[go: up one dir, main page]

CN114155828A - Electric charge fee-prompting method and system based on intelligent voice - Google Patents

Electric charge fee-prompting method and system based on intelligent voice Download PDF

Info

Publication number
CN114155828A
CN114155828A CN202111318798.0A CN202111318798A CN114155828A CN 114155828 A CN114155828 A CN 114155828A CN 202111318798 A CN202111318798 A CN 202111318798A CN 114155828 A CN114155828 A CN 114155828A
Authority
CN
China
Prior art keywords
information
voice
database
cache unit
reminder
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.)
Granted
Application number
CN202111318798.0A
Other languages
Chinese (zh)
Other versions
CN114155828B (en
Inventor
赵斌
陈志伟
李超
邱峥
林启超
庄江颖
陈沧杰
魏林
陈冰勇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
HuiAn Power Supply Co of State Grid Fujian Electric Power Co Ltd
Original Assignee
HuiAn Power Supply Co of State Grid Fujian Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by HuiAn Power Supply Co of State Grid Fujian Electric Power Co Ltd filed Critical HuiAn Power Supply Co of State Grid Fujian Electric Power Co Ltd
Priority to CN202111318798.0A priority Critical patent/CN114155828B/en
Publication of CN114155828A publication Critical patent/CN114155828A/en
Application granted granted Critical
Publication of CN114155828B publication Critical patent/CN114155828B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/02—Methods for producing synthetic speech; Speech synthesisers
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
    • G06F16/63—Querying
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/546—Message passing systems or structures, e.g. queues
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • 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
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
    • G10L25/24—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being the cepstrum
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/12—Messaging; Mailboxes; Announcements
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00—Indexing scheme relating to G06F9/00
    • G06F2209/54—Indexing scheme relating to G06F9/54
    • G06F2209/548—Queue

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Acoustics & Sound (AREA)
  • Human Computer Interaction (AREA)
  • General Physics & Mathematics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Water Supply & Treatment (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to an electric charge fee-prompting method and system based on intelligent voice, wherein the method comprises the following steps: establishing connection between a server and a user side through WebSocket; establishing a cache unit and a database for storing voice information at a server; acquiring electric charge bill information, and searching whether voice information corresponding to the current electric charge bill information exists in a cache unit, if so, pushing the voice information to a message queue, and if not, searching whether voice information corresponding to the current electric charge bill information exists in a database; if the voice information exists in the database, the voice information is pushed to the message queue, if the voice information does not exist in the database, new voice information is synthesized according to the electric bill information, and the voice information is pushed to the message queue after being stored in the database and the cache unit; monitoring whether the message queue has voice messages, if so, pushing the voice messages to the user side of the corresponding user through WebSocket, and generating a fee prompting record.

Description

Electric charge fee-prompting method and system based on intelligent voice
Technical Field
The invention relates to an electric charge fee-prompting method and system based on intelligent voice, and belongs to the technical field of intelligent fee prompting.
Background
After the intelligent voice industry in China is subjected to transient structure adjustment, capacity after falling is eliminated, unqualified enterprises are screened out, and with the transformation of the consumption concept of residents and the improvement of consumption requirements, the intelligent voice industry in China still keeps the growth trend, and will develop towards high quality and high quality in the future, and present new trends of variety increase, consumption diversification and the like. Nowadays, digitalization and intellectualization are rapidly popularized, electric power resources are basic resources of daily work and life of people, and the labor-consuming work of electric charges is inevitably developed towards intellectualization. If urge the expense with intelligent pronunciation applied to the charges of electricity, urge the expense intellectuality with electric power, just can promote the real-time of obtaining the information for the user, also can promote and urge the expense efficiency, reduce the cost of manpower and materials.
The patent with the application number of 200910073824.0 discloses an electric power short message voice charging service system, which can provide two communication modes of mobile phone short messages and telephone voice for an electric power supply department, can automatically operate and maintain data, can improve the service quality and the humanized service level of the electric power department, improve the charging efficiency, and is quick and accurate in work requirement; the daily operation of the system by the staff is reduced, and the manual payment urging strength of the toll collector is reduced; the department work notice and the power failure maintenance notice can be issued through a short message function; the electric power short message voice charging service system can supervise the system application condition and the telephone number maintenance condition of each company through the data query statistical function.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an electric charge fee-prompting method and system based on intelligent voice, wherein a user side is connected with a server side through a WebSocket, so that the user can be automatically reminded to pay the electric charge fee, the problem of untimely fee prompting is solved, the user can know the own electricity utilization condition through a visual voice form, and the user can conveniently pay the electric charge fee; on the other hand can alleviate electric power staff's the work load that the charges of electricity urge to pay and retrieve, realizes the intellectuality of working method, has the advantage of automaticity and high efficiency.
The technical scheme of the invention is as follows:
an electric charge fee-prompting method based on intelligent voice comprises the following steps:
establishing connection between a server and a user side through WebSocket;
establishing a cache unit and a database for storing voice information at a server;
acquiring electric charge bill information, and searching whether voice information corresponding to the current electric charge bill information exists in a cache unit, if so, pushing the voice information to a message queue, and if not, searching whether voice information corresponding to the current electric charge bill information exists in a database;
if the voice information exists in the database, the voice information is pushed to the message queue, if the voice information does not exist in the database, new voice information is synthesized according to the electric bill information, and the voice information is pushed to the message queue after being stored in the database and the cache unit;
monitoring whether the message queue has voice messages, if so, pushing the voice messages to the user side of the corresponding user through WebSocket, and generating a fee prompting record.
As a preferred embodiment, the method for synthesizing new voice information according to the electric bill information specifically includes:
preprocessing, namely generating text data from the electric bill information through a voice synthesis template, and converting the text data into a pinyin representation form;
inputting the preprocessed text data into a pre-trained acoustic feature prediction model, outputting corresponding acoustic features, converting the acoustic features into voice audio through a vocoder, and coding the voice audio to generate the voice information.
As a preferred embodiment, the method further comprises the step of automatically generating the language information at regular time and pushing the charge information at regular time, wherein the step of automatically generating the language information at regular time specifically comprises the following steps:
establishing a timing voice synthesis task, and configuring a time period of the timing voice synthesis task and a corresponding user;
when the time period is reached, starting a timing voice synthesis task;
selecting a voice synthesis template, automatically acquiring the electric bill information generated by the corresponding user in a time period, generating voice information according to the selected voice synthesis template, and storing the voice information into a database and a cache unit;
the timed charge urging task comprises the following steps:
establishing a timing pushing charge-urging task, and configuring a time period of the timing pushing charge-urging task and a corresponding user;
and when the time period is reached, starting a timed pushing charge urging task, inquiring whether voice information of a corresponding user exists in the cache unit and the database, and if so, pushing the voice information to a message queue.
As a preferred embodiment, the database stores at least one speech synthesis template, and the method for establishing the speech synthesis template specifically includes:
inputting newly added template information;
checking whether the newly added template information is legal or not according to a preset rule;
if the information is legal, the information is stored in the database, and if the information is illegal, error information is returned.
As a preferred embodiment, the method further comprises a step for pushing the announcement to all users, which is specifically as follows:
acquiring announcement information;
inquiring whether corresponding notice information exists in the cache unit, if so, pushing the notice information to a message queue, and if not, inquiring whether corresponding notice information exists in a database;
if the database exists, storing the notice information into a cache unit, pushing the notice information to a message queue, and if the database does not exist, returning error information;
monitoring whether the announcement information exists in the message queue or not, and if so, pushing the announcement information to the user sides of all users through WebSocket.
On the other hand, the invention provides an electric charge fee-prompting system based on intelligent voice, which comprises:
the server is provided with a cache unit for storing voice information and a database;
the user side is used for receiving voice information;
the communication module is used for establishing the connection between the server and the user through WebSocket;
the bill management module is used for managing the electric bill information;
the voice synthesis module is used for synthesizing voice information according to the electric bill information;
the fee urging module is used for acquiring the electric bill information and inquiring whether the voice information corresponding to the electric bill information exists in the cache unit and the database; if the voice information exists, outputting the corresponding voice information to a message queue, and if the voice information does not exist, calling a voice synthesis module to synthesize the corresponding voice information and storing the corresponding voice information in a cache unit and a database;
the first monitoring module is used for monitoring whether the voice message exists in the message queue or not, if so, the voice message is pushed to a user side of a corresponding user through the Websocket, and a fee urging record is generated.
As a preferred embodiment, the speech synthesis module comprises:
the preprocessing unit is used for preprocessing, generating text data from the electric bill information through a voice synthesis template, and converting the text data into a pinyin representation form;
the acoustic feature prediction model is used for inputting the preprocessed text data and outputting corresponding acoustic features;
and the synthesis unit is used for converting the acoustic characteristics into voice audio through a vocoder, and carrying out coding processing on the voice audio to generate the voice information.
As a preferred embodiment, the system also comprises a timing automatic generation voice information module and a timing pushing charge prompting information module; wherein the timing automatic generation voice information module is used for:
establishing a timing voice synthesis task, and configuring a time period of the timing voice synthesis task and a corresponding user;
when the time period is reached, starting a timing voice synthesis task;
selecting a voice synthesis template, automatically acquiring the electric bill information generated by the corresponding user in a time period, generating voice information according to the selected voice synthesis template, and storing the voice information into a database and a cache unit;
the timing pushing fee-urging information module is used for:
establishing a timing pushing charge-urging task, and configuring a time period of the timing pushing charge-urging task and a corresponding user;
and when the time period is reached, starting a timed pushing charge urging task, inquiring whether voice information of a corresponding user exists in the cache unit and the database, and if so, pushing the voice information to a message queue.
As a preferred embodiment, the method further comprises a speech synthesis template establishing module, including:
the input unit is used for inputting newly added template information;
the checking unit is used for checking whether the newly added template information is legal or not according to a preset rule; if the information is legal, the information is stored in the database, and if the information is illegal, error information is returned.
As a preferred embodiment, the system further comprises an advertisement pushing module, which includes:
an announcement management unit for managing announcement information;
the pushing unit is used for inquiring whether corresponding notice information exists in the cache unit and the database, if so, pushing the notice information to the message queue, and if not, returning error information;
and the second monitoring module is used for monitoring whether the announcement information exists in the message queue or not, and pushing the announcement information to the user sides of all users through the WebSocket if the announcement information exists in the message queue.
The invention has the following beneficial effects:
1. the intelligent voice-based electric charge reminding method can automatically remind the user to pay the electric charge, solve the problem of untimely reminding, also can enable the user to know the own electricity utilization condition through a visual voice form, and is convenient for the user to pay the electric charge; on the other hand can alleviate electric power staff's the work load that the charges of electricity urge to pay and retrieve, realizes the intellectuality of working method, has the advantage of automaticity and high efficiency.
2. The invention relates to an electric charge fee-urging method based on intelligent voice, which is characterized in that a user side is connected with a server side through WebSocket, when the fee-urging reminding needs to be carried out on a user, the user establishing WebSocket connection with a local machine is inquired from the server side, and a message is sent to the user side; a message queue is introduced. When the system background receives the instant charge request, the request list is sent to the message queue, and the return value can be directly sent to the request without waiting for the completion of message sending, so that the asynchronous effect is realized. After monitoring the message, the monitor of the message queue can send the message to the user establishing WebSocket connection in the local machine, so that the effects of peak clipping, decoupling and asynchronization are realized.
3. The invention relates to an electric charge fee-urging method based on intelligent voice, which subdivides a timing fee-urging task into a timing voice synthesis task and a timing fee-urging pushing task, and accelerates the fee-urging efficiency by synthesizing bill information into voice in the background in advance.
Drawings
FIG. 1 is a flowchart of a method according to a first embodiment of the present invention;
FIG. 2 is a flowchart illustrating the operation of electric charge promotion according to an embodiment of the present invention;
FIG. 3 is a flow diagram of language synthesis in an embodiment of the present invention;
FIG. 4 is a flow chart of time allocation for timing tasks in an embodiment of the present invention;
FIG. 5 is a flow chart of adding a language synthesis template in an embodiment of the present invention;
fig. 6 is a flowchart illustrating advertisement pushing according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be understood that the step numbers used herein are for convenience of description only and are not intended as limitations on the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in the specification of the present invention and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of the described features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to and includes any and all possible combinations of one or more of the associated listed items.
The first embodiment is as follows:
referring to fig. 1 and 2, an electric charge fee-charging method based on intelligent voice includes the following steps:
establishing connection between a server and a user side through WebSocket; in the traditional electric charge reminding, reminding is carried out by dialing a telephone, and because the service control of dialing the telephone is tighter, the real telephone pushing is difficult to realize. Therefore, in the embodiment, the client webpage is adopted to simulate the user, the WebSocket is connected with the server, when the user needs to be reminded of charging, the server queries the user establishing the WebSocket connection with the local computer, and sends the message to the user side;
a Redis cache unit and a MySQL database for storing voice information are established at a server;
acquiring electric charge bill information, searching whether voice information corresponding to the current electric charge bill information exists in a Redis cache unit, if so, pushing the voice information to a RockettMQ message queue, and if not, searching whether voice information corresponding to the current electric charge bill information exists in a MySQL database;
if the MySQL database exists, the voice information is pushed to a RockettMQ message queue, if the MySQL database does not exist, new voice information is synthesized according to the electric bill information, and the voice information is stored in the MySQL database and a Redis cache unit and then pushed to the RockettMQ message queue;
and monitoring whether the RocktMQ message queue has the voice message, if so, pushing the voice message to a user end of the corresponding user through the WebSocket, and generating a fee urging record.
If the message pushing is performed on the fee urging object according to the WebSocket order, when the requested fee urging quantity is large, a long time is needed to process the service logic, which may cause the problems of request blocking, low program execution efficiency, low system availability, and the like. In order to solve the above problem, a message queue is introduced in the present embodiment. When the system background receives the instant charge request, the request list is sent to the message queue, and the return value can be directly sent to the request without waiting for the completion of message sending, so that the asynchronous effect is realized. After monitoring the message, the monitor of the message queue can send the message to the user establishing WebSocket connection in the local machine, so that the effects of peak clipping, decoupling and asynchronization are realized. In addition, multithreading is realized through Java concurrent programming in the interface of the instant charging promotion, so that a plurality of charging promotion tasks are executed in parallel. And a plurality of message queue listeners can be established by deploying a plurality of application instances, each application instance which receives the message can inquire whether a corresponding user exists in the WebSocket connection in the instance, and if the corresponding user exists, the application instance pushes the message to be charged.
Referring to fig. 3, as a preferred embodiment of this embodiment, the method for synthesizing new voice information according to the electric bill information specifically includes:
firstly, generating text data by using the electric bill information through a voice synthesis template, and then performing text preprocessing on an incoming Chinese text. Preprocessing symbols, and judging whether the incoming text has the components of 'v', 'and' or not. ","? ","! "symbols other than these four punctuation marks, if any, are converted into one of the above-mentioned four basic symbols according to a predefined symbol conversion rule. The symbols are processed, followed by a pre-processing of the digits, which are converted to a chinese literal representation if arabic numerals are present. When the preprocessing of the numbers is finished, text-to-pinyin conversion is carried out. The specific method for converting the text into the pinyin comprises the following steps: according to the existing character pinyin file and word pinyin file, Chinese text data is analyzed and converted into a corresponding pinyin representation form. The character pinyin file stores the mapping relation between all Chinese characters and pinyin, and the word pinyin file stores the mapping relation between most of words and pinyin. The word pinyin file mainly aims to eliminate the interference of polyphones as much as possible and ensure the accuracy of generated voice because the polyphones in Chinese characters are more and the same characters have different pronunciations in different words.
After the preprocessing is finished, the text data in the pinyin form can be loaded into the trained acoustic feature prediction model, the model can predict the text data, and the acoustic features represented by the Mel frequency spectrum are output. And restoring the acoustic characteristics represented by the Mel frequency spectrum into voice audio through a Griffin-lim algorithm, carrying out Base64 coding processing on the voice audio to generate the voice information, and returning the voice information to the requester.
As a preferred embodiment of this embodiment, monthly power charge information is required to be pushed to all users once every month, and the voice synthesis and voice charge pushing task for every month is implemented by adopting a back-end configuration timing task, which includes the steps of automatically generating language information at regular time and pushing charge information at regular time, where the step of automatically generating language information at regular time specifically includes:
establishing a timing voice synthesis task, and configuring a time period of the timing voice synthesis task and a corresponding user;
when the time period is reached, starting a timing voice synthesis task;
selecting a voice synthesis template, automatically acquiring the electric bill information generated by the corresponding user in a time period, generating voice information according to the selected voice synthesis template, and storing the voice information into a database and a cache unit;
the timed charge urging task comprises the following steps:
establishing a timing pushing charge-urging task, and configuring a time period of the timing pushing charge-urging task and a corresponding user;
and when the time period is reached, starting a timed pushing charge urging task, inquiring whether voice information of a corresponding user exists in the cache unit and the database, and if so, pushing the voice information to a message queue.
In this embodiment, the charge urging task is executed periodically every month, the last month bill information is synthesized into voice, and the voice is pushed to the user. Because voice synthesis needs a certain time, if the two steps of voice synthesis and voice pushing are carried out together, the problems that the voice pushing efficiency is low, a user cannot timely receive charge prompting information and the like can be caused. In order to solve the problem, the system subdivides the timing task of charge urging into a timing task of voice synthesis and a timing task of charge urging, and synthesizes bill information into voice in the background in advance to accelerate the efficiency of charge urging
Referring to fig. 4, the time period configuration process of the timed speech synthesis task and the timed charge-promoting push task specifically includes: the cron expression is generated according to the obtained parameters, and then the time and the range of the execution of the timing task can be dynamically changed by modifying the related configuration information of the timing task in the database without restarting the whole task. The specific implementation method comprises the steps of adding a timing task for monitoring data change of the database, polling time configuration of the timing task of the database, modifying starting time of the timing task if the time configuration is detected to be changed, and then starting and executing the timing task at regular time. And finally returning the time configuration of the timing task to the requesting party after the time configuration of the timing task is modified.
In this embodiment, the time period is set to one month, when the time period is reached, a speech synthesis timing task is first started to be executed, the timing task will first query whether a currently selected speech synthesis template exists in the Redis cache unit, if not, the currently selected template will be queried in the MySQL database, if not, it is determined that no currently selected speech synthesis template is selected, and a system default speech synthesis template is selected. If the currently selected voice synthesis module exists in the MySQL database, the voice synthesis template is stored in a Redis cache unit, then after the voice synthesis template is confirmed to be finished, monthly bill information of the corresponding user is obtained, then the corresponding voice information is synthesized for each bill according to the template, the bill information is updated, and finally the synthesized voice information is stored in the MySQL database and the Redis cache unit.
Referring to fig. 5, as a preferred embodiment of this embodiment, this embodiment supports a speech synthesis template of the customized toll information, and the established speech synthesis template is stored in a database, where the method for establishing the speech synthesis template specifically includes:
firstly, inputting newly added template information;
checking the template information placeholder conformity, if the placeholder is used correctly, checking the template information placeholder and the element quantity, and if the placeholder is consistent with the element quantity, storing the template information into a database. And if the placeholder is used incorrectly or the placeholder is inconsistent with the element number, returning a result with a transmission parameter error, and returning the result to the requester after the checking is finished.
The method specifically comprises the following steps: the administrator uses the "$" as a placeholder and passes the elements represented by each placeholder to the back-end in sequential columns. The administrator selects the template used by the current voice synthesis, and after the selection is successful, the voice synthesis module can synthesize voice information according to the currently set template, so that diversified voice reminding contents are provided.
Referring to fig. 6, as a preferred embodiment, a step for pushing the announcement to all users is further included, which is specifically as follows:
acquiring announcement information;
inquiring whether corresponding announcement information exists in a Redis cache unit, if so, pushing the announcement information to a RockettMQ message queue, and if not, inquiring whether the corresponding announcement information exists in a MySQL database;
if the MySQL database exists, storing the notice information into a Redis cache unit, pushing the notice information to a RockketMQ message queue, and if the MySQL database does not exist, returning error information;
monitoring whether the notice information exists in the RocktMQ message queue or not, and if yes, pushing the notice information to the user sides of all users through WebSocket.
Example two:
this embodiment provides a charge system is urged to charges of electricity based on intelligence pronunciation, includes:
the server side is established with a Redis cache unit for storing voice information and a MySQL database; the system is also used for managing related user information, including adding user information, deleting user information, modifying user information and inquiring user information;
the user side is used for receiving voice information;
the communication module is used for establishing the connection between the server and the user through WebSocket;
the bill management module is used for managing the electric bill information, and comprises the steps of adding, deleting, modifying and inquiring the electric bill information;
the voice synthesis module is used for synthesizing voice information according to the electric bill information;
the fee urging module is used for acquiring the electric bill information and inquiring whether the voice information corresponding to the electric bill information exists in the cache unit and the database; if the corresponding voice information exists in the message queue, outputting the corresponding voice information to the RockettMQ message queue, and if the corresponding voice information does not exist in the message queue, calling a voice synthesis module to synthesize the corresponding voice information and storing the corresponding voice information in a Redis cache unit and a MySQL database;
the first monitoring module is used for monitoring whether the RocktMQ message queue has the voice message or not, if so, the voice message is pushed to the user side of the corresponding user through the WebSocket, and a fee urging record is generated.
As a preferred implementation of this embodiment, the speech synthesis module includes:
the preprocessing unit is used for preprocessing, generating text data from the electric bill information through a voice synthesis template, and converting the text data into a pinyin representation form;
the acoustic feature prediction model is used for inputting the preprocessed text data and outputting corresponding acoustic features;
and the synthesis unit is used for converting the acoustic characteristics into voice audio through a vocoder, and carrying out coding processing on the voice audio to generate the voice information.
As a preferred implementation manner of this embodiment, the system further includes a module for automatically generating voice information at regular time and a module for pushing fee-prompting information at regular time; wherein the timing automatic generation voice information module is used for:
establishing a timing voice synthesis task, and configuring a time period of the timing voice synthesis task and a corresponding user;
when the time period is reached, starting a timing voice synthesis task;
selecting a voice synthesis template, automatically acquiring the electric bill information generated by the corresponding user in a time period, generating voice information according to the selected voice synthesis template, and storing the voice information into a database and a cache unit;
the timing pushing fee-urging information module is used for:
establishing a timing pushing charge-urging task, and configuring a time period of the timing pushing charge-urging task and a corresponding user;
and when the time period is reached, starting a timed pushing charge urging task, inquiring whether voice information of a corresponding user exists in the cache unit and the database, and if so, pushing the voice information to a message queue.
As a preferred implementation of this embodiment, the method further includes a speech synthesis template establishing module, including:
the input unit is used for inputting newly added template information;
the checking unit is used for checking whether the newly added template information is legal or not according to a preset rule; if the information is legal, the information is stored in the database, and if the information is illegal, error information is returned.
As a preferred implementation manner of this embodiment, the system further includes an advertisement pushing module, which includes:
the system comprises an announcement management unit, an announcement processing unit and an announcement processing unit, wherein the announcement management unit is used for managing announcement information, and comprises newly added announcement information, deleted announcement information, modified announcement information and inquired announcement information;
the pushing unit is used for inquiring whether corresponding notice information exists in the cache unit and the database, if so, pushing the notice information to the message queue, and if not, returning error information;
and the second monitoring module is used for monitoring whether the announcement information exists in the message queue or not, and pushing the announcement information to the user sides of all users through the WebSocket if the announcement information exists in the message queue.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.

Claims (10)

1.一种基于智能语音的电费催费方法，其特征在于，包括以下步骤：1. a method for reminding electricity charges based on intelligent voice, is characterized in that, comprises the following steps: 通过WebSocket建立服务端与用户端的连接；Establish a connection between the server and the client through WebSocket; 在服务端建立用于存储语音信息的缓存单元和数据库；Establish a cache unit and a database for storing voice information on the server; 获取电费账单信息，查找缓存单元中是否存在当前电费账单信息对应的语音信息，缓存单元中存在则将语音信息推送至消息队列，缓存单元中不存在则查找数据库中是否存在当前电费账单信息对应的语音信息；Obtain electricity bill information, find out whether there is voice information corresponding to the current electricity bill information in the cache unit, push the voice information to the message queue if it exists in the cache unit, and check whether there is a current electricity bill information in the database if it does not exist in the cache unit. voice message; 数据库中存在则将语音信息推送至消息队列，数据库中存在不存在则根据电费账单信息合成新的语音信息，并将该语音信息存入数据库和缓存单元后，推送至消息队列；If the voice information exists in the database, the voice information will be pushed to the message queue; if it does not exist in the database, new voice information will be synthesized according to the electricity bill information, and the voice information will be stored in the database and the cache unit, and then pushed to the message queue; 监听消息队列是否存在语音消息，若存在则通过WebSocket将语音信息推送至对应用户的用户端，并生成催费记录。Monitors the message queue to see if there is a voice message, and if so, pushes the voice message to the corresponding user's client through WebSocket, and generates a reminder record. 2.根据权利要求1所述的一种基于智能语音的电费催费方法，其特征在于，所述根据电费账单信息合成新的语音信息的方法具体为：2. a kind of electricity fee reminder method based on intelligent voice according to claim 1 is characterized in that, the described method for synthesizing new voice information according to electricity bill information is specifically: 预处理，将电费账单信息通过语音合成模板生成文本数据，再将文本数据转换为拼音表示形式；Preprocessing, generating text data from the electricity bill information through a speech synthesis template, and then converting the text data into pinyin representation; 将预处理后的文本数据输入至预训练的声学特征预测模型，输出对应的声学特征，再通过声码器将声学特征转换为语音音频，对语音音频进行编码处理，生成所述语音信息。Input the preprocessed text data into the pre-trained acoustic feature prediction model, output the corresponding acoustic features, and then convert the acoustic features into speech audio through a vocoder, and encode the speech audio to generate the speech information. 3.根据权利要求2所述的一种基于智能语音的电费催费方法，其特征在于，还包括定时自动生成语言信息和定时推送催费信息的步骤，其中，所述定时自动生成语言信息的步骤具体为：3. a kind of electricity fee reminder method based on intelligent voice according to claim 2, is characterized in that, also comprises the step of regularly automatically generating language information and regularly pushing the step of reminder information, wherein, described regularly automatically generating the step of language information. The steps are as follows: 建立定时语音合成任务，配置定时语音合成任务的时间周期和对应用户；Create a timed speech synthesis task, configure the time period of the timed speech synthesis task and the corresponding user; 当达到时间周期时，启动定时语音合成任务；When the time period is reached, start the scheduled speech synthesis task; 选择语音合成模板，自动获取时间周期内对应用户产生的电费账单信息，根据选择的语音合成模板生成语音信息，并存入数据库和缓存单元中；Select the speech synthesis template, automatically obtain the electricity bill information generated by the corresponding user in the time period, generate the speech information according to the selected speech synthesis template, and store it in the database and the cache unit; 所述定时催费推送任务的步骤具体为：The steps of the described timed reminder push task are specifically: 建立定时推送催费任务，并配置定时推送催费任务的时间周期和对应用户；Create a scheduled push reminder task, and configure the time period and corresponding user for the scheduled push reminder task; 当达到时间周期时，启动定时推送催费任务，查询缓存单元和数据库中是否存在对应用户的语音信息，若存在则将语音信息推送至消息队列。When the time period is reached, the scheduled push reminder task is started, and the cache unit and the database are queried for the existence of the corresponding user's voice information, and if so, the voice information is pushed to the message queue. 4.根据权利要求3所述的一种基于智能语音的电费催费方法，其特征在于，所述数据库中存储有至少一种所述语音合成模板，所述语音合成模板的建立方法具体为：4. a kind of electricity fee reminder method based on intelligent voice according to claim 3, is characterized in that, at least one described speech synthesis template is stored in described database, and the establishment method of described speech synthesis template is specifically: 输入新增的模板信息；Enter the new template information; 按预定的规则检查新增的模板信息是否合法；Check whether the newly added template information is legal according to the predetermined rules; 合法则存入数据库，不合法则返回错误信息。If it is legal, it will be stored in the database, and if it is illegal, an error message will be returned. 5.根据权利要求1所述的一种基于智能语音的电费催费方法，其特征在于，还包括一用于向所有用户推送公告的步骤，具体如下：5. a kind of electricity fee reminder method based on intelligent voice according to claim 1, is characterized in that, also comprises a step for pushing announcement to all users, is specifically as follows: 获取公告信息；get announcement information; 查询缓存单元中是否存在对应的公告信息，若缓存单元中存在，则将公告信息推送至消息队列，若缓存单元中不存在，则查询数据库中是否存在对应的公告信息；Query whether there is corresponding announcement information in the cache unit, if it exists in the cache unit, push the announcement information to the message queue, and if it does not exist in the cache unit, query whether there is corresponding announcement information in the database; 若数据库中存在，则将公告信息存入缓存单元后，推送至消息队列，若数据库中不存在，则返回错误信息；If it exists in the database, the announcement information will be stored in the cache unit and pushed to the message queue; if it does not exist in the database, an error message will be returned; 监听消息队列是否存在公告信息，若存在则通过WebSocket将公告信息推送至所有用户的用户端。Monitors the message queue to see if there is announcement information, and if so, pushes the announcement information to all users' clients through WebSocket. 6.一种基于智能语音的电费催费系统，其特征在于，包括：6. an electricity bill reminder system based on intelligent voice, is characterized in that, comprises: 服务端，建立有存储语音信息的缓存单元和数据库；On the server side, a cache unit and a database for storing voice information are established; 用户端，用于接收语音信息；The user terminal is used to receive voice information; 通信模块，用于通过WebSocket建立服务端与用户端的连接；The communication module is used to establish the connection between the server and the client through WebSocket; 账单管理模块，用于管理电费账单信息；Bill management module, used to manage electricity bill information; 语音合成模块，用于根据电费账单信息合成语音信息；The speech synthesis module is used to synthesize speech information according to the electricity bill information; 催费模块，用于获取电费账单信息，并查询缓存单元和数据库中是否存在与电费账单信息对应的语音信息；若存在，则将对应语音信息输出至消息队列，若不存在，则调用语音合成模块合成对应的语音信息并存入缓存单元和数据库中；The reminder module is used to obtain the electricity bill information and query whether there is voice information corresponding to the electricity bill information in the cache unit and database; if so, output the corresponding voice information to the message queue, and if not, call speech synthesis The module synthesizes the corresponding voice information and stores it in the cache unit and the database; 第一监听模块，用于监听消息队列是否存在语音消息，若存在则通过WebSocket将语音信息推送至对应用户的用户端，并生成催费记录。The first monitoring module is used to monitor whether there is a voice message in the message queue, and if so, push the voice message to the client of the corresponding user through WebSocket, and generate a reminder record. 7.根据权利要求6所述的一种基于智能语音的电费催费系统，其特征在于，所述语音合成模块包括：7. a kind of electricity bill reminder system based on intelligent voice according to claim 6, is characterized in that, described speech synthesis module comprises: 预处理单元，用于预处理，将电费账单信息通过语音合成模板生成文本数据，再将文本数据转换为拼音表示形式；The preprocessing unit is used for preprocessing, generating text data from the electricity bill information through a speech synthesis template, and then converting the text data into a Pinyin representation; 声学特征预测模型，用于输入预处理后的文本数据，并输出对应的声学特征；The acoustic feature prediction model is used to input the preprocessed text data and output the corresponding acoustic features; 合成单元，用于通过声码器将声学特征转换为语音音频，对语音音频进行编码处理，生成所述语音信息。The synthesis unit is used for converting the acoustic features into speech audio through a vocoder, encoding the speech audio, and generating the speech information. 8.根据权利要求6所述的一种基于智能语音的电费催费系统，其特征在于，还包括定时自动生成语音信息模块和定时推送催费信息模块；其中，所述定时自动生成语音信息模块用于：8. a kind of electricity fee reminder system based on intelligent voice according to claim 6, is characterized in that, also comprises timing automatic generation voice information module and timing push reminder information module; Wherein, described timing automatically generates voice information module Used for: 建立定时语音合成任务，配置定时语音合成任务的时间周期和对应用户；Create a timed speech synthesis task, configure the time period of the timed speech synthesis task and the corresponding user; 当达到时间周期时，启动定时语音合成任务；When the time period is reached, start the scheduled speech synthesis task; 选择语音合成模板，自动获取时间周期内对应用户产生的电费账单信息，根据选择的语音合成模板生成语音信息，并存入数据库和缓存单元中；Select the speech synthesis template, automatically obtain the electricity bill information generated by the corresponding user in the time period, generate the speech information according to the selected speech synthesis template, and store it in the database and the cache unit; 所述定时推送催费信息模块用于：The timing push reminder information module is used for: 建立定时推送催费任务，并配置定时推送催费任务的时间周期和对应用户；Create a scheduled push reminder task, and configure the time period and corresponding user for the scheduled push reminder task; 当达到时间周期时，启动定时推送催费任务，查询缓存单元和数据库中是否存在对应用户的语音信息，若存在则将语音信息推送至消息队列。When the time period is reached, the scheduled push reminder task is started, and the cache unit and the database are queried for the existence of the corresponding user's voice information, and if so, the voice information is pushed to the message queue. 9.根据权利要求8所述的一种基于智能语音的电费催费系统，其特征在于，还包括语音合成模板建立模块，包括：9. a kind of electricity fee reminder system based on intelligent voice according to claim 8, is characterized in that, also comprises speech synthesis template establishment module, comprises: 输入单元，用于输入新增的模板信息；The input unit is used to input the newly added template information; 检查单元，用于按预定的规则检查新增的模板信息是否合法；合法则存入数据库，不合法则返回错误信息。The checking unit is used to check whether the newly added template information is legal according to predetermined rules; if it is legal, it will be stored in the database, and if it is illegal, an error message will be returned. 10.根据权利要求6所述的一种基于智能语音的电费催费系统，其特征在于，还包括一公告推送模块，包括：10. A kind of intelligent voice-based electricity fee reminder system according to claim 6, is characterized in that, also comprises an announcement push module, comprising: 公告管理单元，用于管理公告信息；Announcement management unit, used to manage announcement information; 推送单元，用于查询缓存单元和数据库中是否存在对应的公告信息，若存在，则将公告信息推送至消息队列，若不存在，则返回错误信息；The push unit is used to query whether the corresponding announcement information exists in the cache unit and the database. If it exists, the announcement information is pushed to the message queue, and if it does not exist, an error message is returned; 第二监听模块，用于监听消息队列是否存在公告信息，若存在则通过WebSocket将公告信息推送至所有用户的用户端。The second monitoring module is used to monitor whether there is announcement information in the message queue, and if there is, push the announcement information to the clients of all users through WebSocket.
CN202111318798.0A 2021-11-09 2021-11-09 Electricity bill collection method and system based on intelligent voice Active CN114155828B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111318798.0A CN114155828B (en) 2021-11-09 2021-11-09 Electricity bill collection method and system based on intelligent voice

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111318798.0A CN114155828B (en) 2021-11-09 2021-11-09 Electricity bill collection method and system based on intelligent voice

Publications (2)

Publication Number Publication Date
CN114155828A true CN114155828A (en) 2022-03-08
CN114155828B CN114155828B (en) 2024-06-18

Family

ID=80459596

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111318798.0A Active CN114155828B (en) 2021-11-09 2021-11-09 Electricity bill collection method and system based on intelligent voice

Country Status (1)

Country Link
CN (1) CN114155828B (en)

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140211665A1 (en) * 2013-01-28 2014-07-31 Rackspace Us, Inc. Methods and Systems of Generating a Billing Feed of a Distributed Network
US20170161851A1 (en) * 2015-12-08 2017-06-08 Toast, Inc. Restaurant Notification System
US20170272504A1 (en) * 2016-03-17 2017-09-21 Webtext Holdings Limited Message transfer system, method of transferring messages and software product
CN108345635A (en) * 2017-12-29 2018-07-31 广州品唯软件有限公司 A kind of Billing System and bill generation processing method of High Availabitity
US20190243836A1 (en) * 2018-02-08 2019-08-08 Parallel Wireless, Inc. Data Pipeline for Scalable Analytics and Management
CN111160886A (en) * 2020-04-02 2020-05-15 支付宝(杭州)信息技术有限公司 Electronic bill sending method and device, storage medium and computer equipment
CN111221509A (en) * 2019-12-27 2020-06-02 江苏苏州农村商业银行股份有限公司 Customizable online payment service development platform based on spring cloud architecture system
CN111857880A (en) * 2020-07-23 2020-10-30 中国平安人寿保险股份有限公司 Dialogue configuration item information management method, device, equipment and storage medium
CN112256954A (en) * 2020-09-08 2021-01-22 国网宁夏电力有限公司经济技术研究院 Message pushing processing method and related system
CN112527527A (en) * 2020-12-16 2021-03-19 深圳市分期乐网络科技有限公司 Consumption speed control method and device of message queue, electronic equipment and medium
CN112583638A (en) * 2020-11-25 2021-03-30 新浪网技术(中国)有限公司 Websocket-based voice alarm method and system
CN113129063A (en) * 2021-04-20 2021-07-16 国网江西省电力有限公司信息通信分公司 Electricity charge calculation and distribution method and system based on cloud platform and data center

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140211665A1 (en) * 2013-01-28 2014-07-31 Rackspace Us, Inc. Methods and Systems of Generating a Billing Feed of a Distributed Network
US20170161851A1 (en) * 2015-12-08 2017-06-08 Toast, Inc. Restaurant Notification System
US20170272504A1 (en) * 2016-03-17 2017-09-21 Webtext Holdings Limited Message transfer system, method of transferring messages and software product
CN108345635A (en) * 2017-12-29 2018-07-31 广州品唯软件有限公司 A kind of Billing System and bill generation processing method of High Availabitity
US20190243836A1 (en) * 2018-02-08 2019-08-08 Parallel Wireless, Inc. Data Pipeline for Scalable Analytics and Management
CN111221509A (en) * 2019-12-27 2020-06-02 江苏苏州农村商业银行股份有限公司 Customizable online payment service development platform based on spring cloud architecture system
CN111160886A (en) * 2020-04-02 2020-05-15 支付宝(杭州)信息技术有限公司 Electronic bill sending method and device, storage medium and computer equipment
CN111857880A (en) * 2020-07-23 2020-10-30 中国平安人寿保险股份有限公司 Dialogue configuration item information management method, device, equipment and storage medium
CN112256954A (en) * 2020-09-08 2021-01-22 国网宁夏电力有限公司经济技术研究院 Message pushing processing method and related system
CN112583638A (en) * 2020-11-25 2021-03-30 新浪网技术(中国)有限公司 Websocket-based voice alarm method and system
CN112527527A (en) * 2020-12-16 2021-03-19 深圳市分期乐网络科技有限公司 Consumption speed control method and device of message queue, electronic equipment and medium
CN113129063A (en) * 2021-04-20 2021-07-16 国网江西省电力有限公司信息通信分公司 Electricity charge calculation and distribution method and system based on cloud platform and data center

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
赵斌 等: "智能语音电费催费系统的设计与实现", 福建电脑, vol. 38, no. 3, 31 March 2022 (2022-03-31) *

Also Published As

Publication number Publication date
CN114155828B (en) 2024-06-18

Similar Documents

Publication Publication Date Title
US6868143B1 (en) System and method for advanced unified messaging
CN112434501B (en) Method, device, electronic equipment and medium for intelligent generation of worksheet
CN102456344B (en) System and method for analyzing customer behavior characteristic based on speech recognition technique
CN102135985B (en) Method and system for searching by calling search result of third-party search engine
CN110798578A (en) Incoming call transaction management method and device and related equipment
CN102270363A (en) Intelligent queuing system
CN110795147A (en) Interface protocol file management method and device
CN1141831C (en) Short speech message transmission system
CN106209495A (en) The report form generation method of a kind of cloud monitoring system and device
CN101710927B (en) Method and system for providing information service based on voice platform
CN113448544B (en) Automatic document generation method and system in program development
CN113947368A (en) A contract early warning method, device and computer-readable storage medium
CN117059095A (en) IVR-based service providing method and device, computer equipment and storage medium
CN113055278A (en) Mail filing processing method and device
CN102044006A (en) Control method and control device for associating electronic calendar with address book
CN114155828A (en) Electric charge fee-prompting method and system based on intelligent voice
CN113364669B (en) Message processing method and device, electronic equipment and medium
CN114595972A (en) Smart city management method applying virtual digital people
CN107295029B (en) An information processing method and device
CN108572949A (en) A kind of houseclearing search processing method and device
CN112184481A (en) Smart grid arrearage processing method and device combining RPA and AI
CN114090305B (en) Business auditing method and device
CN115114114A (en) Data report generation method and device in online file feeding mode and related equipment
CN108304219A (en) Secondary developing platform and method
CN101340492B (en) An information query method and device

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