Consumption reminder method, system, storage medium and device based on edge computing

Obtaining and reporting consumption amounts through edge computing technology, solving the reminder problem of mobile payment in an Internet-free environment, realizing multilingual support and unified management, reducing implementation difficulty and system integration costs.

CN114049872BActive Publication Date: 2024-11-05SHENZHEN AEROSPACE SMART CITY SYST TECH RES INST CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111218493.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-10-20
Publication Date
2024-11-05
Estimated Expiration
2041-10-20

AI Technical Summary

Technical Problem

The existing mobile payment methods cannot provide consumption reminders without Internet environments. The consumption reminder service of private cloud deployment is high and difficult to implement. It requires one-to-one account binding for corresponding apps. The broadcast voice only supports Mandarin and cannot match the local dialect. And it cannot be managed uniformly.

Method used

Obtain consumption amount through edge calculation, generate Chinese text using Chinese semantic semantic order model, and output audio files through voice synthesis model for broadcasting, support Mandarin and local dialects, realize consumption amount reminders in the Internet environment, and manage them uniformly through edge calculation .

Benefits of technology

Implementing consumption amount reminders in an Internet environment reduces the difficulty of implementation, avoids third-party reviews, supports multi-language broadcasting, simplifies system integration, and realizes the connection and unified management of any consumer device.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114049872B_ABST
    Figure CN114049872B_ABST
Patent Text Reader

Abstract

The present invention discloses a consumption reminder method, system, storage medium and device based on edge computing, the method comprising: obtaining the consumption amount provided by the consumption device through a message queue service; translating the obtained consumption amount into Chinese text by calling a Chinese semantic word order model; associating and merging the Chinese text according to the initial consonants, finals or audio materials matched by the speech synthesis model, and outputting an audio file; decoding and outputting the synthesized audio file through an audio decoder; and broadcasting the decoded audio file using a broadcasting device. The present invention uses edge computing to integrate settlement server information, and uses edge computing to perform integrated services such as language conversion, speech synthesis and speech broadcasting. Any consumer device can be connected, and there is no difficulty in system integration.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of human-computer interaction technology, and in particular to a consumption reminder method, system, storage medium and device based on edge computing. Background Art

[0002] At present, the mobile payment industry is developing rapidly. The main technical solutions for consumption reminders of existing mobile payment methods are: Alipay's "QR code collection and payment arrival voice reminder" function, the principle of which is that after the server receives the amount of money received, the consumption amount audio is synthesized through the voice synthesis model built into the app, and finally it is broadcast through mobile devices (mobile phones, tablets, etc.); the other is the payment collection audio, which must apply for an aggregated payment code, push the mobile terminal payment information to a third-party platform, and the third-party platform will make a unified settlement. After the settlement is completed, the server synthesizes the consumption voice, and finally pushes it to the payment collection audio through the network for playback. The current consumption amount reminder device usually has the following problems when in use: it needs to be connected to the Internet or mobile terminal when in use, and consumption reminders cannot be made in an environment without the Internet; the cost of deploying consumption reminder services in a private cloud is too high, which is unaffordable for general units and cannot be used out of the box, and the implementation is difficult; when broadcasting, the corresponding app needs to be used for one-to-one account binding. If the same merchant broadcasts multiple stalls, multiple accounts need to be created; sensitive information such as consumption amount needs to be submitted to a third party for review; the broadcast voice only supports Mandarin and cannot match the local dialect; the consumption reminder device cannot be managed uniformly. Summary of the invention

[0003] In response to the above problems, the present invention proposes a consumption reminder method, system, storage medium and device based on edge computing. By obtaining the consumption amount in the Internet / private cloud / customized device, edge computing is used to generate Chinese text of the consumption amount, output the audio file, and decode the audio file for broadcast.

[0004] A first aspect of the present invention provides a consumption amount reminder method based on edge computing, the method comprising:

[0005] Obtain the consumption amount provided by the consumer device through the message queue service;

[0006] The obtained consumption amount is translated into Chinese text by calling the Chinese semantic word order model;

[0007] Associating and merging the Chinese text with the initial consonants, final consonants or audio materials matched by the speech synthesis model, and outputting an audio file;

[0008] Decoding and outputting the synthesized audio file through an audio decoder;

[0009] The decoded audio file is broadcasted using a broadcasting device.

[0010] Furthermore, the method for processing the consumption amount by the Chinese semantic word order model includes: the model receives the consumption amount in yuan;

[0011] Group the consumption amount by decimal point;

[0012] Reverse the digits before the decimal point;

[0013] The reversed numbers are split into arrays again;

[0014] Add units to the numbers in the array except the first one, and add units of ten, hundred, thousand, and ten thousand in order from the second one to get an array with units;

[0015] Convert each number in the array with units to Chinese uppercase;

[0016] Reverse the unit array converted to Chinese uppercase and obtain the Chinese uppercase array greater than zero;

[0017] Add a decimal point Chinese character at the end of the Chinese capital array to get the Chinese character array before the decimal point;

[0018] Split the numbers after the decimal point into an array;

[0019] Convert each digit in the array after the decimal point into Chinese to get the Chinese array after the decimal point;

[0020] Combine the Chinese array after the decimal point with the Chinese array before the decimal point, and add the Chinese array after the decimal point to the Chinese array before the decimal point to get the complete array of the consumption amount;

[0021] Add a Chinese word sequence to the end of the complete array of the consumption amount. If it does not contain a decimal, it ends with "round", otherwise it ends with "round";

[0022] Convert the complete array of consumption amount with Chinese word order added at the end to Chinese text.

[0023] Furthermore, the specific method of associating and merging the Chinese text with the initial consonants, finals or audio materials matched by the speech synthesis model and outputting the audio file includes:

[0024] Record 23 initials, 24 finals and corresponding tones according to the Chinese Pinyin, and number them in order;

[0025] Decompose the commonly used characters of consumption amount into corresponding pinyin, form key-value pairs and store them in hash table;

[0026] Traverse the Chinese text to match the corresponding pinyin and match the corresponding audio material according to the pinyin;

[0027] Sort the audio materials in order, adjust the speed and intonation according to the Chinese context, play and compress them, and synthesize the audio files.

[0028] In some embodiments, the announcement device is a speaker or a speaker.

[0029] A second aspect of the present invention provides a consumption reminder system based on edge computing, the system comprising:

[0030] Message receiving module: used to obtain the consumption amount provided by the consumption device through the message queue service;

[0031] Chinese text generation module: used to translate the acquired consumption amount into Chinese text by calling the Chinese semantic word order model;

[0032] Audio file output module: used to associate and merge the initial consonants, finals or audio materials matched by the speech synthesis model of the Chinese text, and output the audio file;

[0033] Audio file decoding module: used to decode and output the synthesized audio file through an audio decoder;

[0034] Broadcast module: used to broadcast the decoded audio file using the broadcast device.

[0035] According to a third aspect of the present invention, a storage medium is provided. The computer program stored in the storage medium can be executed by one or more processors and can be used to implement the consumption reminder method as described above.

[0036] According to a fourth aspect of the present invention, an electronic device is provided, comprising a memory and a processor, wherein a computer program is stored in the memory, and the memory and the processor are communicatively connected with each other, and when the computer program is executed by the processor, the consumption reminder method as described above is executed.

[0037] The present invention provides a consumption reminder method, system, storage medium and device based on edge computing, which uses edge computing to integrate settlement server information; uses edge computing to perform integrated services such as language conversion, speech synthesis and voice broadcasting; and uses edge computing technology for management. The beneficial effects finally achieved are: consumption amount reminders can be performed in an environment without Internet; there is no need to deploy speech synthesis services, language conversion services, etc., and the implementation difficulty is small; when the amount is broadcast, there is no need to use the corresponding app for one-to-one account binding; information such as consumption amount and subsidies does not need to be submitted to a third party for review, and sensitive information is not exposed; the broadcast voice supports Mandarin, supports voice resource replacement and intonation synthesis training, supports special language (Minnan dialect, Fujian dialect, etc.) training, and dialect localization; any consumption device can be connected, without system integration difficulties; the consumption amount reminder device can be uniformly managed through edge computing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flow chart of a consumption reminder method based on edge computing according to an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of the structure of a consumption reminder system based on edge computing according to an embodiment of the present invention;

[0040] Figure 3 is the architecture of the electronic device in the embodiment of the present invention; DETAILED DESCRIPTION

[0041] In order to further explain the technical solution of the present invention in detail, this embodiment is implemented on the premise of the technical solution of the present invention, and provides a detailed implementation method and specific steps.

[0042] The present invention obtains the consumption amount in the Internet / private cloud / customized device, generates Chinese text of the consumption amount, outputs the audio file, and decodes the audio file before broadcasting it. Figure 1 As shown, the specific steps include:

[0043] S1. Obtain the consumption amount provided by the consumer device through the message queue service. Edge computing uses the settlement server as an intermediary. The consumer device is linked to the settlement server and the consumption amount provided by the consumer device is obtained through the message queue service. The consumption methods of the consumer device include but are not limited to various payment operations such as scanning code payment, card payment, and face payment.

[0044] S2, translating the obtained consumption amount into Chinese text by calling the Chinese semantic word order model;

[0045] S3, associating and merging the Chinese text according to the initial consonants, finals or audio materials matched by the speech synthesis model, and outputting an audio file;

[0046] S4, decoding and outputting the synthesized audio file through an audio decoder;

[0047] S5. Use a broadcasting device to broadcast the decoded audio file.

[0048] The method for processing the consumption amount by the Chinese semantic word order model includes: the model receives the consumption amount in yuan;

[0049] Group the consumption amount by decimal point;

[0050] Reverse the digits before the decimal point;

[0051] The reversed numbers are split into arrays again;

[0052] Add units to the numbers in the array except the first one. Starting from the second number, add the units "Shi", "Bai", "Qian", and "Wan" in sequence to obtain an array with units.

[0053] Convert each number in the array with units to Chinese capital letters.

[0054] Reverse the array with units that have been converted to Chinese capital letters to obtain an array of Chinese capital letters greater than zero.

[0055] Add Chinese decimal point characters at the end of the array of Chinese capital letters to obtain an array of Chinese before the decimal point.

[0056] Split the numbers after the decimal point into an array.

[0057] Convert each digit in the array after the decimal point to Chinese to obtain an array of Chinese after the decimal point.

[0058] Merge the array of Chinese after the decimal point with the array of Chinese before the decimal point, and add the array of Chinese after the decimal point behind the array of Chinese before the decimal point to obtain a complete array of consumption amount.

[0059] Add a Chinese word order ending to the complete array of consumption amount. If there is no decimal part, end with "Yuan Zheng", otherwise end with "Yuan".

[0060] Convert the complete array of consumption amount with the Chinese word order ending added to Chinese text.

[0061] The embodiments are as follows:

[0062] S21. The model receives the consumption amount in yuan. For example, for a consumption of 12.5 yuan, the value passed into the model is 12.5.

[0063] S22. Group the value passed into the model according to the decimal point. For example, if 12.5 is passed in, it is converted into two groups: 12 and 5.

[0064] S23. Reverse the numbers before the decimal point. For example, 12 -> 21.

[0065] S24. Split the reversed number into an array. For example, 21 -> [2, 1].

[0066] S25. Add units to each digit in the array. Do not add anything to the first digit, add "Shi" to the second digit, add "Bai" to the third digit, and so on. For example, [2, 1] -> [2, 1 Shi].

[0067] S26. Convert each number in the array to capital letters. For example, [2, 1 Shi] -> [Er, Yi Shi].

[0068] S27. Reverse the array to obtain the Chinese greater than zero. For example, [Er, Yi Shi] -> [Yi Shi, Er].

[0069] S28. Add Chinese characters for decimal points, e.g., [Ten, Two] -> [Ten, Two, Point];

[0070] S29. Split the digits after the decimal point into an array, e.g., 5 -> [5];

[0071] S210. Convert each digit in the array to Chinese characters, e.g., [5] -> [Five];

[0072] S211. Combine the array with the previous part, e.g., [Ten, Two, Point] -> [Ten, Two, Point, Five];

[0073] S212. Add the Chinese word order ending to the array. If there is no decimal, end with "rounded up", otherwise end with "yuan", e.g., [Ten, Two, Point, Five] -> [Ten, Two, Point, Five, yuan];

[0074] S213. Convert the array to text, e.g., [Ten, Two, Point, Five, yuan] -> Twelve point five yuan; Thus, the process of converting 12.5 to Twelve point five yuan is completed.

[0075] The specific method of associating and combining Chinese text with the initials, finals, or audio materials matched by the speech synthesis model to output an audio file includes:

[0076] Record 23 initials, 24 finals, and their corresponding tones in accordance with Chinese pinyin, and number and arrange them in sequence;

[0077] Decompose the commonly used characters in the consumption amount into corresponding pinyin, form key-value pairs, and store them using a hash table;

[0078] Traverse the Chinese text to match the corresponding pinyin and then match the corresponding audio materials according to the pinyin;

[0079] Sort the audio materials in sequence, adjust the speed and intonation according to the Chinese context, play and compress them to synthesize an audio file.

[0080] In the specific implementation process, professionals record 23 initials, 24 finals and corresponding tones according to Chinese pinyin, and number and arrange them in sequence for easy searching during use, including but not limited to the initials and finals of Mandarin, and the word segmentation and word order of local dialects; disassemble the capital Chinese characters and units from 0 to 9, including Shi, Bai, Qian, Wan, and common words for consumption amounts such as Ling and Dian, into corresponding pinyin to form a key-value pair hash table for storage, which can speed up the processing speed, such as {key: "Er", value: "èr"}; traverse the Chinese text to match the corresponding pinyin and match the corresponding audio materials according to the pinyin; sort the material order, adjust the speed and intonation according to the general technical specifications of the Chinese text-to-speech synthesis system, play and compress to synthesize specific speech; decode and output the synthesized audio file through an audio decoder, and connect to a speaker / speaker through a connecting wire to achieve broadcasting.

[0081] This method solves the problem of performing subsidized consumption without Internet or mobile phone, including but not limited to schools, government agencies, enterprise canteens, and internal supermarkets, and the amount is reminded by artificial intelligence speech synthesis with local dialect.

[0082] Hereinafter, with reference to Figure 2 to describe the system corresponding to the method shown in Figure 1 A consumption reminder system 100 based on edge computing, the system 100 includes: a message receiving module 101: used to obtain the consumption amount provided by the consumption device through the message queue service; a Chinese text generation module 102: used to translate the obtained consumption amount into Chinese text by calling the Chinese semantic word order model; an audio file output module 103: used to associate and merge the Chinese text with the initials, finals or audio materials matched by the speech synthesis model, and output an audio file; an audio file decoding module 104: used to decode and output the synthesized audio file through an audio decoder; a broadcasting module 105: used to broadcast the decoded audio file by using a broadcasting device. In addition to these 5 modules, the system 100 may also include other components. However, since these components are not related to the content of the embodiments of the present disclosure, their illustrations and descriptions are omitted here.

[0083] The specific working process of a consumption reminder system 100 based on edge computing refers to the description of the above-mentioned edge computing-based consumption amount reminder method, and will not be repeated here.

[0084] In addition, the system of the embodiments of the present invention can also be implemented by means of Figure 3 the architecture of the electronic device shown in Figure 3 shows the architecture of the electronic device. As Figure 3As shown, it includes a communication component 201, an input / output component 202, a bus 203, a processor 204, a memory 205, an audio component 206, etc. The processor 204 executes the consumption reminder method as described above, the memory 205 can store various data or files used for electronic device processing and / or communication and program instructions executed by the processor, and the audio component 206 includes at least one speaker for outputting an audio signal. The received audio signal can be further stored in the memory 205 or sent through the communication component 201. Figure 3 The architecture shown is only exemplary and can be adjusted according to actual needs when implementing different electronic devices. Figure 3 One or more components in .

[0085] The embodiment of the present invention may also be implemented as a computer-readable storage medium. The computer-readable storage medium according to the embodiment stores computer-readable instructions. When the computer-readable instructions are executed by a processor, the consumption reminder method according to the embodiment of the present invention described with reference to the above figures may be executed.

[0086] The present invention provides a consumption reminder method, system, storage medium and device based on edge computing, which uses edge computing to integrate settlement server information; uses edge computing to perform integrated services such as language conversion, speech synthesis and voice broadcasting; and uses edge computing technology for management. The beneficial effects finally achieved are: consumption amount reminders can be performed in an environment without Internet; there is no need to deploy speech synthesis services, language conversion services, etc., and the implementation difficulty is small; when the amount is broadcast, there is no need to use the corresponding app for one-to-one account binding; information such as consumption amount and subsidies does not need to be submitted to a third party for review, and sensitive information is not exposed; the broadcast voice supports Mandarin, supports voice resource replacement and intonation synthesis training, supports special language (Minnan dialect, Fujian dialect, etc.) training, and dialect localization; any consumption device can be connected, without system integration difficulties; the consumption amount reminder device can be uniformly managed through edge computing technology.

[0087] In this document, the terms "comprises," "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, such that a step or method that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such step or method.

[0088] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A consumption reminder method based on edge computing, characterized in that: The method comprises: Obtain the consumption amount provided by the consumer device through the message queue service; The obtained consumption amount is translated into Chinese text by calling the Chinese semantic word order model; Associating and merging the Chinese text with the initial consonants, final consonants or audio materials matched by the speech synthesis model, and outputting an audio file; Decoding and outputting the synthesized audio file through an audio decoder; Using a broadcasting device to broadcast the decoded audio file; The Chinese semantic word order model processes the consumption amount in the following ways: The model receives the consumption amount in yuan; Group the consumption amount by decimal point; Reverse the digits before the decimal point; The reversed numbers are split into arrays again; Add units to the numbers in the array except the first one, and add units of ten, hundred, thousand, and ten thousand in order from the second one to get an array with units; Convert each number in the array with units to Chinese uppercase; Reverse the unit array converted to Chinese uppercase and obtain the Chinese uppercase array greater than zero; Add a decimal point Chinese character at the end of the Chinese capital array to get the Chinese character array before the decimal point; Split the numbers after the decimal point into an array; Convert each digit in the array after the decimal point into Chinese to get the Chinese array after the decimal point; Combine the Chinese array after the decimal point with the Chinese array before the decimal point, and add the Chinese array after the decimal point to the Chinese array before the decimal point to get the complete array of the consumption amount; Add a Chinese word sequence to the complete array of the consumption amount. If it does not contain a decimal, it ends with "round", otherwise it ends with "round"; Convert the complete array of consumption amount with Chinese word order added at the end to Chinese text.

2. The consumption reminder method based on edge computing according to claim 1 is characterized in that: The specific method of associating and merging the Chinese text according to the initial consonants, finals or audio materials matched by the speech synthesis model and outputting the audio file includes: Record 23 initials, 24 finals and corresponding tones according to the Chinese Pinyin, and number them in order; Decompose the commonly used characters of consumption amount into corresponding pinyin, form key-value pairs and store them in hash table; Traverse the Chinese text to match the corresponding pinyin and match the corresponding audio material according to the pinyin; Sort the audio materials in order, adjust the speed and intonation according to the general technical specifications of the Chinese speech synthesis system, play and compress them, and synthesize the audio files.

3. The consumption reminder method based on edge computing according to claim 1 is characterized in that: The broadcasting device is a speaker or a speaker.

4. A consumption reminder system based on edge computing, characterized in that: The system comprises: Message receiving module: used to obtain the consumption amount provided by the consumption device through the message queue service; Chinese text generation module: used to translate the acquired consumption amount into Chinese text by calling the Chinese semantic word order model; Audio file output module: used to associate and merge the initial consonants, finals or audio materials matched by the speech synthesis model of the Chinese text, and output the audio file; Audio file decoding module: used to decode and output the synthesized audio file through an audio decoder; Broadcast module: used to broadcast the decoded audio file using the broadcast device; The Chinese semantic word order model processes the consumption amount in the following ways: The model receives the consumption amount in yuan; Group the consumption amount by decimal point; Reverse the digits before the decimal point; The reversed numbers are split into arrays again; Add units to the numbers in the array except the first one, and add units of ten, hundred, thousand, and ten thousand in order from the second one to get an array with units; Convert each number in the array with units to Chinese uppercase; Reverse the unit array converted to Chinese uppercase and obtain the Chinese uppercase array greater than zero; Add a decimal point Chinese character at the end of the Chinese capital array to get the Chinese character array before the decimal point; Split the numbers after the decimal point into an array; Convert each digit in the array after the decimal point into Chinese to get the Chinese array after the decimal point; Combine the Chinese array after the decimal point with the Chinese array before the decimal point, and add the Chinese array after the decimal point to the Chinese array before the decimal point to get the complete array of the consumption amount; Add a Chinese word sequence to the end of the complete array of the consumption amount. If it does not contain a decimal, it ends with "round", otherwise it ends with "round"; Convert the complete array of consumption amount with Chinese word order added at the end to Chinese text.

5. A storage medium, characterized in that: The computer program stored in the storage medium can be processed by one or more processors The device can be executed to implement the consumption reminder method as described in any one of claims 1 to 3.

6. An electronic device, characterized in that: The invention comprises a memory and a processor, wherein the memory stores a computer program The computer program is executed by the processor, the memory and the processor are connected to each other in a communication manner. A consumption reminder method as described in any one of claims 1 to 3.