CN110163630A - Product monitoring and managing method, device, computer equipment and storage medium - Google Patents
Product monitoring and managing method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN110163630A CN110163630A CN201910300383.7A CN201910300383A CN110163630A CN 110163630 A CN110163630 A CN 110163630A CN 201910300383 A CN201910300383 A CN 201910300383A CN 110163630 A CN110163630 A CN 110163630A
- Authority
- CN
- China
- Prior art keywords
- voice
- blacklist
- semantic analysis
- product
- vocal print
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012544 monitoring process Methods 0.000 title claims abstract description 46
- 238000003860 storage Methods 0.000 title claims abstract description 23
- 230000001755 vocal effect Effects 0.000 claims abstract description 113
- 238000004458 analytical method Methods 0.000 claims abstract description 96
- 235000021167 banquet Nutrition 0.000 claims abstract description 62
- 238000004519 manufacturing process Methods 0.000 claims abstract description 44
- 230000004044 response Effects 0.000 claims abstract description 25
- 230000001105 regulatory effect Effects 0.000 claims description 27
- 238000005516 engineering process Methods 0.000 claims description 26
- 238000004590 computer program Methods 0.000 claims description 17
- 238000012790 confirmation Methods 0.000 claims description 14
- 238000012423 maintenance Methods 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 10
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 101000741965 Homo sapiens Inactive tyrosine-protein kinase PRAG1 Proteins 0.000 claims description 5
- 102100038659 Inactive tyrosine-protein kinase PRAG1 Human genes 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000005266 casting Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 239000000047 product Substances 0.000 description 125
- 238000012795 verification Methods 0.000 description 11
- 238000003058 natural language processing Methods 0.000 description 8
- 238000009826 distribution Methods 0.000 description 5
- 235000013399 edible fruits Nutrition 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009432 framing Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000002860 competitive effect Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000006227 byproduct Substances 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
-
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/42—Confirmation, e.g. check or permission by the legal debtor of payment
-
- 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
- G06Q30/018—Certifying business or products
- G06Q30/0185—Product, service or business identity fraud
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Computer Security & Cryptography (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a kind of product monitoring and managing method, device, computer equipment and storage medium, and this method includes the production payment request for obtaining user terminal and sending；The mark that will attend a banquet is matched with blacklist mark, obtains list matching result；If list matching result is successful match, corresponding product type is determined according to product identification, corresponding voice broadcast data are called according to product type, voice broadcast is carried out to voice broadcast data；To be identified voice data of the user terminal based on voice broadcast data feedback is obtained, semantic analysis is carried out to voice data to be identified, obtains semantic analysis result；Target vocal print feature is extracted from voice data to be identified, standard vocal print feature and target vocal print feature are matched, and obtains voice print matching result；According to semantic analysis result and vocal print matching result, corresponding response is carried out to production payment request and is operated, there are frauds in product sales process to solve the problems, such as to attend a banquet.
Description
Technical field
The present invention relates to technical field of data processing more particularly to a kind of product monitoring and managing method, device, computer equipment and
Storage medium.
Background technique
With the development of the global economy, each major company is increasingly competitive, team, intra-company or sales force mutually it
Between there is also competitive relations.Due to corporate supervision deficiency, some sales forces are to promote achievement, in product sales process,
User's purchase product is induced by deceptive information etc. or some Team Managements personnel are the sales achievement for guaranteeing this team, it is hidden
It hides the fraud of sales force from, is complained to will lead to user, influence company reputation, therefore, sale how to be avoided to sit
The fraud of seat becomes urgent problem to be solved.
Summary of the invention
The embodiment of the present invention provides a kind of product monitoring and managing method, device, computer equipment and storage medium, to solve to attend a banquet
There are problems that fraud in product sales process.
A kind of product monitoring and managing method, comprising:
Obtain the production payment request that user terminal is sent, production payment request comprising user identifier, product identification and
It attends a banquet mark；
The mark of attending a banquet is matched with blacklist mark in blacklist list, obtains list matching result；
If the list matching result is successful match, corresponding product type, root are determined according to the product identification
Corresponding voice broadcast data are called according to the product type, voice is carried out to the voice broadcast data using TTS technology and is broadcast
Report；
To be identified voice data of the user terminal based on the voice broadcast data feedback is obtained, to the voice number to be identified
According to semantic analysis is carried out, semantic analysis result is obtained；
Target vocal print feature is extracted from the voice data to be identified, it will standard vocal print corresponding with the user identifier
Feature and the target vocal print feature are matched, and voice print matching result is obtained；
According to the semantic analysis result and the voice print matching as a result, carrying out corresponding sound to production payment request
It should operate.
A kind of product maintenance device, comprising:
Payment request obtains module, for obtaining the production payment request of user terminal transmission, the production payment request packet
Containing user identifier, product identification and mark of attending a banquet；
List matching module is obtained for matching the mark of attending a banquet with blacklist mark in blacklist list
List matching result；
Voice broadcast module determines if being successful match for the list matching result according to the product identification
Corresponding product type is called corresponding voice broadcast data according to the product type, is broadcast using TTS technology to the voice
Count off is according to progress voice broadcast；
Semantic analysis result obtains module, for obtaining to be identified language of the user terminal based on the voice broadcast data feedback
Sound data carry out semantic analysis to the voice data to be identified, obtain semantic analysis result；
Voice print matching result obtains module, will be with for extracting target vocal print feature from the voice data to be identified
The corresponding standard vocal print feature of user identifier and the target vocal print feature are matched, and voice print matching result is obtained；
Operation module is responded, is used for according to the semantic analysis result and the voice print matching as a result, to the product branch
It pays request and carries out corresponding response operation.
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 said goods monitoring and managing method when executing the computer program.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program realizes the said goods monitoring and managing method when being executed by processor.
It is above-mentioned that a kind of product monitoring and managing method, device, computer equipment and storage medium are provided, when obtaining what user terminal was sent
When production payment is requested, the mark that will attend a banquet is matched with blacklist mark in blacklist list, to determine whether to the product
Payment request carries out speech verification.If list matching result is successful match, determine that corresponding voice is broadcast according to product identification
Voice broadcast data are carried out voice broadcast by count off evidence, to determine whether user has a clear understanding of the product purchase of the product of purchase
Notice.To be identified voice data of the user terminal based on voice broadcast data feedback is obtained, voice data to be identified is carried out semantic
Analysis, to get whether user has a clear understanding of product purchase it should be clear that reducing subsequent user complaint.From voice data to be identified
Target vocal print feature is extracted, standard vocal print feature corresponding with user identifier and target vocal print feature are matched, acquisition sound
Line matching result, realization determine whether for my reply.According to semantic analysis result and vocal print matching result, production payment is asked
It asks and carries out corresponding response operation, realize the supervision of product sale.
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 the application environment schematic diagram of product monitoring and managing method in one embodiment of the invention；
Fig. 2 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 3 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 4 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 5 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 6 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 7 is the flow chart of product monitoring and managing method in one embodiment of the invention；
Fig. 8 is the functional block diagram of product maintenance device in one embodiment of the invention；
Fig. 9 is a schematic diagram of computer equipment in one embodiment of the invention.
Specific embodiment
Below by the attached drawing in knot and the embodiment of the present invention, 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.
Product monitoring and managing method provided in an embodiment of the present invention, can be applicable in the application environment such as Fig. 1, the product monitoring party
Method is applied in product distribution system, which includes user terminal and server-side, wherein user terminal by network with
Server-side is communicated.The product monitoring and managing method is particularly applicable in the server-side of Products Show APP, is sent out when receiving user terminal
When the production payment request sent, judge that selling attending a banquet for the product identifies whether to match with blacklist mark, if matching
Corresponding voice broadcast data are broadcasted to user terminal, to understand product purchase it should be clear that avoiding according to voice broadcast data user
It attends a banquet violation operation.According to the voice data to be identified of client feeds back, corresponding response operation is carried out, subsequent user is reduced and throws
It tells.The corresponding target vocal print feature of voice data to be identified is extracted, according to target vocal print feature to determine whether for user
It replys.Wherein, user terminal can be, but not limited to various personal computers, laptop, smart phone, tablet computer and portable
Formula wearable device.Server-side can be realized with the server-side cluster of the either multiple server-side compositions of independent server-side.
In one embodiment, as shown in Fig. 2, providing a kind of product monitoring and managing method, the service in Fig. 1 is applied in this way
It is illustrated, specifically comprises the following steps: for end
S10: obtain user terminal send production payment request, production payment request comprising user identifier, product identification and
It attends a banquet mark.
Wherein, production payment request is the request paid to the product of purchase that user is sent based on user terminal.With
Family mark refers to that server-side is the unique identification of each user distribution, is also possible to the corresponding mark of identification card number user of user
Know, unique user can determine by user identifier.Product identification is the corresponding mark of product of user's purchase, according to product mark
Know and determines unique product.Wherein, server-side is the corresponding product identification of each products configuration in advance, and each product identification is corresponding
One product type, product type refer to classification belonging to the product identification, for example, the corresponding product of product identification be good luck with
Property, affiliated product type is casualty insurance.Mark of attending a banquet refers to server-side in advance and is the mark of distribution of attending a banquet, and passes through mark of attending a banquet
Knowledge, which can determine, uniquely attends a banquet.
Specifically, product sale, after user determines purchase product, user Ke Ji can be carried out by product distribution system by attending a banquet
Production payment request is sent in user terminal, includes user identifier, product identification and mark of attending a banquet in production payment request.
S20: the mark that will attend a banquet is matched with blacklist mark in blacklist list, obtains list matching result.
Wherein, blacklist mark refers to the corresponding mark of attending a banquet for carrying out violation operation.
Specifically, it is stored with blacklist list in database, has each blacklist identify, is corresponding in blacklist list
Monitor duration and corresponding monitoring time limit.The server-side each blacklist mark in mark and blacklist list that will attend a banquet carries out
Match, obtains matching result.Wherein, matching result may include attend a banquet mark and blacklist mark successful match and attending a banquet identify with it is black
List mark it fails to match two kinds.If mark of attending a banquet and blacklist identify successful match, server-side needs to grasp this payment
Make to carry out IVR speech verification, wherein IVR (Interactive Voice Response), i.e. interactive voice answering.It realizes
Supervision to product sale.If it fails to match for mark of attending a banquet and blacklist mark, user can directly carry out delivery operation, be not necessarily to
Carry out IVR speech verification.By IVR speech verification, to determine whether user understands product purchase it should be clear that and guaranteeing to reply confirmation
The user of information is purchase product.
S30: if list matching result is successful match, corresponding product type is determined according to product identification, according to production
Category type calls corresponding voice broadcast data, carries out voice broadcast to voice broadcast data using TTS technology.
Wherein, voice broadcast data refer to preconfigured lteral data.It is to be appreciated that each product type corresponding one
Voice broadcast data, by voice broadcast data user will be clearly understood that product buy notice it is noted that data, i.e. product
Purchase is it should be clear that avoid violation operation of attending a banquet.Wherein, voice broadcast data should be simple and clear clear.
Specifically, if attending a banquet mark and blacklist mark successful match, needing to attend a banquet to this identifies corresponding payment request
Carry out IVR speech verification.Firstly, inquiring database according to product identification, it is corresponding to be stored with each product identification in database
Product type, product type includes plus protects, comprehensive golden loan/loan is small to disappear, activates first group with card/is turned over dozen and other types etc..For example,
Product identification corresponding product is good luck retinue, according to the table of comparisons in the casual lookup database of good luck, with the casual corresponding production of good luck
Category type is casualty insurance.Corresponding voice broadcast data are called according to product type.For example, (1) product type is unexpected protects
Danger, corresponding voice broadcast data are as follows: " whether do you understand that you currently buy a casualty insurance, rather than income type insures? it is clear
Chu's R. S. V. P. is clear, and the R. S. V. P. that such as has a question has a question, and thanks." (2) product type is that loan/loan is small disappears for comprehensive gold, it is corresponding
Voice broadcast data are as follows: " whether you understand that other synthesis financial services such as the insurance that you are currently bought and loan have no directly
Association? understand that R. S. V. P. is clear, the R. S. V. P. that such as has a question has a question, and thanks." (3) product type be credit card activate first group/
It turns over and beats, corresponding voice broadcast data are as follows: " whether you understand that insurance that you are currently bought and credit card amount promote etc. other
Does comprehensive financial service have no direct correlation? understand that R. S. V. P. is clear, the R. S. V. P. that such as has a question has a question, and thanks." according to product
Type search database, obtains corresponding with product type voice broadcast data, and use TTS technology by voice broadcast data into
Row voice broadcast will carry out IVR speech verification, and avoid the violation operation attended a banquet, and realize the supervision of product sale.
S40: obtain to be identified voice data of the user terminal based on voice broadcast data feedback, to voice data to be identified into
Row semantic analysis obtains semantic analysis result.
Wherein, voice data to be identified refers to voice number of the user based on user terminal according to voice broadcast data feedback
According to.
Specifically, server-side provides a voice and obtains interface, which obtains interface and user terminal is connected to the network, Yong Huke
Based on user terminal, voice data to be identified is fed back to server-side according to voice broadcast data, server-side obtains interface by voice
The voice data to be identified is got, and semantic analysis is carried out to voice data to be identified, is determined in voice data to be identified
Key content obtains semantic analysis result, wherein semantic analysis result includes confirmation message, denies information and uncertain information.
It is to be appreciated that comprising " understanding R. S. V. P. in the corresponding voice broadcast data of each product type in voice broadcast data
Clear, the R. S. V. P. that such as has a question has a question " voice broadcast data, i.e., the voice data to be identified that user replys includes clearly
Chu, i.e. semantic analysis result are confirmation message, include to have a question in the voice data to be identified that user replys, then semantic analysis knot
Fruit is to deny information；Understand and had a question if not including in the voice data to be identified that user replys, then semantic analysis knot
Fruit is uncertain information.
S50: extracting target vocal print feature from voice data to be identified, will standard vocal print feature corresponding with user identifier
It is matched with target vocal print feature, obtains voice print matching result.
Wherein, target vocal print feature refers to the feature extracted from voice data to be identified, and MFCC calculation specifically can be used
Method extracts target vocal print feature from voice data to be identified.Standard vocal print feature refers to the corresponding with user identifier of preparatory typing
Vocal print feature.Wherein, MFCC (Mel-scale Frequency Cepstral Coefficients, mel cepstrum coefficients)
It is characterized in the cepstrum parameter extracted in Mel scale frequency domain, Mel scale describes the nonlinear characteristic of human ear frequency, adopts
Vocal print feature extraction is carried out to voice data to be identified with MFCC algorithm, the MFCC feature got is target vocal print feature.
Specifically, target vocal print feature is extracted first from voice data to be identified, and is searched according to user identifier is corresponding
Database obtains corresponding with user identifier standard vocal print feature, using similarity algorithm by target vocal print feature and standard sound
Line feature carries out similarity calculation, obtains voice print matching result according to calculated vocal print similarity, wherein voice print matching result
Including successful match and it fails to match.
Further, in the database by the storage of voice broadcast data, user identifier and voice data to be identified, i.e., from adopting
Voice broadcast is carried out to voice broadcast data with TTS technology to start, carries out voice admission, until the voice to be identified that user sends
End of Tape after data saves the voice data of admission.If subsequent user pair product corresponding with product identification into
When row is complained, data base call voice data to be identified can be inquired and determine whether it is purchase product, user is avoided to cheat row
For to guarantee the validity of voice data to be identified.
S60: according to semantic analysis result and vocal print matching result, corresponding response is carried out to production payment request and is operated.
Specifically, if semantic analysis result is confirmation message, illustrate that user has a clear understanding of the product that current collection is bought, if
Semantic analysis result is to deny information, then illustrates that there is also queries for product of the user to purchase, i.e. semantic analysis result is different, institute
Corresponding response operation should be different.If voice print matching result is successful match, illustrate target vocal print feature and standard vocal print
Characteristic matching success, voice data to be identified is to reply in person；If voice print matching result is that it fails to match, illustrate target vocal print
Feature and standard vocal print feature successful match, voice data to be identified is not to reply in person, i.e., voice print matching result is different, and institute is right
The response operation answered should be different.According to semantic analysis result and vocal print matching result, production payment request is carried out corresponding
Response operation, with guarantee user have a clear understanding of product purchase it should be clear that and be I reply voice data to be identified, avoid subsequent
The supervision of product sale is realized in customer complaint.
Step S10-S60, when obtaining the production payment request that user terminal is sent, in the mark that will attend a banquet and blacklist list
Blacklist mark is matched, to determine whether to request the production payment to carry out speech verification.If list matching result is
With success, then corresponding voice broadcast data are determined according to product identification, voice broadcast data is subjected to voice broadcast, to determine
Whether user has a clear understanding of the product purchase notice of the product of purchase.User terminal is obtained based on voice broadcast data feedback wait know
Other voice data, to voice data to be identified carry out semantic analysis, with get user whether have a clear understanding of product purchase it should be clear that
Subsequent user is reduced to complain.Target vocal print feature is extracted from voice data to be identified, it will standard sound corresponding with user identifier
Line feature and target vocal print feature are matched, and are obtained voice print matching and are determined whether as a result, realizing for my reply.According to semanteme
Result and vocal print matching result are analyzed, corresponding response is carried out to production payment request and is operated, realizes the supervision of product sale.
In one embodiment, as shown in figure 3, in step S40, i.e., semantic analysis is carried out to voice data to be identified, obtained
Semantic analysis result specifically comprises the following steps:
S41: pre-processing voice data to be identified, obtains voice messaging.
Wherein, pretreatment, which refers to treat, knows voice data progress framing, adding window and preemphasis etc..Framing is by N number of sampling
Point set synthesizes an observation unit, referred to as frame.The value of N is 256 or 512 under normal conditions, and the time covered is about the left side 20-30ms
It is right.To avoid the variation of adjacent two frame excessive, by making have one section of overlapping region between adjacent two frame, this overlapping region is contained
M sampled point, the value of usual M are about the 1/2 or 1/3 of N, this process is known as framing.
Adding window is each frame multiplied by Hamming window (i.e. Hamming Window), since the amplitude-frequency characteristic of Hamming window is that secondary lobe declines
Subtract it is larger, server-side by single frames voice data carry out windowing process, the continuity of frame left end and frame right end can be increased.Pre-add
It is to keep the frequency spectrum of signal more flat to promote high frequency section by a high-pass filter single frames voice data after adding window again
It is sliding, low frequency is maintained at into the entire frequency band of high frequency, frequency spectrum can be sought with same signal-to-noise ratio, and the formant of prominent high frequency obtains
Preemphasis treated voice messaging.
S42: feature extraction is carried out to voice messaging, obtains phonetic feature.
Wherein, phonetic feature includes but is not limited to use filter characteristic.Filter (Filter-Bank, abbreviation Fbank)
It is characterized in common phonetic feature in speech recognition process.Since Meier feature commonly used in the prior art is carrying out model identification
Dimension-reduction treatment can be carried out to information in the process, lead to the loss of partial information, it in order to avoid the above problems, can in the present embodiment
Common Meier feature is replaced using filter characteristic.
S43: identifying phonetic feature using speech recognition modeling, obtains target text data.
Wherein, speech recognition modeling includes preparatory trained acoustic model and language model.Wherein, acoustic model is to use
To obtain the model of the corresponding aligned phoneme sequence of phonetic feature.Phoneme is by unit the smallest in voice, it will be appreciated that for inside Chinese character
Phonetic, word includes at least one phoneme.Such as: Chinese syllable ā () only one phoneme, there are two phonemes by à i (love)
Deng.The training method of acoustic model includes but is not limited to that GMM-HMM (mixed Gauss model) is used to be trained.Language model is
For aligned phoneme sequence to be converted to the model of natural language text.Specifically, phonetic feature is input to preparatory training by server
It is identified in good acoustic model, obtain the corresponding aligned phoneme sequence of target voice feature, the aligned phoneme sequence that then will acquire is defeated
Enter into preparatory trained language model and converted, obtains corresponding target text data.
S44: semantic analysis is carried out to target text data using NLP technology, is obtained corresponding with target text data
Semantic analysis result.
Wherein, NLP (Natural Language Processing, natural language processing) is computer with a kind of effective
Mode analyze, understand and from human language obtain meaning a kind of language processing techniques.By utilizing NLP technology, exploitation
Person can organize and construct knowledge hierarchy to execute autoabstract, translation, name Entity recognition, relationship extraction, sentiment analysis, language
The tasks such as sound identification and topic segmentation.In the present embodiment, semantic analysis interface provided by open source NLP technology can be used, to mesh
Mark lteral data carries out intention analysis, obtains analysis result corresponding with target text data.
Specifically, target text data are input in semantic analysis interface and carry out intention analysis, acquisition and target text
The corresponding semantic analysis result of data, semantic analysis result include confirmation message, deny information and uncertain information.By right
Target text data carry out semantic analysis, to determine whether user has a clear understanding of product purchase it should be clear that in turn according to semanteme point
It analyses result and carries out corresponding response operation, avoid violation operation of attending a banquet.
In step S41-S44, by being pre-processed to voice data to be identified, to obtain more smooth voice letter
Breath, then feature extraction is carried out to voice messaging, phonetic feature is obtained, to know using speech recognition modeling to phonetic feature
Not, target text data are obtained, semantic analysis are carried out to target text data by NLP technology, to determine whether user is clear
Product purchase notice operates as a result, responding accordingly convenient for subsequent progress.
It in one embodiment, i.e., will standard vocal print feature corresponding with user identifier and mesh as shown in figure 4, in step S50
Mark vocal print feature is matched, and obtains voice print matching as a result, specifically comprising the following steps:
S51: using cosine similarity algorithm pair standard vocal print feature corresponding with user identifier and target vocal print feature into
Row similarity calculation obtains vocal print similarity.
Wherein, vocal print similarity refers to target vocal print feature value similar with standard vocal print feature.
Specifically, standard vocal print feature corresponding with user identifier is first obtained, using cosine similarity formula, to standard sound
Line feature and target vocal print feature carry out similarity calculation, obtain vocal print similarity.Wherein, cosine similarity calculation formula isS is vocal print similarity, AiFor target vocal print feature, BiFor standard vocal print
Feature, i are i-th dimension feature, and n is number of dimensions.
S52: if vocal print similarity is greater than similarity threshold, target vocal print feature and standard vocal print feature successful match,
Obtaining voice print matching result is successful match.
Wherein, similarity threshold refers to preset for determining whether the corresponding user of target vocal print feature is preparatory
The user of the standard vocal print feature of storage.
Specifically, the vocal print similarity that server-side will acquire is compared with similarity threshold, if vocal print similarity is big
In similarity threshold, then target vocal print feature and standard vocal print feature successful match, then obtain voice print matching result for matching at
Function.
S53: if vocal print similarity is not more than similarity threshold, target vocal print feature matches mistake with standard vocal print feature
It loses, obtaining voice print matching result is that it fails to match.
Specifically, if vocal print similarity is not more than similarity threshold, target vocal print feature is matched with standard vocal print feature
Failure, then obtaining voice print matching result is that it fails to match.
It is similar with the progress of standard vocal print feature to target vocal print feature using cosine similarity algorithm in step S51-S53
Degree calculates, and obtains vocal print similarity, and calculation method is simple and quick, and is determined whether according to vocal print feature for my reply.If sound
Line similarity is greater than similarity threshold, then target vocal print feature and standard vocal print feature successful match, obtain voice print matching result
For successful match；If vocal print similarity is not more than similarity threshold, it fails to match with standard vocal print feature for target vocal print feature,
Obtaining voice print matching result is that it fails to match, to improve the acquisition accuracy of voice print matching result.
In one embodiment, as shown in figure 5, in step S60, according to semantic analysis result and vocal print matching result, to production
Product payment request carries out corresponding response operation, specifically comprises the following steps:
S61: if voice print matching result is successful match, and semantic analysis result is confirmation message, then is based on production payment
Request enters corresponding delivery operation interface.
Specifically, if voice print matching result is successful match, voice data to be identified is to reply in person, and semantic analysis
As a result it is confirmation message, then illustrates that user understands that product is bought it should be clear that then entering corresponding payment behaviour based on production payment request
Make interface, user can complete delivery operation based on delivery operation interface.
S62: if voice print matching result is successful match, and semantic analysis result is to deny information, then to client feeds back
It answers questions information, obtains user terminal based on information feedback of answering questions and continue with information.
Wherein, information of answering questions refers to the information that doubt is released for user corresponding with product identification, it is possible to understand that ground is answerred questions
Information is art if FAQs corresponding with product identification and answer information, is that user carries out clear one's mind of doubt by the words art.Continue
Handling information is that user determines whether the information explained the puzzle based on information of answering questions.Continuing with information includes continuing payment information and refusing
Exhausted payment information.
Specifically, if voice print matching result is successful match, and semantic analysis result is to deny information, then voice to be identified
Data are to reply in person, but the user has the product of purchase and feels uncertain, then first determines the doubt problem of user, further according to
A gang of's problem calling at family is answerred questions information, is answerred questions information using TTS technology voice broadcast, and user can carry out according to information of answering questions
Explain the puzzle, and feeds back and continue with information.
S63: it if continuing with information to continue payment information, executes and voice broadcast data is broadcasted using TTS technology.
Specifically, if getting continue with information of the user terminal based on information feedback of answering questions is to continue payment information,
The user has explained the puzzle according to information of answering questions, then executes the step of broadcasting voice broadcast data using TTS technology, the step and step
Step is identical in S30, does not repeat specifically herein.
S64: if continuing with information as refusal to pay information, information is handled to client feeds back objection, obtains user
End group is in the final determining information of objection processing information feedback.
Specifically, information is continued with as refusal to pay information based on information feedback of answering questions if obtaining user terminal, be somebody's turn to do
User does not explain the puzzle according to information of answering questions, then handles information to server-side feedback objection, wherein objection processing information refers to by artificial
The information explained the puzzle obtains user based on the final determining information of objection processing information feedback, wherein final to determine that information is
Refer to that user determines whether the information explained the puzzle based on objection processing information, it is final to determine that information includes non-payment information and continues to prop up
Pay information.
S65: if finally determining, information is non-payment information, exits current interface.
Specifically, if the final determining information of client feeds back is non-payment information, user is handled according to objection to be believed
Breath is not explained the puzzle, and attending a banquet, there may be violation operations, then not can be carried out delivery operation, exit current interface.By non-payment information
Associated storage is identified with attending a banquet, identifies corresponding monitor duration convenient for subsequent increase blacklist.
S66: if finally determining, information is to continue payment information, executes and broadcasts voice broadcast data using TTS technology.
Specifically, if the final determining information of client feeds back is to continue payment information, user is handled according to objection to be believed
Breath has been explained the puzzle, then executes and broadcast voice broadcast data using TTS technology, the step is identical as step in step S30, does not make herein
Specifically repeat.It is to be appreciated that passing through step S30 if executing step S30 after user explains the puzzle and reacquiring language to be identified
Sound data can when complaining convenient for subsequent user by associated storages such as voice data to be identified, user identifier and final determining information
It is used as evidence.
Further, it if voice print matching result is that it fails to match, sends and reminds to the corresponding mobile terminal of user identifier
Information.Wherein, prompting message is " current reply is not to operate in person " etc., by this step, with determination voice data to be identified
For my reply.Wherein, mobile terminal is the terminal device of user.
Further, if voice print matching result is successful match, and semantic analysis result is uncertain information, then repeats to hold
Row carries out voice broadcast to voice broadcast data using TTS technology.Specifically, when semantic analysis result is uncertain information, i.e.,
The voice data to be identified and voice broadcast data that user replys be not corresponding, then executes using TTS technology to voice broadcast data
Voice broadcast is carried out, to determine whether user has a clear understanding of product purchase it should be clear that avoiding violation operation of attending a banquet.
In step S61-S66, if voice print matching result is successful match, and semantic analysis result is confirmation message, then root
It requests to enter corresponding delivery operation interface according to production payment, avoids violation operation of attending a banquet, reduce subsequent user and complain.If vocal print
Matching result is successful match, and semantic analysis result is to deny information, then according to information and the objection processing information equipotential of answering questions
User explains the puzzle, so that user has a clear understanding of product purchase it should be clear that improving user experience.
In one embodiment, as shown in fig. 6, before step S20, i.e., the black name in will attend a banquet mark and blacklist list
Before single mark is matched, product monitoring and managing method further include:
S201: it obtains each attend a banquet and identifies corresponding violation number, if violation number is greater than frequency threshold value, will attend a banquet mark
Know and is identified as blacklist.
Wherein, frequency threshold value is preset for determining the threshold value attended a banquet and identify whether to identify for blacklist.Blacklist
Mark refers to that violation number is greater than the mark of attending a banquet of frequency threshold value.
Specifically, it obtains each attend a banquet and identifies corresponding violation number, wherein the channel of acquisition violation number can be divided into more
Kind, one is clients to complain to attending a banquet, accessed to identify corresponding violation number with attending a banquet；One is supervisors
Feedback identifies corresponding violation number with attending a banquet, another is the non-payment information fed back when client pays, root
According to non-payment acquisition of information to identify corresponding violation number with attending a banquet.It counts each attend a banquet and identifies corresponding all violations
Number compares all violation numbers with frequency threshold value, if violation number is greater than frequency threshold value, which is made
For blacklist mark, when identifying sale product so as to the subsequent blacklist, corresponding payment request is identified to the blacklist
It carries out IVR speech verification (i.e. the step of step S30-S50), realizes that identifying corresponding sales behavior to blacklist supervises,
To ensure that user has a clear understanding of the product of purchase, subsequent user is avoided to complain.If violation number is not more than frequency threshold value, not to this
It attends a banquet and identifies corresponding product progress IVR speech verification.
S202: identifying corresponding violation number according to blacklist, determine blacklist identify corresponding supervision duration and with prison
The pipe duration corresponding regulatory time limit.
Wherein, supervision duration refers to the time corresponding with violation number, for example, violation number is 2 times, a length of one when supervision
A month.Regulatory time limit refers to that blacklist identifies corresponding initial time and deadline, for example, the violation time of a certain mark of attending a banquet
Number is 2 times, and frequency threshold value is greater than at January 1, then a length of one month when supervising, then the regulatory time limit is January 1 to 1 day 2 months.
Specifically, the corresponding supervision duration of different violation numbers is different, it is possible to understand that ground, violation number is more, corresponding
Supervision duration is longer, conversely, violation number is fewer, corresponding supervision duration is shorter.It is pre-configured with each violation number and supervision
The corresponding relationship of duration, and store in the database.Identify corresponding violation number by blacklist and search database, obtain with
Corresponding supervision duration determines the regulatory time limit as supervision duration corresponding with blacklist mark, and according to supervision duration.Pass through
This step can quickly determine that each blacklist identifies corresponding supervision duration and regulatory time limit, determine that method is simple and quick.
S203: blacklist list is formed based on each blacklist mark and regulatory time limit, and is stored in the database.
Wherein, blacklist list is to store the list of blacklist mark and the corresponding relationship between the regulatory time limit.
Specifically, it is identified based on each blacklist and the regulatory time limit forms blacklist list, convenient for subsequent determining needs pair
Which production payment request carries out IVR speech verification.
In step S201-S203, according to it is each attend a banquet identify corresponding violation number determine blacklist identify, so as to subsequent
Corresponding production payment emotion is identified to blacklist and seeks progress IVR speech verification.Corresponding violation number is identified according to blacklist,
Determine that blacklist identifies corresponding supervision duration and regulatory time limit corresponding with duration is supervised, so as within the regulatory time limit to black
List identifies corresponding production payment request and is supervised.
In one embodiment, as shown in fig. 7, after step S60, i.e., according to semantic analysis result and voice print matching knot
Fruit, after carrying out corresponding response operation to production payment request, product monitoring and managing method further include:
S601: it when the current time in system is that each blacklist identifies the deadline of corresponding regulatory time limit, obtains black
List identifies at least one semantic analysis result within the regulatory time limit.
Specifically, current_date function can be used and obtain the current time in system, be blacklist when the current time in system
When identifying the deadline of corresponding regulatory time limit, obtains blacklist and identify at least one semantic analysis knot within the regulatory time limit
Fruit, by semantic analysis result to determine that the blacklist identifies whether corresponding attend a banquet has violation operation.
S602: if each semantic analysis result is confirmation message, blacklist mark is rejected from blacklist list.
Specifically, if each semantic analysis result corresponding with blacklist mark is confirmation message, then illustrate the black name
Corresponding attend a banquet of single mark does not carry out violation operation within the regulatory time limit, needs to propose blacklist mark from blacklist list
It rejects, is not required to supervise the corresponding production payment request of blacklist mark.
Further, it can also determine blacklist mark within the regulatory time limit by other means, if to exist and grasp in violation of rules and regulations
Make, violation operation if it exists, then execute and handled according to preset rules quantity is summarized, determines the step of newly-increased supervision duration
Suddenly；Violation operation if it does not exist then executes the step of rejecting blacklist mark from blacklist list.
S603: at least one NACK messages if it exists, then statistics identifies the remittance of corresponding non-payment information with blacklist
Total quantity is handled according to preset rules quantity is summarized, and determines newly-increased supervision duration.
Wherein, summarize quantity and refer to the quantity for identifying the non-payment information of corresponding client feeds back with blacklist.In advance
If rule refers to preset for according to the rule of the determining supervision duration of quantity is summarized, for example, summarizing, quantity to be bigger, and institute is right
The newly-increased supervision duration answered is longer.Newly-increased monitor duration refers to the increased duration of institute on the basis of monitor duration, for example, a certain
It is 1 month a length of when the corresponding supervision of blacklist mark, according to preset rules and summarize quantity, determine it is 10 days a length of when newly-increased supervision,
Increase by 10 days supervision time on the basis of then in supervision a length of 1 month.
Specifically, if there are at least one NACK messages, users couple for semantic analysis result corresponding with blacklist mark
The product of purchase is counted and summarizes quantity with what blacklist identified corresponding non-payment information, according to preset rules there are query
With quantity is summarized to determine newly-increased monitor duration.
S604: based on newly-increased supervision duration, the monitoring time limit corresponding with blacklist mark in blacklist list is updated.
Specifically, after server-side gets newly-increased supervision duration corresponding with blacklist mark, when according to newly-increased supervision
Length redefines the monitoring time limit corresponding with blacklist mark, and prison corresponding with blacklist mark is updated in blacklist list
Control the time limit.
In step S601-S604, when the current time in system is that blacklist identifies the deadline of corresponding regulatory time limit,
Determine that blacklist identifies corresponding semantic analysis result；If each semantic analysis result is confirmation message, the blacklist mark
Know corresponding attend a banquet without violation operation within the monitoring time limit, without being monitored to blacklist mark.If it exists at least one
A NACK messages, then corresponding attend a banquet of blacklist mark exists within the monitoring time limit against operation is advised, then statistics and blacklist
Identify corresponding non-payment information summarizes quantity, identifies corresponding supervision duration to increase the blacklist.Based on newly-increased prison
Pipe duration updates the monitoring time limit corresponding with blacklist mark in blacklist list, to monitor in the time limit to the black name
List identifies corresponding production payment request and is supervised.
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 product maintenance device is provided, product is supervised in the product maintenance device and above-described embodiment
Pipe method corresponds.As shown in figure 8, the product maintenance device include payment request obtain module 10, list matching module 20,
Voice broadcast module 30, semantic analysis result obtain module 40, voice print matching result obtains module 50 and response operation module 60.
Detailed description are as follows for each functional module:
Payment request obtains module 10, and for obtaining the production payment request of user terminal transmission, production payment request includes
User identifier, product identification and mark of attending a banquet.
List matching module 20 matches with blacklist mark in blacklist list for that will attend a banquet mark, obtains name
Single matching result.
Voice broadcast module 30 determines corresponding if being successful match for list matching result according to product identification
Product type calls corresponding voice broadcast data according to product type, carries out voice to voice broadcast data using TTS technology
Casting.
Semantic analysis result obtains module 40, for obtaining to be identified voice of the user terminal based on voice broadcast data feedback
Data carry out semantic analysis to voice data to be identified, obtain semantic analysis result.
Voice print matching result obtain module 50, for from voice data to be identified extract target vocal print feature, will with
Family identifies corresponding standard vocal print feature and target vocal print feature is matched, and obtains voice print matching result.
Operation module 60 is responded, for requesting to carry out to production payment according to semantic analysis result and vocal print matching result
Corresponding response operation.
In one embodiment, semantic analysis result obtain module 40, including pretreatment unit 41, feature extraction unit 42,
Word recognition unit 43 and semantic analysis unit 44.
Pretreatment unit 41 obtains voice messaging for pre-processing to voice data to be identified.
Feature extraction unit 42 obtains phonetic feature for carrying out feature extraction to voice messaging.
Word recognition unit 43 obtains target text number for identifying using speech recognition modeling to phonetic feature
According to.
Semantic analysis unit 44 obtains and target text for carrying out semantic analysis to target text data using NLP technology
The corresponding semantic analysis result of digital data.
In one embodiment, voice print matching result obtains module 50, including vocal print similarity acquiring unit, the first vocal print
With result acquiring unit and the second vocal print matching result acquiring unit.
Vocal print similarity acquiring unit, for special using cosine similarity algorithm pair standard vocal print corresponding with user identifier
Target of seeking peace vocal print feature carries out similarity calculation, obtains vocal print similarity.
First voice print matching result acquiring unit, if being greater than similarity threshold for vocal print similarity, target vocal print is special
Sign and standard vocal print feature successful match, obtaining voice print matching result is successful match.
Second vocal print matching result acquiring unit, if being not more than similarity threshold, target vocal print for vocal print similarity
It fails to match with standard vocal print feature for feature, and obtaining voice print matching result is that it fails to match.
In one embodiment, operation module 60 is responded, including delivery operation interface acquiring unit, continues with acquisition of information
Unit, finally determines information acquisition unit, the second response unit and third response unit at the first response unit.
Delivery operation interface acquiring unit, if being successful match for voice print matching result, and semantic analysis result is true
Recognize information, then corresponding delivery operation interface is entered based on production payment request.
Information acquisition unit is continued with, if being successful match for voice print matching result, and semantic analysis result is no
Recognize information, then answer questions information to client feeds back, obtains user terminal based on information feedback of answering questions and continue with information.
First response unit, if executing to continue payment information for continuing with information and broadcasting language using TTS technology
Sound broadcasts data.
It is final to determine information acquisition unit, if for continuing with information for refusal to pay information, to client feeds back
Objection handles information, obtains user terminal based on the final determining information of objection processing information feedback.
Second response unit, if exiting current interface for finally determining that information is non-payment information.
Third response unit, if executing for finally determining that information is to continue payment information and broadcasting language using TTS technology
Sound broadcasts data.
In one embodiment, before list matching module 20, product maintenance device further includes that blacklist mark determines list
Member, regulatory time limit determination unit and blacklist list form unit.
Blacklist identifies determination unit, corresponding violation number is identified for obtaining each attend a banquet, if violation number is greater than
Frequency threshold value, the then mark that will attend a banquet are identified as blacklist.
Regulatory time limit determination unit determines that blacklist mark corresponds to for identifying corresponding violation number according to blacklist
Supervision duration and with supervision duration corresponding regulatory time limit.
Blacklist list forms unit, for forming blacklist list based on each blacklist mark and regulatory time limit, and
Storage is in the database.
In one embodiment, after responding operation module 60, product maintenance device further includes that semantic analysis result obtains
Unit, blacklist mark culling unit, newly-increased supervision duration determination unit and monitoring time limit updating unit.
Semantic analysis result acquiring unit, for being that blacklist identifies cutting for corresponding regulatory time limit when the current time in system
Only when the time, obtains blacklist and identify at least one semantic analysis result within the regulatory time limit.
Blacklist mark culling unit identifies blacklist if being confirmation message for each semantic analysis result
It is rejected from blacklist list.
Newly-increased supervision duration determination unit then counts and blacklist mark pair at least one NACK messages if it exists
The non-payment information answered summarizes quantity, is handled according to preset rules quantity is summarized, and determines newly-increased supervision duration.
Time limit updating unit is monitored, for updating in blacklist list and identifying phase with blacklist based on newly-increased supervision duration
The corresponding monitoring time limit.
Specific about product maintenance device limits the restriction that may refer to above for product monitoring and managing method, herein not
It repeats again.Modules in the said goods maintenance device can be fully or partially through software, hardware and its group and to realize.On
Stating each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also store in a software form
In memory in computer equipment, 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-side, 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 database of machine equipment, which is used to store, executes product monitoring and managing method generation or the data obtained etc. in the process, for example, blacklist arranges
Table.The network interface of the computer equipment is used to communicate with external terminal by network connection.The computer program is processed
To realize a kind of product monitoring and managing method when device executes.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, processor realize product monitoring party in above-described embodiment when executing computer program
The step of method, for example, S10 shown in Fig. 2 to step S60 or Fig. 3 is to step shown in fig. 7.Alternatively, processor executes meter
The function of each module in above-described embodiment in product maintenance device is realized when calculation machine program, for example, module 10 shown in Fig. 8 is to mould
The function of block 60.To avoid repeating, details are not described herein again.
In one embodiment, a kind of computer readable storage medium is provided, computer program, computer are stored thereon with
Product monitoring and managing method in above method embodiment is realized when program is executed by processor, for example, step S10 shown in Fig. 2 is extremely walked
Rapid S60 or Fig. 3 is to step shown in fig. 7.Alternatively, the computer program realizes above-described embodiment when being executed by processor
The function of each module in middle product maintenance device, for example, function of the module 10 shown in Fig. 8 to module 60.To avoid repeating, herein
It repeats no more.
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, computer program can be stored in a non-volatile computer and can be read
In storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the application
To any reference of memory, storage, database or other media used in provided each embodiment, may each comprise non-
Volatibility 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 arbitrary access
Memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms, such as static
RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhanced SDRAM
(ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (RambuS) directly RAM (RDRAM), straight
Connect 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 device are divided into different functional unit or module, to complete above description
All or part of function.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification 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 include
Within protection scope of the present invention.
Claims (10)
1. a kind of product monitoring and managing method characterized by comprising
The production payment request that user terminal is sent is obtained, the production payment request is comprising user identifier, product identification and attends a banquet
Mark；
The mark of attending a banquet is matched with blacklist mark in blacklist list, obtains list matching result；
If the list matching result is successful match, corresponding product type is determined according to the product identification, according to institute
It states product type and calls corresponding voice broadcast data, voice broadcast is carried out to the voice broadcast data using TTS technology；
Obtain to be identified voice data of the user terminal based on the voice broadcast data feedback, to the voice data to be identified into
Row semantic analysis obtains semantic analysis result；
Target vocal print feature is extracted from the voice data to be identified, it will standard vocal print feature corresponding with the user identifier
It is matched with the target vocal print feature, obtains voice print matching result；
It is grasped according to the semantic analysis result and the voice print matching as a result, carrying out corresponding response to production payment request
Make.
2. product monitoring and managing method as described in claim 1, which is characterized in that described to carry out language to the voice data to be identified
Justice analysis, obtains semantic analysis result, comprising:
The voice data to be identified is pre-processed, voice messaging is obtained；
Feature extraction is carried out to the voice messaging, obtains phonetic feature；
The phonetic feature is identified using speech recognition modeling, obtains target text data；
Semantic analysis is carried out to the target text data using NLP technology, is obtained corresponding with the target text data
Semantic analysis result.
3. product monitoring and managing method as described in claim 1, which is characterized in that it is described will standard corresponding with the user identifier
Vocal print feature and the target vocal print feature are matched, and voice print matching result is obtained, comprising:
Using cosine similarity algorithm pair standard vocal print feature corresponding with the user identifier and the target vocal print feature into
Row similarity calculation obtains vocal print similarity；
If the vocal print similarity is greater than similarity threshold, the target vocal print feature is matched into the standard vocal print feature
Function, obtaining voice print matching result is successful match；
If the vocal print similarity is not more than similarity threshold, the target vocal print feature is matched with the standard vocal print feature
Failure, obtaining voice print matching result is that it fails to match.
4. product monitoring and managing method as described in claim 1, which is characterized in that described according to the semantic analysis result and described
Voice print matching operates as a result, carrying out corresponding response to production payment request, comprising:
If the voice print matching result is successful match, and the semantic analysis result is confirmation message, then is based on the product
Payment request enters corresponding delivery operation interface；
If the voice print matching result is successful match, and the semantic analysis result is to deny information, then to the user terminal
Information of answering questions is fed back, the user terminal is obtained based on the information feedback of answering questions and continues with information；
If described continue with information to continue payment information, execute described using TTS technology casting voice broadcast data；
If the information that continues with handles information to the client feeds back objection, described in acquisition for refusal to pay information
Final determining information of the user terminal based on objection processing information feedback；
If the final determining information is non-payment information, current interface is exited；
If the final determining information is to continue payment information, execute described using TTS technology casting voice broadcast data.
5. product monitoring and managing method as described in claim 1, which is characterized in that described by mark and the blacklist column of attending a banquet
Before blacklist mark is matched in table, the product monitoring and managing method further include:
It obtains each described attend a banquet and identifies corresponding violation number, if the violation number is greater than frequency threshold value, by the seat
Seat mark is identified as blacklist；
Identify corresponding violation number according to the blacklist, determine the blacklist identify corresponding supervision duration and with it is described
Supervise the duration corresponding regulatory time limit；
Blacklist list is formed based on each blacklist mark and the regulatory time limit.
6. product monitoring and managing method as claimed in claim 4, which is characterized in that described according to the semantic analysis result and institute
After voice print matching is stated as a result, carrying out corresponding response operation to production payment request, the product monitoring and managing method is also wrapped
It includes:
When the current time in system is that the blacklist identifies the deadline of corresponding regulatory time limit, the blacklist mark is obtained
Know at least one semantic analysis result within the regulatory time limit；
If each semantic analysis result is confirmation message, blacklist mark is rejected from blacklist list；
At least one NACK messages if it exists, then statistics identifies the total amount of corresponding non-payment information with the blacklist
Amount, is handled the quantity that summarizes according to preset rules, determines newly-increased supervision duration；
Based on the newly-increased supervision duration, the monitoring phase corresponding with blacklist mark in the blacklist list is updated
Limit.
7. a kind of product maintenance device characterized by comprising
Payment request obtains module, and for obtaining the production payment request of user terminal transmission, the production payment request is comprising using
Family mark, product identification and mark of attending a banquet；
List matching module obtains list for matching the mark of attending a banquet with blacklist mark in blacklist list
Matching result；
Voice broadcast module is determined according to the product identification and is corresponded to if being successful match for the list matching result
Product type, corresponding voice broadcast data are called according to the product type, using TTS technology to the voice broadcast number
According to progress voice broadcast；
Semantic analysis result obtains module, for obtaining to be identified voice number of the user terminal based on the voice broadcast data feedback
According to the voice data progress semantic analysis to be identified, acquisition semantic analysis result；
Voice print matching result obtain module, for from the voice data to be identified extract target vocal print feature, will with it is described
The corresponding standard vocal print feature of user identifier and the target vocal print feature are matched, and voice print matching result is obtained；
Respond operation module, for according to the semantic analysis result and the voice print matching as a result, being asked to the production payment
It asks and carries out corresponding response operation.
8. product maintenance device as claimed in claim 7, which is characterized in that the semantic analysis result obtains module, comprising:
Pretreatment unit obtains voice messaging for pre-processing to the voice data to be identified；
Feature extraction unit obtains phonetic feature for carrying out feature extraction to the voice messaging；
Word recognition unit obtains target text data for identifying using speech recognition modeling to the phonetic feature；
Semantic analysis unit obtains and the target for carrying out semantic analysis to the target text data using NLP technology
The corresponding semantic analysis result of lteral data.
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
The step of any one of 6 product monitoring and managing method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization product monitoring and managing method as described in any one of claim 1 to 6 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910300383.7A CN110163630B (en) | 2019-04-15 | 2019-04-15 | Product supervision method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910300383.7A CN110163630B (en) | 2019-04-15 | 2019-04-15 | Product supervision method, device, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110163630A true CN110163630A (en) | 2019-08-23 |
| CN110163630B CN110163630B (en) | 2024-04-05 |
Family
ID=67639532
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910300383.7A Active CN110163630B (en) | 2019-04-15 | 2019-04-15 | Product supervision method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163630B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112651787A (en) * | 2021-01-06 | 2021-04-13 | 上海明略人工智能(集团)有限公司 | Payment method and device for orally-played advertisement |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016119359A1 (en) * | 2015-01-26 | 2016-08-04 | 中兴通讯股份有限公司 | Voice control-based mobile payment method and apparatus |
| WO2018113526A1 (en) * | 2016-12-20 | 2018-06-28 | 四川长虹电器股份有限公司 | Face recognition and voiceprint recognition-based interactive authentication system and method |
| CN108347441A (en) * | 2018-02-08 | 2018-07-31 | 深圳壹账通智能科技有限公司 | Payment processing method, device, computer equipment and storage medium |
| CN108428134A (en) * | 2017-06-24 | 2018-08-21 | 平安科技（深圳）有限公司 | Electronic device, IVR voice payments method and computer readable storage medium |
| WO2018149209A1 (en) * | 2017-02-15 | 2018-08-23 | 腾讯科技（深圳）有限公司 | Voice recognition method, electronic device, and computer storage medium |
-
2019
- 2019-04-15 CN CN201910300383.7A patent/CN110163630B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016119359A1 (en) * | 2015-01-26 | 2016-08-04 | 中兴通讯股份有限公司 | Voice control-based mobile payment method and apparatus |
| WO2018113526A1 (en) * | 2016-12-20 | 2018-06-28 | 四川长虹电器股份有限公司 | Face recognition and voiceprint recognition-based interactive authentication system and method |
| WO2018149209A1 (en) * | 2017-02-15 | 2018-08-23 | 腾讯科技（深圳）有限公司 | Voice recognition method, electronic device, and computer storage medium |
| CN108428134A (en) * | 2017-06-24 | 2018-08-21 | 平安科技（深圳）有限公司 | Electronic device, IVR voice payments method and computer readable storage medium |
| CN108347441A (en) * | 2018-02-08 | 2018-07-31 | 深圳壹账通智能科技有限公司 | Payment processing method, device, computer equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112651787A (en) * | 2021-01-06 | 2021-04-13 | 上海明略人工智能(集团)有限公司 | Payment method and device for orally-played advertisement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110163630B (en) | 2024-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10728384B1 (en) | System and method for redaction of sensitive audio events of call recordings | |
| US10210870B2 (en) | Method for verification and blacklist detection using a biometrics platform | |
| US8145562B2 (en) | Apparatus and method for fraud prevention | |
| US11714793B2 (en) | Systems and methods for providing searchable customer call indexes | |
| US10636047B2 (en) | System using automatically triggered analytics for feedback data | |
| EP1902442B1 (en) | Selective security masking within recorded speech utilizing speech recognition techniques | |
| US11336771B2 (en) | Method and system for accurate automatic call tracking and analysis | |
| US20150142446A1 (en) | Credit Risk Decision Management System And Method Using Voice Analytics | |
| KR20200092499A (en) | Method and apparatus for counseling support using interactive artificial intelligence technology | |
| CN110177182A (en) | Sensitive data processing method, device, computer equipment and storage medium | |
| CN110851576A (en) | Question and answer processing method, device, equipment and readable medium | |
| CN113343058B (en) | Voice session supervision method, device, computer equipment and storage medium | |
| CN110766442A (en) | Client information verification method, device, computer equipment and storage medium | |
| Shim et al. | Phonetic analytics technology and big data: real-world cases | |
| CN109346082A (en) | Sales order acquisition methods, device, equipment and medium based on speech recognition | |
| GB2521050A (en) | Credit risk decision management system and method using voice analytics | |
| US11335351B2 (en) | Cognitive automation-based engine BOT for processing audio and taking actions in response thereto | |
| EP4016355B1 (en) | Anonymized sensitive data analysis | |
| Weber et al. | Validations of an alpha version of the E3 Forensic Speech Science System (E3FS3) core software tools | |
| CN113064983B (en) | Semantic detection method, semantic detection device, computer equipment and storage medium | |
| CN110163630A (en) | Product monitoring and managing method, device, computer equipment and storage medium | |
| CN113438374A (en) | Intelligent outbound call processing method, device, equipment and storage medium | |
| KR20180042763A (en) | Chatting type financial robot and method for providing financial service using the same | |
| US11704585B2 (en) | System and method to determine outcome probability of an event based on videos | |
| CN113314103B (en) | Illegal information identification method and device based on real-time speech emotion analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |