CN110110321A - Products Show method, apparatus, equipment and storage medium based on voice data - Google Patents
Products Show method, apparatus, equipment and storage medium based on voice data Download PDFInfo
- Publication number
- CN110110321A CN110110321A CN201910207372.4A CN201910207372A CN110110321A CN 110110321 A CN110110321 A CN 110110321A CN 201910207372 A CN201910207372 A CN 201910207372A CN 110110321 A CN110110321 A CN 110110321A
- Authority
- CN
- China
- Prior art keywords
- user
- portrait
- group
- time
- preset
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000004891 communication Methods 0.000 claims abstract description 69
- 238000004458 analytical method Methods 0.000 claims abstract description 68
- 238000000605 extraction Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 17
- 230000001755 vocal effect Effects 0.000 claims description 11
- 238000007405 data analysis Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 description 8
- 230000003542 behavioural effect Effects 0.000 description 7
- 238000013507 mapping Methods 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The Products Show method based on voice data that the invention discloses a kind of, device, equipment and storage medium, the described method includes: obtaining the user identifier and history calling record of user, the history calling record of each user is converted into text file and carries out semantic analysis, obtain user characteristics, it is analyzed using history calling record of the speech analysis model to each user, settling time portrait, it is drawn a portrait using user characteristics and time portrait building user, and user identifier and user are drawn a portrait into corresponding storage into user information database, according to the user identifier of user to be linked up, user's portrait corresponding with the user identifier of user to be linked up is obtained from user information database, target product is determined according to the user characteristics that the user draws a portrait, and it can time for communication according to the time portrait determination of user's portrait, by target product and can time for communication it is defeated Client is arrived out.The embodiment of the present invention can effectively be provided for client and targetedly be serviced, and improve the popularization success rate of service product.
Description
Technical field
The present invention relates to data analysis technique field more particularly to a kind of Products Show methods based on voice data, dress
It sets, equipment and storage medium.
Background technique
Telemarketing is a kind of common marketing mode, and telemarketing refers to that sales force and client carry out ditch by phone
Communications and liaison system, to determine client to the satisfaction of product or in a manner of new order can be received.
Currently, the financial institutions such as bank, security and insurance in the popularization of financial product or other service products, sell
Personnel generally can be by phone to lead referral product, to promote and sell itself service product, however, telemarketing person
It is not and client is aspectant exchanges that not can determine that client, whether having time carries out telephonic communication, also, telemarketing person exists
When contacting strange client, the demand of client cannot be accurately held, it, can not recommended user's sense in the unfamiliar situation of client
The service product of interest is targetedly serviced therefore, because telemarketing person cannot effectively provide for client, is faced with objective
The case where family hangs up the telephone or directly refuses call is relatively conventional, causes the popularization success rate of service product generally relatively low.
Summary of the invention
A kind of Products Show method, apparatus, equipment and storage medium based on voice data is provided in the embodiment of the present invention,
With solve the problems, such as currently in telemarketing it is low to the specific aim Generalization Ability of service product with promote success rate it is low.
A kind of Products Show method based on voice data, comprising:
Obtain the user identifier and history calling record of at least one user；
The history calling record of each user is converted to by text file using preset speech recognition modeling,
And semantic analysis is carried out to the corresponding text file of each user, obtain the user characteristics of each user；
According to preset speech analysis model, the history calling record of each user is analyzed, is established
The time of each user draws a portrait；
It is drawn a portrait using the user that the user characteristics of the user and time portrait construct the user, and by institute
The user identifier and the user for stating user draw a portrait corresponding storage into preset user information database；
Obtain the user identifier of user to be linked up；
If there is the user identifier of the user to be linked up in the preset user information database, obtain described wait link up
The corresponding user's portrait of the user identifier of user, draws a portrait as target user；
According to the user characteristics in target user portrait, first object product is determined, and according to the mesh
Mark the time portrait in user's portrait, determine first can time for communication, and by the first object product and described first
Can time for communication be output to client.
A kind of Products Show device based on voice data, comprising:
Voice data obtains module, for obtaining the user identifier and history calling record of at least one user；
Fisrt feature obtains module, for being led to the history of each user using preset speech recognition modeling
Words recording is converted to text file, and carries out semantic analysis to the corresponding text file of each user, obtains each
The user characteristics of the user；
First data analysis module is used for according to preset speech analysis model, to the history of each user
Calling record is analyzed, and the time portrait of each user is established；
First portrait memory module, for using described in the user characteristics and the time portrait building of the user
The user of user draws a portrait, and the user identifier of the user and the user are drawn a portrait to correspond to store to preset user and believed
It ceases in library；
First identifier obtains module, for obtaining the user identifier of user to be linked up；
First portrait obtains module, if the user for there is the user to be linked up in the preset user information database
Mark then obtains the corresponding user's portrait of user identifier of the user to be linked up, draws a portrait as target user；
First result output module, for determining the first mesh according to the user characteristics in target user portrait
Mark product, and drawn a portrait according to the time in target user portrait, determine first can time for communication, and by described the
One target product and described first can time for communication be output to client.
A kind of computer equipment, including memory, processor and storage are in the memory and can be in the processing
The computer program run on device, the processor realize the above-mentioned product based on voice data when executing the computer program
Recommended method.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program realizes the above-mentioned Products Show method based on voice data when being executed by processor.
Above-mentioned Products Show method, apparatus, equipment and storage medium based on voice data, by obtaining at least one use
The user identifier and history calling record at family；History calling record is converted into text file, text corresponding to each user
File is analyzed, and the user characteristics of user are obtained, to determine the behavioural habits and hobby of user；According to preset language
Sound analysis model analyzes the history calling record of each user, the time portrait of user is established, so as to analyze
The busy or idle period of user out formulates the time book table of each user；Use user characteristics and time
Portrait building user's portrait, draws a portrait corresponding storage into user information database for user identifier and user, improves the receipts of user information
Collect efficiency；The user identifier for obtaining user to be linked up obtains corresponding with the user identifier of user to be linked up from user information database
User portrait；Drawn a portrait according to the user that gets determine target product and can time for communication, and by target product and can link up
Time is output to client, realize treat link up user can time for communication and product of interest intelligent recognition and push away automatically
It send, so that business personnel can be that user to be linked up formulates marketing strategy according to the information being pushed, so as to targeted
Ground is that user recommends service product, the service quality that service product is promoted is improved, to improve the marketing success rate of service product.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below by institute in the description to the embodiment of the present invention
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is an application environment schematic diagram of the Products Show method in one embodiment of the invention based on voice data；
Fig. 2 is a flow chart of the Products Show method in one embodiment of the invention based on voice data；
Fig. 3 is the one specific of building group's portrait in Products Show method in one embodiment of the invention based on voice data
Flow chart；
Fig. 4 is user characteristics are obtained in Products Show method in one embodiment of the invention based on voice data one specific
Flow chart；
Fig. 5 is the time portrait that user is established in Products Show method in one embodiment of the invention based on voice data
One specific flow chart；
Fig. 6 is one that group's time portrait is established in Products Show method in one embodiment of the invention based on voice data
Specific flow chart；
Fig. 7 is target user is obtained in Products Show method in one embodiment of the invention based on voice data one specific
Flow chart；
Fig. 8 is a functional block diagram of the Products Show device in one embodiment of the invention based on voice data；
Fig. 9 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
Products Show method provided by the embodiments of the present application based on voice data, can be applicable to the application environment such as Fig. 1
In, which includes server-side and client, wherein is attached, is serviced by network between server-side and client
End constructs user's portrait according to the history calling record of user, and business personnel inputs user identifier in client, and client will be used
Family mark is sent to server-side, and server-side searches corresponding user according to user identifier and draws a portrait, and is drawn a portrait according to user and determine mesh
Mark product and can time for communication, and export to client.Client specifically can be, but not limited to be various personal computers, notes
This computer, smart phone, tablet computer and portable wearable device, server-side can specifically use independent server or more
The server cluster of a server composition is realized.The method application of Products Show of the offer of the embodiment of the present invention based on voice data
In server-side.
In one embodiment, Fig. 2 shows a flow charts of the Products Show method in the present embodiment based on voice data, should
Method applies the server-side in Fig. 1, for construct user portrait, and according to user draw a portrait to business personnel recommend with wait link up
The corresponding target product of user and can time for communication.As shown in Fig. 2, the Products Show method based on voice data of being somebody's turn to do includes step
S11 is to step S17, and details are as follows:
S11: the user identifier and history calling record of at least one user are obtained.
In the present embodiment, the corresponding user identifier of each user, user identifier is the body for capableing of unique identification user
The identification information of part, the user identifier specifically can be the ID card No. or telephone number of user, be also possible to user's note
The account information etc. of volume, can specifically be configured, herein with no restrictions according to the needs of practical application.
Specifically, history calling record is the recording of the business exchange process of business personnel and user.Generally, in silver
The financial institutions such as row, security and insurance in the popularization of financial product or other service products, business personnel by phone to
User's recommended products, meanwhile, the dialog context of marketing process is recorded using sound pick-up outfit, obtains history calling record,
And accordingly store the user identifier of user and history calling record in the database for being used to store history calling record, often
The corresponding user identifier of a history calling record enables server-side to determine the history call record of user according to user identifier
Sound, server-side can obtain the user identifier and history calling record of user from the database of the storage history calling record.
S12: the history calling record of each user is converted to by text file using preset speech recognition modeling, and right
The corresponding text file of each user carries out semantic analysis, obtains the user characteristics of each user.
In the present embodiment, preset speech recognition modeling is the model for voice content to be converted to word content,
Server-side obtains the spectrum signature of history calling record, and the frequency of usage history calling record using preset speech recognition modeling
Spectrum signature, physical spectrum feature corresponding with vowel pre-stored in speech recognition modeling, word tone or word are matched,
The special corresponding text of physical spectrum identical with the spectrum signature of history calling record is exported, so that the history of each user be led to
Words recording is converted to corresponding text file.
Optionally, server-side is using Chinese natural language handling implement FudanNLP to the corresponding text file of each user
Semantic analysis is carried out, understands the semantic informations such as meaning, theme and the similarity of word and sentence in text file, and according to word
Related information present in language, sentence and text identifies to obtain relevant semantic information, thus obtain semantic analysis as a result,
To the user characteristics of each user, wherein user characteristics refer to the features such as behavioural habits and the hobby of user, each user
Including at least one user characteristics.
For example, the text file of a certain user is directed to, after business personnel recommends service product, if in the answer of user
Holding includes the keywords such as " being out of use " and " not needing ", then it represents that user holds the attitude of refusal to service product, can analyze
The user characteristics of the service product are not liked to the user；If the answer content of user includes " consideration " and " introduction " or and industry
The relevant keyword of business product, then it represents that user is interested in service product, can analyze to obtain the user to the service product
Interested user characteristics.
S13: according to preset speech analysis model, analyzing the history calling record of each user, establishes each
The time of user draws a portrait.
In the present embodiment, time portrait refers to the time book table of user, for recording when linking up of user
Between and the rush hour.Preset speech analysis model is extracted and is analyzed for the phonetic feature to history calling record
Model.Server-side analyzes the history calling record of each user, according to the duration of call and history of history calling record
The phonetic feature of calling record, judges whether user conveniently answers sales calls in the history calling record corresponding time,
In, phonetic feature is the parameter for describing the voice signal of user's sending, which includes but is not limited to word speed, sound
Tune, volume and speech frequency etc..
For example, when the air time of history calling record A is 10 days 10 October in 2018, the call of history calling record A
Shi Changwei 1 minute, and in history calling record A user word speed quickly, indicate user may be inconvenient to answer in the time
Sales calls, on October 10th, 2018 is Wednesday, then by rush hour when the 10 of Wednesday labeled as user.History calling record
When the air time of B is 10 days 16 October in 2018, the duration of call of the calling record is 30 minutes, and history calling record
The volume of user is larger in B, indicates that user is free during this period of time and carries out business exchange, then will be labeled as using when the 16 of Wednesday
Family can time for communication.
Specifically, server-side analyzes history calling record of each user under different time, determines that user exists
The different periods can time for communication and the rush hour, and according to the obtained result of analysis establish each user include can ditch
The time portrait of logical time and the rush hour, to determine that user conveniently answers the time of sales calls.
It should be noted that it should be noted that do not have between step S12 and step S13 it is inevitable it is successive execute sequence,
It is also possible to the relationship executed side by side, herein with no restrictions.
S14: it is drawn a portrait using the user of the user characteristics of user and time portrait building user, and by the user identifier of user
Storage corresponding with user's portrait is into preset user information database.
In the present embodiment, it is drawn a portrait according to the time of the user characteristics of user and user, the corresponding user of building user draws
Picture, wherein be coordination between user characteristics and time portrait, be described for the different dimensions to user.
Specifically, after building obtains user's portrait, the user of the user identifier of user and user is drawn a portrait and correspond to storage
Into preset user information database, wherein the preset user information database is that server-side is preset for storing representation data
Database, user identifier and user portrait have one-to-one relationship.
S15: the user identifier of user to be linked up is obtained.
Specifically, user to be linked up is the target user that business personnel needs business of distributor product, the use of user to be linked up
Family mark can specifically be inputted in the client by business personnel, be also possible to obtain in the customer resources of server-side subsidiary company,
It can specifically be obtained according to the needs of practical application, herein with no restrictions.
S16: if there is the user identifier of user to be linked up in preset user information database, the use of user to be linked up is obtained
Family identifies corresponding user's portrait, draws a portrait as target user.
Specifically, whether server-side inquire according to the user identifier of user to be linked up in preset user information database
The corresponding user's portrait of the stored user identifier, if there is the user identifier of user to be linked up in preset user information database,
The corresponding user's portrait of user identifier for then obtaining user to be linked up, draws a portrait as target user, links up user's to treat
Information is analyzed.
S17: according to target user draw a portrait in user characteristics, determine first object product, and draw according to target user
As in time portrait, determine first can time for communication, and by first object product and first can time for communication be output to client
End.
Specifically, according to target user draw a portrait in user characteristics, the hobby of user to be linked up is obtained, in business people
In the product library of company where member, service product relevant to the hobby of user to be linked up is selected as first object
Product.Meanwhile according to target user draw a portrait in time draw a portrait, obtain the time that user to be linked up conveniently answers sales calls,
Obtain user to be linked up first can time for communication.
It should be noted that first object product can be a service product, it is also possible to multiple service products, first
Can time for communication can be a period, be also possible to multiple periods, can be determined according to the actual situation.
For example, comprising a user characteristics being " possessing two cars ", and the product library of company in target user's portrait
In it is related to user characteristics " possessing two cars " comprising " private savings vehicle insurance " and " traffic accident dangerous ", then by " private savings vehicle insurance " and
" traffic accident danger " is used as first object product.
By first object product and first can time for communication be output to client, for prompt service personnel user to be linked up
Interested product and Convenient telephone link up period, enable business personnel according to wait link up user demand and when
Between, targetedly recommend service product for user, provides personalized service, and the busy of user to be linked up can be avoided
Time carries out telephonic communication, promotes the quality of service.
In the corresponding embodiment of Fig. 2, by the user identifier and history calling record that obtain at least one user；It will go through
History calling record is converted to text file, analyzes the corresponding text file of each user, obtains the user characteristics of user,
To determine the behavioural habits and hobby of user；According to preset speech analysis model, converse the history of each user
Recording is analyzed, and the time portrait of user, so as to analyze the busy or idle period for obtaining user, system are established
The time book table of fixed each user；It is drawn a portrait using user characteristics and time portrait building user, by user identifier and use
Portrait corresponding storage in family improves the collection efficiency of user information into user information database；The user identifier of user to be linked up is obtained,
From user information database, user's portrait corresponding with the user identifier of user to be linked up is obtained；According to the user's portrait got
Determine target product and can time for communication, and by target product and can time for communication be output to client, realize and treat communication
User can time for communication and product of interest intelligent recognition and automatic push so that business personnel can be according to being pushed
Information is that user to be linked up formulates marketing strategy, so as to targetedly be that user recommends service product, improves business and produces
The service quality that product are promoted, to improve the marketing success rate of service product.
In one embodiment, the corresponding at least types of populations mark of user, as shown in figure 3, after step s 13, and
Before step S15, i.e., described according to preset speech analysis model, to the history calling record of each user into
Row analysis is established after the time portrait of each user, and before the user identifier for obtaining user to be linked up,
The Products Show method based on voice data further include:
S21: identifying according to group, in the user characteristics that every types of populations identifies corresponding user, chooses identical user
The quantity of feature is more than the user characteristics of preset amount threshold, and the user characteristics chosen are identified as group's mark and are corresponded to
Population characteristic.
In the present embodiment, the also corresponding at least types of populations mark of user, group's mark is for identifying belonging to the user
Group identification information, the group mark can be age or gender, be also possible to educational background, for example, " male " or " undergraduate course
Or more " etc. groups mark.
Specifically, the user characteristics obtained according to step S12 can determine that every types of populations identifies the user of corresponding user
Feature, in the user characteristics that every types of populations identifies corresponding user, it is more than preset for choosing the quantity of identical user characteristics
The user characteristics of amount threshold, and the user characteristics chosen are identified as group and identify corresponding population characteristic, for describing
Behavioural habits and hobby of the types of populations etc., wherein preset amount threshold specifically can be 10, be also possible to 100, tool
Body can need to be configured according to the actual situation, herein with no restrictions.
S22: it is carried out using the history calling record that preset speech analysis model identifies corresponding user to every types of populations
Analysis establishes every types of populations and identifies corresponding group's time portrait.
In the present embodiment, group's time portrait refers to the time book table of the user in group, for recording this
User in group is corresponding can time for communication and the rush hour.Preset speech analysis model is for history calling record
The phonetic feature model that extracts and analyze.Server-side to every types of populations identify the history calling record of corresponding user into
Row analysis, according to the phonetic feature of the duration of call of history calling record and history calling record, judges the use in the types of populations
Whether sales calls are conveniently answered in the history calling record corresponding time in family, wherein phonetic feature is for describing user
The parameter of the voice signal of sending, the phonetic feature include but is not limited to word speed, tone, volume and speech frequency etc..
Specifically, server-side identifies history calling record of the corresponding user under different time to every types of populations and divides
Analysis, determine user in the types of populations in different times section can time for communication and the rush hour, also, in the same time
Under, if user in the types of populations be in can the number of users of time for communication be greater than the number of users in the rush hour, will
The period can time for communication labeled as the group；Otherwise, if user in the types of populations be in can time for communication user
Quantity is less than the number of users in the rush hour, then the period is labeled as to the rush hour of the group.According to analyzing
To result establish every types of populations include can time for communication and group's time of the rush hour portrait, to determine the types of populations
In user conveniently answer time of sales calls.
It should be noted that it should be noted that do not have between step S21 and step S22 it is inevitable it is successive execute sequence,
It is also possible to the relationship executed side by side, herein with no restrictions.
S23: use groups identify corresponding population characteristic and group's time portrait building group identifies corresponding group and draws
Picture, and by the storage corresponding with group's portrait of group's mark into preset user information database.
In the present embodiment, corresponding population characteristic is identified according to group and group identifies corresponding group's time portrait,
Building group identifies corresponding group's portrait, wherein is coordination between population characteristic and group's time portrait, for group
The different dimensions of user in body are described.
Specifically, after building obtains group's portrait, preset user is arrived into the storage corresponding with group's portrait of group's mark
In information bank, which is that server-side presets database for storing representation data, group's mark
Drawing a portrait with group has one-to-one relationship.
Preferably, after the step s 15, i.e., described after the user identifier for obtaining user to be linked up in the present embodiment
Products Show method based on voice data further include:
S24: if the user identifier of user to be linked up is not present in preset user information database, obtain user's to be linked up
Group's mark.
Specifically, according to the user identifier of the step S15 user to be linked up obtained, server-side is in preset user information database
In, whether the corresponding user of the stored user identifier draws a portrait for inquiry, if there is no wait link up in preset user information database
The user identifier of user then obtains group's mark of user to be linked up.
For example, the gender of user to be linked up is women, and the age is 65 years old, then " female " and " the elderly " is general to ditch
The group at family identifies.
S25: obtaining group's portrait corresponding with the group of user to be linked up mark from preset user information database, as
Target group's portrait.
Specifically, the group's mark for the user to be linked up that essential step S124 is obtained, is obtained from preset user information database
It takes group corresponding with the group of user to be linked up mark to draw a portrait, draws a portrait as target group, so as to according to user institute to be linked up
The information of the group stated is treated communication user and is assessed.
S26: according to target group draw a portrait in population characteristic, determine the second target product, and draw according to target group
As in group's time portrait, determine second can time for communication, and by the second target product and second can time for communication be output to
Client.
Specifically, according to target group draw a portrait in population characteristic, obtain the hobby of the user in the types of populations,
In the product library of company where business personnel, service product relevant to the hobby of the user in the types of populations is selected
As the second target product.Meanwhile according to target group draw a portrait in group's time draw a portrait, obtain the user side in the types of populations
Just the time for answering sales calls, obtain user to be linked up second can time for communication.
It should be noted that the second target product can be a service product, it is also possible to multiple service products, second
Can time for communication can be a period, be also possible to multiple periods, can be determined according to the actual situation.
For example, comprising a population characteristic being " more ailing " in target group's portrait, and comprising " strong in the product library of company
Health danger " and " accident insurance " are related to population characteristic " how ailing ", then " health insurance " and " accident insurance " are used as the second target product.
By the second target product and second can time for communication be output to client, for prompt service personnel user to be linked up
The period that the interested product and Convenient telephone of the group at place are linked up, enable business personnel according to user to be linked up
The group at place needs integration time, targetedly recommends service product for user, provides personalized service, and can
The rush hour for avoiding user to be linked up carries out telephonic communication, promotes the quality of service.
In the corresponding embodiment of Fig. 3, the group by obtaining user is identified, and establishes group corresponding with every types of populations mark
Body characteristics and group's time portrait, obtain in the behavioural habits and hobby and every types of populations of the user in every types of populations
User the busy or idle period.Use groups identify corresponding population characteristic and group's time portrait building group
Corresponding group's portrait is identified, and the storage corresponding with group's portrait of group's mark is improved into group into preset user information database
The information collecting efficiency of body user.And group corresponding with the group of user to be linked up mark is obtained from preset user information database
Body portrait, drawn a portrait according to the group that gets determine target product and can time for communication, and by target product and can time for communication
It is pushed to business personnel, so that business personnel will appreciate that requirement product and working time wait link up the group where user
Section, enables business personnel to provide its interested product for client, targetedly recommends service product for user, improves industry
The service quality for product promotion of being engaged in, to improve the marketing success rate of service product.
In one embodiment, the present embodiment carries out the corresponding text file of each user to mentioned in step S12
Semantic analysis, the concrete methods of realizing for obtaining the user characteristics of each user are described in detail.
Referring to Fig. 4, Fig. 4 shows a specific flow chart of step S12, details are as follows:
S121: using preset semantic analysis model, carries out keyword extraction to the corresponding text file of each user, and
Semantic analysis is carried out to the keyword extracted, obtains target keyword.
In the present embodiment, preset semantic analysis model is for carrying out semantic analysis to word, sentence or article
Model, which specifically can be natural language processing tool FudanNLP, is also possible to Chinese
Adopted analysis system bosonnlp etc., can specifically be configured, herein with no restrictions according to the needs of practical application.
Specifically, server-side uses preset semantic analysis model, to the sentence in the corresponding text file of each user
It is analyzed and processed, extraction is able to reflect the theme of text file or the word of main contents as keyword, and to extraction
The keyword arrived carries out semantic analysis, understands the semantic informations such as meaning, theme and the similarity that antistop list reaches, and according to key
Related information present in word identifies to obtain relevant semantic information, and is spliced keyword according to the result of semantic analysis, or
Keyword is mapped to the behavior of user by method of the person in the corpus in preset semantic analysis model using fuzzy query
Habit or hobby, obtain target keyword.
For example, in one embodiment, a text file is that " I intends to go for a company's loan, uses mortgage
Mode buys the suite of rooms ", word segmentation processing is carried out to text file, and extract the theme or main contents for being able to reflect text file
Word as keyword, segmented keywords such as " plans ", " company ", " loan ", " mortgage ", " buying " and " house ", and
Semantic analysis is carried out to the keyword extracted, obtains target keyword " intending loan " and " buying house ".
S122: using target keyword as the user characteristics of user.
Specifically, the target keyword obtained according to step S121 is used using target keyword as the user characteristics of user
In the behavioural habits and hobby of description user.
In the corresponding embodiment of Fig. 4, by using preset semantic analysis model, to each user corresponding text text
Part carries out keyword extraction, and carries out semantic analysis to the keyword extracted, the user characteristics of user is obtained, will pass through use
The information of the call voice data-gathering user at family determines the behavioural habits and hobby of user.
In one embodiment, the present embodiment to mentioned in step S13 according to preset speech analysis model, to each
The history calling record of user is analyzed, and the concrete methods of realizing for establishing the time portrait of each user is described in detail.
Referring to Fig. 5, Fig. 5 shows a specific flow chart of step S13, details are as follows:
S131: according to the user identifier of user, the standard call recording of user is obtained from preset successful case library.
In the present embodiment, preset successful case library is the database for storing standard call recording, standard call
Recording refers to that the recording for the business exchange process that successfully service product is promoted or sold, each standard call recording correspond to
One user identifier, according to the user identifier of user, all standards that the user is obtained from preset successful case library are logical
Words recording.
S132: the standard call recording of user is analyzed using preset speech analysis model, establishes the use of user
Family baseline characteristic.
Specifically, the phonetic feature recorded using each standard call of preset speech analysis model extraction user,
In, phonetic feature is the parameter for describing the voice signal of user's sending, which includes but is not limited to word speed, sound
Tune, volume and speech frequency etc..
Processing of averaging is carried out to the phonetic feature of all standard call recording of user, the word speed for calculating user is equal
Value, tone mean value, volume mean value and speech frequency mean value etc., obtain the phonetic feature mean value of user, and the phonetic feature is equal
It is worth user's baseline characteristic as user, for describing the phonetic feature of the user in business transaction success.
S133: speech feature extraction is carried out to each history calling record of user, obtains each history calling record
User vocal feature.
Specifically, server-side carries out speech feature extraction to history calling record using preset speech analysis model, can
To obtain the user vocal feature of each history calling record.
S134: being compared and analyzed using user's baseline characteristic of user vocal feature and user, obtains user time note
Record, and drawn a portrait according to the time that user time record establishes user.
Specifically, the user vocal feature of each history calling record of user is obtained, and uses the call record of each history
The user vocal feature of sound and user's baseline characteristic of the user compare, according to preset grade form to history calling record
It gives a mark, obtains the score value of each history calling record, to determine and use according to the score value of each history calling record
Whether family facilitates carry out telephonic communication in the history calling record corresponding period, for example, the score value of history calling record
It is bigger, user the history calling record corresponding period can time for communication it is longer.
Wherein, which includes the mapping relations between the voice attributes and score value of history calling record, language
Sound attribute is the parameter for describing history calling record, which includes but is not limited to phonetic feature and long recording time,
Mapping relations between the voice attributes and score value of the history calling record can be configured according to the needs of actual conditions, this
Place is with no restrictions.
For example, in a specific embodiment, as shown in table 1, a preset history call is shown in the grade form of table 1
Mapping relations between the voice attributes and score value of recording:
1. grade form of table
| Score value | Voice attributes |
| +1 | Average word speed is faster than baseline word speed in calling record |
| -1 | Average word speed is slower than baseline word speed in calling record |
| +1 | Volume is greater than baseline volume in calling record |
| -1 | Volume is less than baseline volume in calling record |
| +1 | Volume is greater than baseline tone in calling record |
| -1 | Volume is less than baseline tone in calling record |
| -10 | The long recording time of calling record was less than 1 minute |
| +10 | The long recording time of calling record is greater than 10 minutes |
| …… | …… |
Wherein, if the score value of history calling record is positive, indicate user in the history calling record corresponding period
Facilitate carry out telephonic communication, and score value is bigger, indicate user can time for communication it is longer；If the scoring of history calling record
Value is negative, and indicates that user is inconvenient to carry out telephonic communication in the history calling record corresponding period.
It is analyzed by all history calling records to user, it is logical in history in different time periods to obtain the user
Talk about the score value of recording, and user time record obtained according to the score value of history calling record, so as to determine user it is convenient into
It the period of row telephonic communication, and the period of inconvenient telephonic communication, is recorded according to user time, sequentially in time structure
Build the time portrait whether user conveniently answers sales calls.
It is understood that since the job placement time of different industries and different users is different, so different
User between the rush hour it is also different, and analyzed, can be obtained by the history calling record to each user
The period of sales calls is conveniently answered to user, and is inconvenient to answer the period of sales calls, and user is constructed with this
User portrait, for describing the busy or idle period of user.
In the corresponding embodiment of Fig. 5, the standard call by obtaining user from preset successful case library is recorded, and is built
User's baseline characteristic of vertical user, to obtain the phonetic feature of user in transaction successful story, to each history of user
Calling record carries out speech feature extraction, obtains the user vocal feature of each history calling record, and special using user speech
Sign is compared and analyzed with user's baseline characteristic of user, according to the standard call of user record in difference, judge to use
The period of sales calls is conveniently answered at family, and is inconvenient to answer the period of sales calls, to obtain user time note
Record establish the time portrait of user, so that the sequence according to the time describes the busy or idle period of user, from
And obtain user can time for communication.
In one embodiment, the present embodiment to mentioned in step S22 using preset speech analysis model to every class
The history calling record that group identifies corresponding user is analyzed, and is established every types of populations and is identified corresponding group's time portrait
Concrete methods of realizing is described in detail.
Referring to Fig. 6, Fig. 6 shows a specific flow chart of step S22, details are as follows:
S221: identifying according to group, from preset successful case library, obtains the mark that every types of populations identifies corresponding user
Quasi- calling record.
In the present embodiment, preset successful case library is the database for storing standard call recording, standard call
Recording refers to that the recording for the business exchange process that successfully service product is promoted or sold, each standard call recording correspond to
At least one group mark, identifies according to group, and every types of populations is obtained from preset successful case library and identifies corresponding user
Standard call recording.
S222: according to preset speech analysis model, to the standard call that every types of populations identifies corresponding user record into
Row analysis establishes every types of populations and identifies corresponding group's baseline characteristic.
Specifically, it is recorded using the standard call that the preset every types of populations of speech analysis model extraction identifies corresponding user
Phonetic feature, wherein phonetic feature is the parameter for describing the voice signal of user's sending, which includes but not
It is limited to word speed, tone, volume and speech frequency etc..
The phonetic feature for identifying the standard call recording of corresponding user to every types of populations carries out processing of averaging, and calculates every
Word speed mean value, tone mean value, volume mean value and speech frequency mean value of user in types of populations etc., obtain the use in every types of populations
The phonetic feature mean value at family, and using phonetic feature mean value as group's baseline characteristic, so that it is corresponding to establish every types of populations mark
Group's baseline characteristic, for describing the phonetic feature of the user in business transaction success in every types of populations.
S223: the history calling record for identifying corresponding user to every types of populations carries out speech feature extraction, and use mentions
The phonetic feature got group's baseline characteristic corresponding with types of populations mark compares and analyzes, and obtains group's time record.
In the present embodiment, server-side identifies corresponding user to every types of populations using preset speech analysis model and goes through
History calling record carries out speech feature extraction, and available every types of populations identifies the user of the history calling record of corresponding user
Phonetic feature.
Specifically, it is carried out pair using the user vocal feature extracted group's baseline characteristic corresponding with types of populations mark
Than being given a mark according to preset grade form to history calling record, the score value of each history calling record being obtained, so as to root
According to the score value of each history calling record, determine that the user in the types of populations is in the history calling record corresponding period
It is no to facilitate carry out telephonic communication, obtain group's time record.For example, the score value of history calling record is bigger, user goes through at this
The history calling record corresponding period can time for communication it is longer.
Wherein, which includes the mapping relations between the voice attributes and score value of history calling record, language
Sound attribute is the parameter for describing history calling record, which includes but is not limited to phonetic feature and long recording time,
Mapping relations between the voice attributes and score value of the history calling record can be configured according to the needs of actual conditions, example
Such as, the concrete form of preset grade form may refer to table 1.
S224: corresponding group's time record is identified according to every types of populations, when constructing every corresponding group of types of populations mark
Between draw a portrait.
Specifically, corresponding group's time record is identified according to every types of populations, under the same time, if in the types of populations
User be in can time for communication number of users be greater than in the rush hour number of users, then by the period be labeled as the group
Body can time for communication；Otherwise, if user in the types of populations be in can time for communication number of users be less than in it is busy when
Between number of users, then by the period be labeled as the group the rush hour.And according to group can time for communication and group
The rush hour, establish sequentially in time every types of populations group's time portrait, to determine the user side in the types of populations
Just the time of sales calls is answered.
In the corresponding embodiment of Fig. 6, corresponding use is identified by from preset successful case library, obtaining every types of populations
The standard call at family is recorded, and is established every types of populations and is identified corresponding group's baseline characteristic, to obtain in transaction successful story
The phonetic feature of user in every types of populations, the history calling record for identifying corresponding user to every types of populations carry out phonetic feature
It extracts, and is compared and analyzed using the phonetic feature group's baseline characteristic corresponding with types of populations mark extracted, judged
The user in every types of populations conveniently answers the period of sales calls out, and is inconvenient to answer the period of sales calls, from
And the time portrait that group's time record establish group is obtained, so that the sequence according to the time describes the use in every types of populations
The busy or idle period at family, thus obtain the user in different groups can time for communication information.
In one embodiment, as described in Figure 7, after step S23, i.e., corresponding institute is identified using the group described
State population characteristic and group's time portrait construct the group and identifies corresponding group's portrait, and by group mark and
After the corresponding storage of group's portrait is into preset user information database, the Products Show method based on voice data is also
Include:
S31: obtaining the service attribute of product to be recommended, and in preset user information database, is distinguished using service attribute
It is matched with the population characteristic of the user characteristics of user's portrait and group's portrait.
In the present embodiment, product to be recommended refers to that the service product that company needs to promote, service attribute refer to due to retouching
The information of service product is stated, service attribute can specifically include but be not limited to product category, name of product and scope of application etc., example
Such as, the service attribute of product " 2017 old men danger " to be recommended includes " old man danger ", " insurance ", " the elderly " and " 60 years old or more " etc.
Service attribute.
Specifically, server-side obtains the service attribute of product to be recommended, in preset user information database, uses business category
Property matched respectively with the population characteristic of the user characteristics of user's portrait and group's portrait, inquiry matches with service attribute
Characteristic information.
S32: user's portrait corresponding with the user characteristics that service attribute matches is obtained, and is drawn a portrait and is determined according to the user
Target user.
Specifically, in preset user information database, if existing in the user characteristics of user's portrait identical as service attribute
User characteristics, then it represents that the user characteristics match with service attribute, obtain the user characteristics pair to match with service attribute
The user's portrait answered, and according to the corresponding user identifier of user's portrait, determine target user, wherein the target user can be
One user, is also possible to multiple users, can be determined according to the actual situation.
S33: group's portrait corresponding with the population characteristic that service attribute matches is obtained, and is drawn a portrait and is determined according to the group
Target group.
Specifically, in preset user information database, if existing in the population characteristic of group's portrait identical as service attribute
Population characteristic, then it represents that the population characteristic matches with service attribute, obtains the population characteristic pair to match with service attribute
The group's portrait answered, and identified according to the corresponding group of group's portrait, determine target group, wherein the target group can be
One group, is also possible to multiple groups, can be determined according to the actual situation.
S34: target user and target group are output to client.
Specifically, target user indicates to treat the interested user of recommended products, and target group indicate to treat recommended products
Target user and target group are output to client by interested group, so that business personnel can determine product to be recommended
User or crowd, and can according to target user and target group in preset user information database, obtain target use
Family it is corresponding can time for communication and target group it is corresponding can time for communication, so as to formulate corresponding marketing strategy, can
Time for communication to corresponding user recommend service product, improve the specific aim Generalization Ability of service product, targetedly for
Its service product interested is recommended at family, improves the popularization success rate of service product.
In the corresponding embodiment of Fig. 7, by obtaining the service attribute of product to be recommended, in preset user information database
In, it is matched respectively with the population characteristic of the user characteristics of user's portrait and group's portrait using service attribute, acquisition and industry
The corresponding user's portrait of the user characteristics that business attribute matches, draws a portrait according to the user and determines target user, so that determination is treated
The interested user of recommended products obtains group's portrait corresponding with the population characteristic that service attribute matches, and according to the group
Body, which is drawn a portrait, determines target group, determines and treats the interested group of recommended products, and target user and target group are exported
To client, so that business personnel can determine the user or crowd of product to be recommended, so as to formulate corresponding marketing
Strategy, can time for communication to corresponding user recommend service product, improve the specific aim Generalization Ability of service product, be directed to
Property recommend its service product interested for user, improve the popularization success rate of service product.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
In one embodiment, a kind of Products Show square law device based on voice data is provided, it should be based on voice data
Products Show method in Products Show square law device and above-described embodiment based on voice data corresponds.As shown in figure 8, should
Products Show square law device based on voice data includes: that voice data obtains module 811, fisrt feature obtains module 812, the
One data analysis module 813, first portrait memory module 814, first identifier obtain the portrait of module 815, first and obtain module 816
With the first result output module 817.Detailed description are as follows for each functional module:
Voice data obtains module 811, for obtaining the user identifier and history calling record of at least one user；
Fisrt feature obtains module 812, records for the history of each user to be conversed using preset speech recognition modeling
Sound is converted to text file, and carries out semantic analysis to the corresponding text file of each user, and the user for obtaining each user is special
Sign；
First data analysis module 813, for conversing the history of each user and recording according to preset speech analysis model
Sound is analyzed, and the time portrait of each user is established；
First portrait memory module 814, for using the user characteristics of user and user's picture of time portrait building user
Picture, and the user identifier of user and user are drawn a portrait into corresponding storage into preset user information database；
First identifier obtains module 815, for obtaining the user identifier of user to be linked up；
First portrait obtains module 816, if the user identifier for there is user to be linked up in preset user information database,
The corresponding user's portrait of user identifier for then obtaining user to be linked up, draws a portrait as target user；
First result output module 817 determines that first object produces for the user characteristics in drawing a portrait according to target user
Product, and according to target user draw a portrait in time draw a portrait, determine first can time for communication, and by first object product and first
Can time for communication be output to client.
Preferably, it is somebody's turn to do the Products Show device based on voice data further include:
Second feature obtains module 821, special in the user that every types of populations identifies corresponding user for being identified according to group
In sign, the quantity for choosing identical user characteristics is more than the user characteristics of preset amount threshold, and the user chosen is special
Sign is identified as group and identifies corresponding population characteristic；
Second data analysis module 822, for identifying corresponding use to every types of populations using preset speech analysis model
The history calling record at family is analyzed, and is established every types of populations and is identified corresponding group's time portrait；
Second portrait building module 823 identifies corresponding population characteristic and the portrait building of group's time for use groups
Group identifies corresponding group's portrait, and by the storage corresponding with group's portrait of group's mark into preset user information database；
Second identifier obtains module 824, if the user for user to be linked up to be not present in preset user information database marks
Know, then obtains group's mark of user to be linked up；
Second portrait obtains module 825, marks for obtaining from preset user information database with the group of user to be linked up
Know corresponding group's portrait, draws a portrait as target group；
Second result output module 826 determines that the second target produces for the population characteristic in drawing a portrait according to target group
Product, and according to target group draw a portrait in group's time draw a portrait, determine second can time for communication, and by the second target product and
Second can time for communication be output to client.
Preferably, fisrt feature acquisition module 812 includes:
Keyword extraction submodule, for using preset semantic analysis model, text file corresponding to each user
Keyword extraction is carried out, and semantic analysis is carried out to the keyword extracted, obtains target keyword；
User characteristics determine submodule, for using target keyword as the user characteristics of user.
Preferably, the first data analysis module 813 includes:
First voice acquisition submodule is obtained from preset successful case library and is used for the user identifier according to user
The standard call at family is recorded；
First baseline setting up submodule, for using preset speech analysis model to carry out the standard call recording of user
Analysis, establishes user's baseline characteristic of user；
First phonetic feature acquisition submodule carries out speech feature extraction for each history calling record to user,
Obtain the user vocal feature of each history calling record；
Portrait setting up submodule at the first time, for being carried out pair using user's baseline characteristic of user vocal feature and user
Than analysis, user time record is obtained, and is drawn a portrait according to the time that user time record establishes user.
Preferably, the second data analysis module 822 includes:
Second voice acquisition submodule from preset successful case library, obtains every types of populations for identifying according to group
Identify the standard call recording of corresponding user；
Second baseline setting up submodule, for identifying corresponding use to every types of populations according to preset speech analysis model
The standard call recording at family is analyzed, and is established every types of populations and is identified corresponding group's baseline characteristic；
Second phonetic feature acquisition submodule, the history calling record for identifying corresponding user to every types of populations carry out
Speech feature extraction, and compared point using phonetic feature group's baseline characteristic corresponding with types of populations mark for extracting
Analysis obtains group's time record；
Second time portrait setting up submodule, for identifying corresponding group's time record according to every types of populations, building is often
Types of populations identifies corresponding group's time portrait.
Preferably, it is somebody's turn to do the Products Show device based on voice data further include:
Product information matching module, for obtaining the service attribute of product to be recommended, and in preset user information database,
It is matched respectively with the population characteristic of the user characteristics of user's portrait and group's portrait using service attribute；
First object establishes module, and the corresponding user of user characteristics for obtaining with service attribute matches draws a portrait, and
It is drawn a portrait according to the user and determines target user；
Second goal module, the corresponding group's portrait of population characteristic for obtaining with service attribute matches, and
It is drawn a portrait according to the group and determines target group；
Target information output module, for target user and target group to be output to client.
Specific restriction about the Products Show square law device based on voice data may refer to above for based on language
The restriction of the Products Show method of sound data, details are not described herein.In the above-mentioned Products Show square law device based on voice data
Modules can be realized fully or partially through software, hardware and combinations thereof.Above-mentioned each module can embed in the form of hardware
In processor in computer equipment, the memory that can also be stored in a software form in computer equipment
In, the corresponding operation of the above modules is executed in order to which processor calls.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 9.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The network interface of machine equipment is used to communicate with external terminal by network connection.When the computer program is executed by processor with
Realize a kind of Products Show method based on voice data.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor are realized in above-described embodiment when executing computer program based on voice
Step in the Products Show method of data, such as step S11 shown in Fig. 2 to step S17, alternatively, processor executes calculating
The function of each module of the Products Show square law device in above-described embodiment based on voice data, such as Fig. 8 are realized when machine program
Shown module 811 to module 817 function.To avoid repeating, which is not described herein again.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program realizes the step in the Products Show method in above-described embodiment based on voice data when being executed by processor, such as schemes
Step S11 shown in 2 to step S17, alternatively, processor is realized in above-described embodiment when executing computer program based on voice number
According to Products Show square law device each module function, such as module 811 shown in Fig. 8 is to the function of module 817.To avoid weight
Multiple, which is not described herein again.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of Products Show method based on voice data, which is characterized in that the Products Show side based on voice data
Method includes:
Obtain the user identifier and history calling record of at least one user；
The history calling record of each user is converted to by text file using preset speech recognition modeling, and right
The corresponding text file of each user carries out semantic analysis, obtains the user characteristics of each user；
According to preset speech analysis model, the history calling record of each user is analyzed, is established each
The time of the user draws a portrait；
It is drawn a portrait using the user that the user characteristics of the user and time portrait construct the user, and by the use
The user identifier at family and the user draw a portrait corresponding storage into preset user information database；
Obtain the user identifier of user to be linked up；
If there is the user identifier of the user to be linked up in the preset user information database, the user to be linked up is obtained
User identifier corresponding user portrait, draw a portrait as target user；
According to the user characteristics in target user portrait, first object product is determined, and used according to the target
Time portrait in the portrait of family, determine first can time for communication, and can ditch by the first object product and described first
The logical time is output to client.
2. the Products Show method based on voice data as described in claim 1, which is characterized in that the user is corresponding at least
One types of populations mark, described according to preset speech analysis model, to the history calling record of each user into
Row analysis is established after the time portrait of each user, and before the user identifier for obtaining user to be linked up,
The Products Show method based on voice data further include:
It is identified according to the group, the group described in every class identifies in the user characteristics of the corresponding user, chooses phase
The quantity of the same user characteristics is more than the user characteristics of preset amount threshold, and the user chosen is special
Sign is identified as the group and identifies corresponding population characteristic；
It is conversed using the history that preset speech analysis model group described in every class identifies the corresponding user
Recording is analyzed, and is established group described in every class and is identified corresponding group's time portrait；
The corresponding population characteristic is identified using the group and group's time portrait constructs group's mark and corresponds to
Group portrait, and by the group mark it is corresponding with group portrait store into the preset user information database；
After the user identifier for obtaining user to be linked up, the Products Show method based on voice data further include:
If there is no the user identifiers of the user to be linked up in the preset user information database, obtain described general to ditch
The group at family identifies；
Group's portrait corresponding with the group of the user to be linked up mark is obtained from the preset user information database, as
Target group's portrait；
According to the population characteristic in target group portrait, the second target product is determined, and according to the target complex
Group's time portrait in body portrait, determine second can time for communication, and by second target product and described second
Can time for communication be output to the client.
3. the Products Show method based on voice data as described in claim 1, which is characterized in that described to each use
The corresponding text file in family carries out semantic analysis, and the user characteristics for obtaining each user include:
Using preset semantic analysis model, keyword extraction is carried out to the corresponding text file of each user, and
Semantic analysis is carried out to the keyword extracted, obtains target keyword；
Using the target keyword as the user characteristics of the user.
4. the Products Show method based on voice data as described in claim 1, which is characterized in that described according to preset language
Sound analysis model analyzes the history calling record of each user, and the time for establishing each user draws
As including:
According to the user identifier of the user, the standard call record of the user is obtained from preset successful case library
Sound；
The standard call recording of the user is analyzed using the preset speech analysis model, establishes the user's
User's baseline characteristic；
Speech feature extraction is carried out to each of the user history calling record, obtains each history calling record
User vocal feature；
It is compared and analyzed using the user vocal feature and user's baseline characteristic of the user, obtains user time
Record, and drawn a portrait according to the time that user time record establishes the user.
5. the Products Show method based on voice data as claimed in claim 2, which is characterized in that described using described default
Speech analysis model group described in every class identify the history calling record of the corresponding user and analyze, establish
Group described in every class identifies corresponding group's time portrait
It is identified according to the group, from the preset successful case library, it is corresponding described to obtain the mark of group described in every class
The standard call of user is recorded；
According to the preset speech analysis model, the group described in every class identifies the standard call of the corresponding user
Recording is analyzed, and is established group described in every class and is identified corresponding group's baseline characteristic；
The history calling record that the group described in every class identifies the corresponding user carries out speech feature extraction, and uses
The phonetic feature extracted group's baseline characteristic corresponding with types of populations mark compares and analyzes, and obtains group's time note
Record；
Corresponding group's time record is identified according to group described in every class, group described in every class is constructed and identifies corresponding group's time
Portrait.
6. the Products Show method based on voice data as claimed in claim 2, which is characterized in that use the group described
Body identifies the corresponding population characteristic and group's time portrait constructs the group and identifies corresponding group's portrait, and will
Group mark is corresponding with group portrait store into preset user information database after, it is described based on voice data
Products Show method further include:
The service attribute of product to be recommended is obtained, and in the preset user information database, is distinguished using the service attribute
It is matched with the population characteristic of the user characteristics of user portrait and group portrait；
User's portrait corresponding with the user characteristics that the service attribute matches is obtained, and is drawn a portrait according to the user and determines target
User；
Group's portrait corresponding with the population characteristic that the service attribute matches is obtained, and is drawn a portrait according to the group and determines target
Group；
The target user and the target group are output to the client.
7. a kind of Products Show device based on voice data, which is characterized in that the Products Show device packet of the voice data
It includes:
Voice data obtains module, for obtaining the user identifier and history calling record of at least one user；
Fisrt feature obtains module, records for the history of each user to be conversed using preset speech recognition modeling
Sound is converted to text file, and carries out semantic analysis to the corresponding text file of each user, obtains each described
The user characteristics of user；
First data analysis module, for conversing the history of each user according to preset speech analysis model
Recording is analyzed, and the time portrait of each user is established；
First portrait memory module, the user characteristics and time portrait for using the user construct the user
User's portrait, and preset user information database is arrived into the user identifier of the user and the user corresponding storage of drawing a portrait
In；
First identifier obtains module, for obtaining the user identifier of user to be linked up；
First portrait obtains module, if user's mark for there is the user to be linked up in the preset user information database
Know, then obtains the corresponding user's portrait of user identifier of the user to be linked up, draw a portrait as target user；
First result output module, for determining that first object is produced according to the user characteristics in target user portrait
Product, and being drawn a portrait according to the time in target user portrait, determine first can time for communication, and by first mesh
Mark product and described first can time for communication be output to client.
8. the Products Show device based on voice data as claimed in claim 7, which is characterized in that described to be based on voice data
Products Show device further include:
Second feature obtains module, and for being identified according to the group, the group described in every class identifies the corresponding user's
In the user characteristics, the quantity for choosing the identical user characteristics is more than the user characteristics of preset amount threshold,
And the user characteristics chosen are identified as the group and identify corresponding population characteristic；
Second data analysis module, for identifying corresponding institute using preset speech analysis model group described in every class
The history calling record for stating user is analyzed, and is established group described in every class and is identified corresponding group's time portrait；
Second portrait building module, for being drawn a portrait using the group corresponding population characteristic of mark and group's time
It constructs the group and identifies corresponding group's portrait, and the group is identified into store to described pre- corresponding with group portrait
If user information database in；
Second identifier obtains module, if for user's mark in the preset user information database there is no the user to be linked up
Know, then obtains group's mark of the user to be linked up；
Second portrait obtains module, marks for obtaining from the preset user information database with the group of the user to be linked up
Know corresponding group's portrait, draws a portrait as target group；
Second result output module, for determining that the second target is produced according to the population characteristic in target group portrait
Product, and being drawn a portrait according to group's time in target group portrait, determine second can time for communication, and by described the
Two target products and described second can time for communication be output to the client.
9. a kind of computer equipment, including memory, processor and storage are in the memory and can be in the processor
The computer program of upper operation, which is characterized in that the processor realized when executing the computer program as claim 1 to
Products Show method described in 6 any one based on voice data.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In realization is as described in any one of claim 1 to 6 based on the product of voice data when the computer program is executed by processor
Recommended method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910207372.4A CN110110321A (en) | 2019-03-19 | 2019-03-19 | Products Show method, apparatus, equipment and storage medium based on voice data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910207372.4A CN110110321A (en) | 2019-03-19 | 2019-03-19 | Products Show method, apparatus, equipment and storage medium based on voice data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110110321A true CN110110321A (en) | 2019-08-09 |
Family
ID=67484366
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910207372.4A Pending CN110110321A (en) | 2019-03-19 | 2019-03-19 | Products Show method, apparatus, equipment and storage medium based on voice data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110321A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110782318A (en) * | 2019-10-21 | 2020-02-11 | 五竹科技(天津)有限公司 | Marketing method and device based on audio interaction and storage medium |
| CN110990545A (en) * | 2019-11-28 | 2020-04-10 | 重庆锐云科技有限公司 | Artificial intelligent telephone customer service expansion marketing management system and method |
| CN111177541A (en) * | 2019-12-20 | 2020-05-19 | 上海淇玥信息技术有限公司 | Data analysis method and device based on user tag generation time, server and storage medium |
| CN111402931A (en) * | 2020-03-05 | 2020-07-10 | 云知声智能科技股份有限公司 | Voice boundary detection method and system assisted by voice portrait |
| CN111614843A (en) * | 2020-04-09 | 2020-09-01 | 深圳追一科技有限公司 | Call configuration determining method and device, computer equipment and storage medium |
| CN111667316A (en) * | 2020-06-16 | 2020-09-15 | 中国银行股份有限公司 | Service processing method, device and system |
| CN111949776A (en) * | 2020-07-17 | 2020-11-17 | 上海淇馥信息技术有限公司 | Method and device for evaluating user tag and electronic equipment |
| CN112364203A (en) * | 2020-11-30 | 2021-02-12 | 未来电视有限公司 | Television video recommendation method, device, server and storage medium |
| CN112487039A (en) * | 2020-12-16 | 2021-03-12 | 平安养老保险股份有限公司 | Data processing method, device and equipment and readable storage medium |
| CN112734458A (en) * | 2020-12-25 | 2021-04-30 | 珠海大横琴科技发展有限公司 | Method and device for inviting business, electronic equipment and readable storage medium |
| CN112967721A (en) * | 2021-02-03 | 2021-06-15 | 上海明略人工智能(集团)有限公司 | Sales lead information identification method and system based on voice identification technology |
| CN113094578A (en) * | 2021-03-16 | 2021-07-09 | 平安普惠企业管理有限公司 | Deep learning-based content recommendation method, device, equipment and storage medium |
| CN113127751A (en) * | 2019-12-30 | 2021-07-16 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device and equipment and computer readable storage medium |
| CN113158052A (en) * | 2021-04-23 | 2021-07-23 | 平安银行股份有限公司 | Chat content recommendation method and device, computer equipment and storage medium |
| CN113283934A (en) * | 2021-05-28 | 2021-08-20 | 北京量冠科技有限公司 | Information determination method and device, electronic equipment and storage medium |
| CN113438374A (en) * | 2021-07-27 | 2021-09-24 | 中国平安人寿保险股份有限公司 | Intelligent outbound call processing method, device, equipment and storage medium |
| CN115098793A (en) * | 2022-03-30 | 2022-09-23 | 陈应书 | User portrait analysis method and system based on big data |
| CN117033801A (en) * | 2023-10-08 | 2023-11-10 | 太平金融科技服务（上海）有限公司 | Service recommendation method, device, equipment and storage medium |
| CN117440091A (en) * | 2023-09-22 | 2024-01-23 | 杭州东岸网络信息服务有限公司 | Voice data processing method and call control method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017000623A1 (en) * | 2015-07-02 | 2017-01-05 | 中兴通讯股份有限公司 | Information recommendation method and apparatus |
| CN108366287A (en) * | 2018-01-30 | 2018-08-03 | 广州虎牙信息科技有限公司 | Direct broadcasting room action message methods of exhibiting and computer storage media, terminal |
-
2019
- 2019-03-19 CN CN201910207372.4A patent/CN110110321A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017000623A1 (en) * | 2015-07-02 | 2017-01-05 | 中兴通讯股份有限公司 | Information recommendation method and apparatus |
| CN108366287A (en) * | 2018-01-30 | 2018-08-03 | 广州虎牙信息科技有限公司 | Direct broadcasting room action message methods of exhibiting and computer storage media, terminal |
Non-Patent Citations (2)
| Title |
|---|
| SHUAI YANG ET AL.: "1+N fusion: Cascaded self-portrait enhancement", 《2017 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)》 * |
| 傅军 等: "电力用户行为画像构建技术研究", 《电气应用》 * |
Cited By (27)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110782318A (en) * | 2019-10-21 | 2020-02-11 | 五竹科技(天津)有限公司 | Marketing method and device based on audio interaction and storage medium |
| CN110990545A (en) * | 2019-11-28 | 2020-04-10 | 重庆锐云科技有限公司 | Artificial intelligent telephone customer service expansion marketing management system and method |
| CN111177541A (en) * | 2019-12-20 | 2020-05-19 | 上海淇玥信息技术有限公司 | Data analysis method and device based on user tag generation time, server and storage medium |
| CN111177541B (en) * | 2019-12-20 | 2023-08-22 | 上海淇玥信息技术有限公司 | Data analysis method and device based on user tag generation time |
| CN113127751A (en) * | 2019-12-30 | 2021-07-16 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device and equipment and computer readable storage medium |
| CN113127751B (en) * | 2019-12-30 | 2023-10-27 | 中移(成都)信息通信科技有限公司 | User portrait construction method, device, equipment and computer readable storage medium |
| CN111402931A (en) * | 2020-03-05 | 2020-07-10 | 云知声智能科技股份有限公司 | Voice boundary detection method and system assisted by voice portrait |
| CN111614843A (en) * | 2020-04-09 | 2020-09-01 | 深圳追一科技有限公司 | Call configuration determining method and device, computer equipment and storage medium |
| CN111667316A (en) * | 2020-06-16 | 2020-09-15 | 中国银行股份有限公司 | Service processing method, device and system |
| CN111667316B (en) * | 2020-06-16 | 2023-11-10 | 中国银行股份有限公司 | Service processing method, device and system |
| CN111949776A (en) * | 2020-07-17 | 2020-11-17 | 上海淇馥信息技术有限公司 | Method and device for evaluating user tag and electronic equipment |
| CN111949776B (en) * | 2020-07-17 | 2023-09-22 | 上海淇馥信息技术有限公司 | User tag evaluation method and device and electronic equipment |
| CN112364203B (en) * | 2020-11-30 | 2023-04-28 | 未来电视有限公司 | Television video recommendation method, device, server and storage medium |
| CN112364203A (en) * | 2020-11-30 | 2021-02-12 | 未来电视有限公司 | Television video recommendation method, device, server and storage medium |
| CN112487039A (en) * | 2020-12-16 | 2021-03-12 | 平安养老保险股份有限公司 | Data processing method, device and equipment and readable storage medium |
| CN112487039B (en) * | 2020-12-16 | 2024-04-30 | 平安养老保险股份有限公司 | Data processing method, device, equipment and readable storage medium |
| CN112734458A (en) * | 2020-12-25 | 2021-04-30 | 珠海大横琴科技发展有限公司 | Method and device for inviting business, electronic equipment and readable storage medium |
| CN112967721A (en) * | 2021-02-03 | 2021-06-15 | 上海明略人工智能(集团)有限公司 | Sales lead information identification method and system based on voice identification technology |
| CN112967721B (en) * | 2021-02-03 | 2024-05-31 | 上海明略人工智能(集团)有限公司 | Sales lead information recognition method and system based on voice recognition technology |
| CN113094578A (en) * | 2021-03-16 | 2021-07-09 | 平安普惠企业管理有限公司 | Deep learning-based content recommendation method, device, equipment and storage medium |
| CN113158052A (en) * | 2021-04-23 | 2021-07-23 | 平安银行股份有限公司 | Chat content recommendation method and device, computer equipment and storage medium |
| CN113283934A (en) * | 2021-05-28 | 2021-08-20 | 北京量冠科技有限公司 | Information determination method and device, electronic equipment and storage medium |
| CN113438374A (en) * | 2021-07-27 | 2021-09-24 | 中国平安人寿保险股份有限公司 | Intelligent outbound call processing method, device, equipment and storage medium |
| CN115098793A (en) * | 2022-03-30 | 2022-09-23 | 陈应书 | User portrait analysis method and system based on big data |
| CN117440091A (en) * | 2023-09-22 | 2024-01-23 | 杭州东岸网络信息服务有限公司 | Voice data processing method and call control method |
| CN117033801A (en) * | 2023-10-08 | 2023-11-10 | 太平金融科技服务（上海）有限公司 | Service recommendation method, device, equipment and storage medium |
| CN117033801B (en) * | 2023-10-08 | 2023-12-22 | 太平金融科技服务（上海）有限公司 | Service recommendation method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110321A (en) | Products Show method, apparatus, equipment and storage medium based on voice data | |
| CN109767787B (en) | Emotion recognition method, device and readable storage medium | |
| CN104598445B (en) | Automatically request-answering system and method | |
| US20230334034A1 (en) | Systems and methods for providing searchable customer call indexes | |
| CN109767318A (en) | Loan product recommended method, device, equipment and storage medium | |
| CN108388558B (en) | Question matching method and device, customer service robot and storage medium | |
| CN107886949A (en) | A kind of content recommendation method and device | |
| US12002454B2 (en) | Method and apparatus for intent recognition and intent prediction based upon user interaction and behavior | |
| CN110516057B (en) | Petition question answering method and device | |
| CN111882431A (en) | Intelligent message pushing method based on NLP deep learning | |
| CN113297365B (en) | User intention judging method, device, equipment and storage medium | |
| CN114936727A (en) | Work order distribution system, method and computer equipment | |
| CN110347810A (en) | Method, apparatus, computer equipment and storage medium are answered in dialog mode retrieval | |
| CN111798118B (en) | Enterprise operation risk monitoring method and device | |
| CN113342948A (en) | Intelligent question and answer method and device | |
| CN111881273A (en) | Question and answer processing method and device | |
| CN115017289A (en) | Method and device for serving customers based on knowledge base | |
| CN112860873A (en) | Intelligent response method, device and storage medium | |
| CN111309882B (en) | Method and device for realizing intelligent customer service question and answer | |
| CN111916110A (en) | Voice quality inspection method and device | |
| CN112287091A (en) | Intelligent question-answering method and related products | |
| CN115687754B (en) | Active network information mining method based on intelligent dialogue | |
| CN113190660A (en) | Questionnaire survey method and device | |
| CN113111157B (en) | Question-answer processing method, device, computer equipment and storage medium | |
| CN114254088A (en) | Method for constructing automatic response model and automatic response method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190809 |