cycle respectively based on the historical washing record data and the characteristic data of the clothing to be washed; The product of the clothing feature similarity and the clothing washing cycle weight is used as the clothing washing matching degree, and the historical washing program control instructions corresponding to the historical washing clothes with the highest clothing washing matching degree are used as voice control reference instructions.

[0007] Optionally, determining the similarity of clothing features between the clothes to be washed and the clothes washed in history and the weight of the clothing washing cycles includes: Construct a feature vector of the clothes to be washed and a feature vector of the clothes to be washed according to the features of the clothes to be washed and the features of the clothes to be washed in history respectively; The cosine similarity between the feature vector of the clothes to be washed and the feature vector of the clothes washed in the past is used as the clothing feature similarity; Extract historical washing record data of clothing features whose similarity is greater than a preset clothing feature similarity threshold and construct a set of washing records of similar clothing; The clothing washing cycle is determined by the distribution of washing time in a set of washing records of similar clothing, and the clothing washing cycle weight is determined according to the clothing washing cycle.

[0008] Optionally, determining the laundry washing cycle weight according to the laundry washing cycle includes: Extract the historical cleaning timestamps corresponding to each historically cleaned garment in the same type of garment cleaning record set and determine the garment cleaning cycle. The calculation formula for the garment cleaning cycle is: ; In the formula Indicates the number of historical cleaning records in the set of cleaning records of the same type of clothing. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the laundry washing cycle; Get the current washing time of the clothes to be washed and determine the washing cycle weight of the clothes in combination with the washing cycle of the clothes. The calculation formula of the washing cycle weight of the clothes is: ; wherein represents the current washing time of the laundry to be washed, represents the historical washing timestamp of the last historical washing record in the same type of laundry washing record set, represents the weight of the laundry washing cycle.

[0009] Optionally, the step of correcting the recognition result of the user voice control instruction comprises: obtaining the user voice control instruction and the voice control reference instruction and extracting instruction keywords to construct a user instruction keyword set and a reference instruction keyword set, respectively; analyzing the keyword features in the user instruction keyword set and the reference instruction keyword set and determining a common instruction keyword set and a similar instruction keyword set; performing intersection operation on the common instruction keyword set and the similar instruction keyword set to obtain a corrected user instruction keyword set and convert it into a corrected user voice control instruction.

[0010] Optionally, the step of determining the common instruction keyword set and the similar instruction keyword set comprises: performing intersection operation on the user instruction keyword set and the reference instruction keyword set to obtain the common instruction keyword set; taking the complement of the common instruction keyword set in the user instruction keyword set as a first difference instruction keyword set and taking the complement of the common instruction keyword set in the reference instruction keyword set as a second difference instruction keyword set; determining the word vector similarity between each difference instruction keyword in the first difference instruction keyword set and the second difference instruction keyword set and constructing a similar instruction keyword set for the keywords with a word vector similarity less than a preset word vector similarity threshold.

[0011] In a second aspect, the present application provides a laundry machine data analysis system based on voice recognition, comprising: a data acquisition module for acquiring user voice data, laundry feature data to be washed and historical washing record data; a reference instruction matching module for analyzing the user's laundry washing cycle and washing program instruction preference based on the historical washing record data and matching the corresponding voice control reference instruction in combination with the laundry feature data to be washed; a control instruction correction module for recognizing the user voice control instruction based on the user voice data and correcting the recognition result of the user voice control instruction in combination with the voice control reference instruction; a washing operation execution module for matching the washing parameters according to the corrected user voice control instruction and executing the washing operation.

[0012] In a third aspect, the present application provides an electronic device, comprising a processor and a memory, wherein the memory stores a computer program that can be invoked by the processor, and the processor executes the voice recognition-based washing machine data analysis method by invoking the computer program stored in the memory.

[0013] In a fourth aspect, the present application provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to execute the voice recognition-based washing machine data analysis method.

[0014] Compared with the prior art, the present application has the following advantages and beneficial effects: The present application intelligently matches voice control reference instructions by combining user historical cleaning records, characteristics of clothes to be cleaned, and long-term clothes cleaning cycles of the user, and intelligently corrects the voice instruction recognition result by keyword similarity analysis on the user voice control instruction and the reference instruction, thereby significantly improving the accuracy and stability of voice control instruction recognition and further improving the user operation convenience. BRIEF DESCRIPTION OF DRAWINGS

[0015] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, made with reference to the accompanying drawings: Figure 1 is a schematic diagram of the overall process of the voice recognition-based washing machine data analysis method provided by the embodiments of the present application; Figure 2 is a schematic diagram of the process of determining the similarity of clothes characteristics and the weight of clothes cleaning cycles provided by the embodiments of the present application; Figure 3 is a schematic diagram of the structure of the voice recognition-based washing machine data analysis system provided by the embodiments of the present application; Figure 4 is a schematic diagram of the structure of the electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0016] The technical solutions of the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific features in the embodiments and the embodiments of the present application are detailed descriptions of the technical solutions of the present application, and are not limitations of the technical solutions of the present application. In the case of no conflict, the technical features in the embodiments and the embodiments of the present application can be combined with each other.

[0017] Please refer to Figure 1 , Figure 1 is a schematic diagram of the overall process of the voice recognition-based washing machine data analysis method provided by the embodiments of the present application, specifically comprising the following steps: S110: Obtain user voice data, to-be-cleaned clothes feature data, and historical cleaning record data.

[0018] S120: Analyze the clothes cleaning cycle and cleaning program instruction preference of the user based on the historical cleaning record data, and match the corresponding voice control reference instruction combined with the to-be-cleaned clothes feature data; By comparing and analyzing the historical cleaning record data of the user and the to-be-cleaned clothes feature data, the most similar historical cleaning record to the current to-be-cleaned clothes is first extracted, and the cleaning cycle information of similar clothes is combined to comprehensively determine the clothes feature similarity and the clothes cleaning cycle weight, and then the comprehensive clothes cleaning matching degree is calculated, so that the voice control reference instruction most close to the current cleaning scene is accurately matched from numerous historical cleaning records. The voice control reference instruction serves as a prediction basis for the user's real intention, and is compared and corrected with the initial control instruction input by the user through voice, which can significantly improve the fault tolerance and accuracy of instruction recognition, avoid the problem of incorrect washing program selection caused by voice recognition deviation or environmental noise, and ensure that the washing machine can better match the user's long-term use habits when intelligently performing washing operations. The corresponding voice control reference instruction includes: The historical cleaning record data and the to-be-cleaned clothes feature data are obtained, and the historical cleaning record data includes historical cleaning clothes features, historical cleaning program control instructions, and corresponding historical cleaning time stamps; The clothes feature similarity and the clothes cleaning cycle weight of the to-be-cleaned clothes and the historical cleaning clothes are determined respectively through the historical cleaning record data and the to-be-cleaned clothes feature data. The clothes feature similarity and the clothes cleaning cycle weight are used to comprehensively consider the consistency of the clothes features and the user's past cleaning habits, which helps to filter out the historical cleaning record that is most similar to the current to-be-cleaned clothes and meets the user's periodic cleaning rules, and improves the pertinence and accuracy of the matching result; The product of the clothes feature similarity and the clothes cleaning cycle weight is taken as the clothes cleaning matching degree, and the historical cleaning program control instruction corresponding to the historical cleaning clothes with the highest clothes cleaning matching degree is taken as the voice control reference instruction, which can ensure that the selected reference instruction is highly related to the current clothes type and meets the user's personalized cleaning habits, thereby providing more reliable and accurate semantic reference when correcting the user's voice control instruction in the future; Please refer to Figure 2 , Figure 2 is a flowchart provided by the embodiment of the present application for determining the clothes feature similarity and the clothes cleaning cycle weight. The clothes feature similarity and the clothes cleaning cycle weight of the to-be-cleaned clothes and the historical cleaning clothes are determined, including: The to-be-cleaned clothes feature vector and the historical cleaning clothes feature vector are constructed respectively through the to-be-cleaned clothes feature and the historical cleaning clothes feature; The cosine similarity between the feature vector of the laundry to be washed and the feature vector of the laundry previously washed is used as the laundry feature similarity. This can accurately determine the degree of feature similarity between the current laundry and the laundry previously washed in a multi-dimensional and quantitative manner, significantly improving the scientificity and objectivity of similarity judgment. Extract historical washing record data with clothing feature similarity greater than a preset clothing feature similarity threshold and construct a set of washing record data for similar clothing. Constructing a set of washing record data for similar clothing helps to eliminate interference samples with large feature differences and retain only valid reference samples that are highly similar to the clothing to be washed in terms of weight, material, color, stain degree, etc. The step of determining the preset clothing feature similarity threshold comprises: performing feature vector similarity calculation and analysis on historical clothing washing data of a large number of users, obtaining statistically the feature similarity distribution patterns of various types of clothing (such as cotton, silk, blended fabrics, and heavy clothing) under different washing programs, and selecting an empirical threshold in the similarity distribution based on the actual washing results of different clothing categories and user usage habits. For example, the 75th percentile in the similarity distribution is selected as the preset clothing feature similarity threshold; The washing cycle of clothes is determined by the distribution of washing time in a collection of washing records of similar clothes, and the washing cycle weight of clothes is determined based on the washing cycle of clothes. The washing cycle of clothes is used to reflect the user's periodic washing habits of similar clothes, and the washing cycle weight is then determined. The time factor is introduced into the subsequent washing matching degree calculation to further improve the accuracy of predicting voice control reference commands; The weight of the laundry washing cycle is determined according to the laundry washing cycle, including: Extract the historical washing timestamps corresponding to each historical washing item in the washing record set of the same type of clothing and determine the washing cycle of the clothing. The washing cycle of clothing is used to dynamically capture the user's personalized washing habits and provides a key reference in the time dimension for subsequent matching. The calculation formula for the washing cycle of clothing is: ; In the formula Indicates the number of historical cleaning records in the set of cleaning records of the same type of clothing. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the interval between two adjacent washes. It is used to take the arithmetic average of all intervals of the same type of clothing to capture the periodic pattern, avoiding the interference of single abnormal data (such as temporary emergency washing) in the overall cycle judgment. Indicates the laundry washing cycle; The current washing time of the clothes to be washed is obtained and combined with the washing cycle of the clothes to determine the washing cycle weight of the clothes. The washing cycle weight of the clothes can reflect the degree of compliance between the current washing time and the user's historical washing habits. By assigning different weights to different washing time points, it ensures that the recommended washing program is more in line with the user's actual needs and usage habits. The calculation formula of the washing cycle weight of clothes is: ; In the formula Indicates the current washing time of the clothes to be washed. Indicates the historical cleaning timestamp of the last historical cleaning record in the set of similar clothing cleaning records. Indicates the time difference between the current cleaning time and the last cleaning time. Used to convert the absolute time difference into a ratio relative to the clothing washing cycle, eliminating the dimension difference of different clothing washing cycles, using the exponential decay form The weight is described as non-linearly decaying as the deviation increases, that is, the control instructions that conform to the periodic law are adopted first. Indicates the laundry cycle weight of the laundry to be washed.

[0019] S130: Recognizing a user voice control instruction based on the user voice data and correcting a recognition result of the user voice control instruction in combination with a voice control reference instruction; The system obtains the user's current voice control command and the voice control reference command obtained based on historical matching, and decomposes the two into keyword sets. By comparing and analyzing the keyword features of the two, it determines the common command keyword set and the similar command keyword set obtained by word vector similarity calculation. Then, it performs an intersection operation on the common keywords and similar keywords to generate a revised command set that is closer to the user's true intention, and outputs the revised user voice control command accordingly. The correction process not only combines the user's usage habits and scenario characteristics, but also makes full use of the semantic verification information provided by the reference command, realizing automatic and intelligent secondary proofreading of the voice command, significantly improving the accuracy and reliability of command execution, and correcting the recognition results of the user's voice control command, including: Obtaining user voice control instructions and voice control reference instructions and extracting instruction keywords to respectively construct a user instruction keyword set and a reference instruction keyword set, wherein the step of obtaining the user voice control instructions includes: using a speech recognition engine (such as an ASR system based on an acoustic model, a language model, and a decoding algorithm) to preprocess, extract features, and match acoustic units on user voice data, transcribe the speech into corresponding text instructions, and perform word segmentation, intent recognition, and command parsing on the recognized text in combination with a preset instruction vocabulary or a semantic parsing model (NLP), thereby outputting user voice control instructions corresponding to the speech content; analyze the keyword features in the user instruction keyword set and the reference instruction keyword set, and determine a common instruction keyword set and a similar instruction keyword set, the common instruction keyword set and the similar instruction keyword set being used to describe the core consistent content between the user voice instruction and the reference instruction and the similar but different keywords existing; perform an intersection operation on the common instruction keyword set and the similar instruction keyword set to obtain a modified user instruction keyword set and convert it into a modified user voice control instruction, the intersection operation being used to comprehensively integrate the core information of the user original instruction and the supplementary content of the reference instruction, effectively correct the recognition errors and ambiguities, improve the accuracy of voice recognition and the integrity of instruction understanding, and ensure that the final executed washing control instruction is more consistent with the real intention of the user; determining the common instruction keyword set and the similar instruction keyword set includes: perform an intersection operation on the user instruction keyword set and the reference instruction keyword set to obtain a common instruction keyword set, the common instruction keyword set being used to extract the key information commonly contained in the user voice instruction and the reference instruction and determine the basic content of the user's real intention; the complement of the common instruction keyword set in the user instruction keyword set is taken as a first difference instruction keyword set, and the complement of the common instruction keyword set in the reference instruction keyword set is taken as a second difference instruction keyword set, aiming to identify the keyword parts that are different between the user instruction and the reference instruction and provide targets for further semantic similarity analysis, and effectively distinguish the special words expressed by the user and possible recognition errors; determine the word vector similarity between the difference instruction keywords in the first difference instruction keyword set and the second difference instruction keyword set, and construct a similar instruction keyword set for the keywords with a word vector similarity less than a preset word vector similarity threshold, to realize intelligent matching and identification of keywords with similar semantics but different expressions, wherein the step of determining the preset word vector similarity threshold includes: constructing a standard word vector library of washing machine control instructions based on domain knowledge, and statistically analyzing the similarity distribution of correct recognition and misrecognition word pairs through large-scale voice recognition test data, combining voice recognition error mode analysis, and selecting a similarity value that can cover more than 85% of typical misrecognition cases as the preset word vector similarity threshold.

[0020] S140: match the washing parameters according to the modified user voice control instruction and perform the washing operation; After the user voice control instruction is corrected, the core operation intention and parameter keywords such as the washing mode, temperature, rotating speed, time, and whether pre-washing are extracted according to the corrected instruction content, and are matched with the pre-set washing parameter database; according to the matching result, the specific washing program configuration such as the water level, washing time length, rinsing times, dehydration rotating speed, and water temperature is determined, and a complete washing control instruction set is generated; the control module executes the corresponding washing operation process according to the matched washing parameters, so as to ensure that the washing process is consistent with the user intention Figure 1 , and improve the washing effect and use convenience.

[0021] Please refer to Figure 3 , Figure 3 is a structural schematic diagram of a washing machine data analysis system based on voice recognition provided by the embodiment of the application. The embodiment provides a washing machine data analysis system based on voice recognition, which comprises: a data acquisition module 210, configured to acquire user voice data, to-be-cleaned clothes feature data, and historical cleaning record data; a reference instruction matching module 220, configured to analyze the clothes cleaning period and cleaning program instruction preference of a user based on the historical cleaning record data, and match corresponding voice control reference instructions in combination with the to-be-cleaned clothes feature data; a control instruction correction module 230, configured to recognize a user voice control instruction based on the user voice data, and correct the recognition result of the user voice control instruction in combination with the voice control reference instruction; a washing operation execution module 240, configured to match washing parameters and execute a washing operation according to the corrected user voice control instruction.

[0022] In the embodiment of the application, the reference instruction matching module 220 is configured to analyze the clothes cleaning period and cleaning program instruction preference of a user based on the historical cleaning record data, and match corresponding voice control reference instructions in combination with the to-be-cleaned clothes feature data, comprising: acquiring historical cleaning record data and to-be-cleaned clothes feature data, the historical cleaning record data containing historical cleaning clothes features, historical cleaning program control instructions, and corresponding historical cleaning time stamps; determining the clothes feature similarity and clothes cleaning period weight of to-be-cleaned clothes and historical cleaning clothes based on the historical cleaning record data and the to-be-cleaned clothes feature data, comprising: constructing a to-be-cleaned clothes feature vector and a historical cleaning clothes feature vector based on the to-be-cleaned clothes feature and the historical cleaning clothes feature, respectively; taking the cosine similarity of the to-be-cleaned clothes feature vector and the historical cleaning clothes feature vector as the clothes feature similarity; extract the historical cleaning record data with a clothes feature similarity greater than a preset clothes feature similarity threshold and construct a same-type clothes cleaning record set; determine a clothes cleaning cycle through the cleaning time distribution in the same-type clothes cleaning record set and determine a clothes cleaning cycle weight according to the clothes cleaning cycle; multiply the clothes feature similarity and the clothes cleaning cycle weight to obtain a clothes cleaning matching degree and take the historical cleaning program control instruction corresponding to the historical cleaning clothes with the highest clothes cleaning matching degree as the voice control reference instruction.

[0023] In the embodiments of the present application, the control instruction modification module 230 is configured to identify the user voice control instruction based on the user voice data and modify the identification result of the user voice control instruction in combination with the voice control reference instruction, including: obtain the user voice control instruction and the voice control reference instruction and extract instruction keywords to construct a user instruction keyword set and a reference instruction keyword set, including: perform an intersection operation on the user instruction keyword set and the reference instruction keyword set to obtain a common instruction keyword set; take the complement of the common instruction keyword set in the user instruction keyword set as a first difference instruction keyword set and take the complement of the common instruction keyword set in the reference instruction keyword set as a second difference instruction keyword set; determine the word vector similarity between each difference instruction keyword in the first difference instruction keyword set and the second difference instruction keyword set and construct a similar instruction keyword set for the keywords with a word vector similarity less than a preset word vector similarity threshold; analyze the keyword features in the user instruction keyword set and the reference instruction keyword set and determine the common instruction keyword set and the similar instruction keyword set; perform an intersection operation on the common instruction keyword set and the similar instruction keyword set to obtain a modified user instruction keyword set and convert it to a modified user voice control instruction.

[0024] The steps of implementing the functions of each parameter and each unit module in the voice recognition-based washing machine data analysis system of the present application can refer to the parameters and steps in the embodiments of the voice recognition-based washing machine data analysis method described above, and will not be repeated here.

[0025] Please refer to Figure 4Embodiments of the present application also provide an electronic device 300, comprising a memory 310, a processor 320 and a communication bus 330; the memory 310 and the processor 320 are connected through the communication bus 330. The memory 310 stores instructions that can be loaded and executed by the processor 320 to implement the voice recognition-based washing machine data analysis method provided by the above embodiments.

[0026] The memory 310 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 310 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function, and instructions for implementing the voice recognition-based washing machine data analysis method provided by the above embodiments, etc.; the data storage area can store data involved in the voice recognition-based washing machine data analysis method provided by the above embodiments, etc.

[0027] The processor 320 can include one or more processing cores. The processor 320 executes various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 310, calling data stored in the memory 310. The processor 320 can be at least one of an Application Specific Integrated Circuit (ASIC), a Digital Signal Processor (DSP), a Digital Signal Processing Device (DSPD), a Programmable Logic Device (PLD), a Field Programmable Gate Array (FPGA), a Central Processing Unit (CPU), a controller, a microcontroller and a microprocessor. It can be understood that for different devices, the electronic devices used to implement the functions of the processor 320 described above can also be other devices, and the embodiments of the present application are not limited specifically.

[0028] The communication bus 330 can include a channel for transmitting information between the above components. The communication bus 330 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus 330 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4Only one bus or type of bus can be present. However, busses can be implemented using different technologies, such as optical or electrical technologies.

[0029] The embodiment of the present application provides a computer readable storage medium, which stores a computer program capable of being loaded and executed by a processor to execute the voice recognition based washing machine data analysis method provided by the above embodiment.

[0030] In the embodiment of the present application, the computer readable storage medium can be a tangible device that keeps and stores instructions for use by an instruction execution device. The computer readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device or any combination of the above. Specifically, the computer readable storage medium can be a portable computer disk, a hard disk, a U disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a platform random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, an optical disk, a magnetic disk, a mechanical coding device and any combination of the above.

[0031] The term "comprising" or "including" or any other variant thereof is intended to cover non-exclusive inclusions, so that a process, method, article or apparatus including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or apparatus.

[0032] The above description is merely preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or equivalent features without departing from the above application concept. For example, the above features are replaced with technical features with similar functions applied in the present application (but not limited to) to form technical solutions.

Claims

1. A washing machine data analysis method based on speech recognition, characterized in that: The steps include: Obtain user voice data, characteristic data of the clothes to be washed, and historical washing record data; Analyze the user's clothing washing cycle and washing program instruction preferences based on historical washing record data and match the corresponding voice control reference instructions based on the characteristic data of the clothes to be washed; Recognizing a user voice control instruction based on the user voice data and correcting a recognition result of the user voice control instruction in combination with a voice control reference instruction; Match the washing parameters and perform the washing operation according to the corrected user voice control instructions.

2. The method for analyzing washing machine data based on speech recognition according to claim 1, characterized in that: The matching corresponding voice control reference instruction includes: Acquire historical cleaning record data and characteristic data of the clothes to be cleaned, wherein the historical cleaning record data includes the characteristics of the clothes to be cleaned, historical cleaning program control instructions, and corresponding historical cleaning timestamps; Determine the similarity of clothing characteristics between the clothing to be washed and the clothing historically washed and the weight of the clothing washing cycle respectively based on the historical washing record data and the characteristic data of the clothing to be washed; The product of the clothing feature similarity and the clothing washing cycle weight is used as the clothing washing matching degree, and the historical washing program control instructions corresponding to the historical washing clothes with the highest clothing washing matching degree are used as voice control reference instructions.

3. The method for analyzing washing machine data based on speech recognition according to claim 2, characterized in that: The determining of the similarity of clothing features between the clothing to be washed and the clothing washed in history and the clothing washing cycle weight includes: Construct a feature vector of the clothes to be washed and a feature vector of the clothes to be washed according to the features of the clothes to be washed and the features of the clothes to be washed in history respectively; The cosine similarity between the feature vector of the clothes to be washed and the feature vector of the clothes washed in the past is used as the clothing feature similarity; Extract historical washing record data of clothing features whose similarity is greater than a preset clothing feature similarity threshold and construct a set of washing records of similar clothing; The clothing washing cycle is determined by the distribution of washing time in a set of washing records of similar clothing, and the clothing washing cycle weight is determined according to the clothing washing cycle.

4. The method for analyzing washing machine data based on speech recognition according to claim 3, characterized in that: Determining the weight of the laundry washing cycle according to the laundry washing cycle includes: Extract the historical cleaning timestamps corresponding to each historically cleaned garment in the same type of garment cleaning record set and determine the garment cleaning cycle. The calculation formula for the garment cleaning cycle is: ； In the formula Indicates the number of historical cleaning records in the set of cleaning records of the same type of clothing. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the The historical cleaning timestamp of the historical cleaning record. Indicates the laundry washing cycle; Get the current washing time of the clothes to be washed and determine the washing cycle weight of the clothes in combination with the washing cycle of the clothes. The calculation formula of the washing cycle weight of the clothes is: ； In the formula Indicates the current washing time of the clothes to be washed. Indicates the historical cleaning timestamp of the last historical cleaning record in the set of similar clothing cleaning records. Indicates the weight of the laundry washing cycle.

5. The method for analyzing washing machine data based on speech recognition according to claim 1, characterized in that: The correcting of the recognition result of the user's voice control instruction includes: Obtain user voice control commands and voice control reference commands and extract command keywords to construct a user command keyword set and a reference command keyword set respectively; Analyze keyword features in the user instruction keyword set and the reference instruction keyword set and determine a common instruction keyword set and a similar instruction keyword set; An intersection operation is performed on the common instruction keyword set and the similar instruction keyword set to obtain a modified user instruction keyword set and convert it into a modified user voice control instruction.

6. The method for analyzing washing machine data based on speech recognition according to claim 5, characterized in that: The determining of the common instruction keyword set and the similar instruction keyword set includes: Perform an intersection operation on the user instruction keyword set and the reference instruction keyword set to obtain a common instruction keyword set; The complement of the common instruction keyword set in the user instruction keyword set is used as the first difference instruction keyword set, and the complement of the common instruction keyword set in the reference instruction keyword set is used as the second difference instruction keyword set; Determine the word vector similarity between each difference instruction keyword in the first difference instruction keyword set and the second difference instruction keyword set, and construct a similar instruction keyword set from keywords whose word vector similarity is less than a preset word vector similarity threshold.

7. A washing machine data analysis system based on speech recognition, applied to the washing machine data analysis method based on speech recognition according to any one of claims 1 to 6, characterized in that: The system comprises: The data acquisition module is used to obtain user voice data, characteristic data of the clothes to be washed, and historical washing record data; A reference instruction matching module is used to analyze the user's clothing washing cycle and washing program instruction preferences based on historical washing record data and match corresponding voice control reference instructions based on the characteristic data of the clothes to be washed; A control instruction correction module, configured to identify a user voice control instruction based on the user voice data and correct the recognition result of the user voice control instruction in combination with a voice control reference instruction; The washing operation execution module is used to match the washing parameters and execute the washing operation according to the corrected user voice control instruction.

8. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; characterized in that the processor executes the washing machine data analysis method based on voice recognition as described in any one of claims 1 to 6 by calling the computer program stored in the memory.

9. A computer-readable storage medium, characterized in that Instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the washing machine data analysis method based on voice recognition according to any one of claims 1 to 6.

Images