CN106446205A - Internet-based method and device for forming paid knowledgea of expert into audio compilation album - Google Patents
Internet-based method and device for forming paid knowledgea of expert into audio compilation album Download PDFInfo
- Publication number
- CN106446205A CN106446205A CN201610870493.3A CN201610870493A CN106446205A CN 106446205 A CN106446205 A CN 106446205A CN 201610870493 A CN201610870493 A CN 201610870493A CN 106446205 A CN106446205 A CN 106446205A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- answer
- question
- unit
- audio
- 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 47
- 238000012549 training Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000005070 sampling Methods 0.000 claims abstract description 8
- 238000004891 communication Methods 0.000 claims description 22
- 238000001556 precipitation Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 21
- 230000000694 effects Effects 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 7
- 239000002244 precipitate Substances 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 4
- 230000001902 propagating effect Effects 0.000 claims description 4
- 238000012795 verification Methods 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 2
- 230000005540 biological transmission Effects 0.000 claims description 2
- 230000006835 compression Effects 0.000 claims description 2
- 238000007906 compression Methods 0.000 claims description 2
- 238000007418 data mining Methods 0.000 claims description 2
- 238000013461 design Methods 0.000 claims description 2
- 239000012634 fragment Substances 0.000 claims description 2
- 239000003550 marker Substances 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 claims description 2
- 239000002243 precursor Substances 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 238000010200 validation analysis Methods 0.000 claims description 2
- 238000012790 confirmation Methods 0.000 claims 2
- 241000208340 Araliaceae Species 0.000 claims 1
- 241000196324 Embryophyta Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 230000001186 cumulative effect Effects 0.000 claims 1
- 230000005611 electricity Effects 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 230000000644 propagated effect Effects 0.000 claims 1
- 239000011435 rock Substances 0.000 claims 1
- 238000005065 mining Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 3
- 230000002860 competitive effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 238000013475 authorization Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000009223 counseling Methods 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
- 238000003306 harvesting Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012360 testing method Methods 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
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The invention provides a method and device and system for mining paid knowledge of an expert and forming an expert audio compilation album based on a mobile Internet sharing network model. A knowledgea response unit sends voice through an input unit for searching, and an audio sampling unit calls an acoustic model formed in an audio training database for performing voice recognition and sends the recognized similar text to a server for searching; an audio index processing unit of the server conducts searching to call out the relevant audio and the relevant text in a knowledge base; a content display control unit of a mobile client obtains a relevance feedback result, and an audio index processing unit calculates the text similarity; based on the display result of the content display control unit, the knowledge response unit conducts editorial selection on the compilation album; a server knowledge rule engine unit calculates the price of the compilation album. The audio index processing unit of the server collects the attribute data of multiple roles together again, establishes an index to form the compilation album, and stores the compilation album in the knowledge base.
Description
Technical field
The present invention is to be related to a kind of sharing based on mobile Internet to excavate the paid knowledge of expert under economic model and formed specially
The method and device of family's audio frequency album.
Background technology
At present, share under economic tide although free knowledge sharing information can be passed through in Internet era
In internet, search obtains, but actually paid high value knowledge sharing has become as a kind of demand, because what individuality had
Knowledge, particularly implicit knowledge are often the unique experience knowledge that it is accumulated for a long time, typically they be unwilling by its with
Other people gratuitously share, and in the tissue ring of competition, the competitive pressure that individuality is experienced makes them think oneself institute
The knowledge having is the guarantee that oneself is worth in the tissue with status, if the knowledge being had gratuitously is emptied one's purse to authorize other people,
The competitive advantage of this uniqueness will be lost, then the individual interests of itself cannot be guaranteed.So, in order to encourage these to have
The individuality enriched one's knowledge, shared knowledge, need one fast and effectively paid knowledge sharing system come to knowledge provider with know
Know the docking of Requirement Acquisition side.
It is proposed that a kind of knowledge transaction system centered on knowledge services and method in referenced patent document 1.Although should
Document also refer to paid Knowledge Transaction, but its system is to service publisher to pass through network or terminal software game server, leads to
Cross the data containing knowledge services information（Service content, title, charging way etc.）It is sent in server, and server is made
For intermediary, the dealing money that both sides are consulted carries out trustship, after the completion of payment, facilitates Knowledge Transaction.
But corresponding at present similar conventional internet knowledge sharing framework, people are simply by this knowledge question active procedure
Simply it is decomposed into two sides：Demand for counseling enquirement side and the supplying party answering consulting, both sides pass through the platform Consulting Trade System amount of money, enter
Row Knowledge Transaction.But under Novel Internet epoch and knowledge paid pattern, there is drawback in this traditional pattern in fact,
Because party in request will undertake completely cost to paid knowledge, and it is low to promote efficiency.And in fact, during asking, its
Real quizmaster, also create a kind of knowledge and put question to scene, the individuality enriched one's knowledge has been carried out with knowledge experience and has excavated and explore
Activity, in this sense, in question and answer, he is also the knowledge provider of the value drawing up a questionnaire, and pattern originally is neglected
Omit the value of quizmaster, the knowledge process that in addition also have ignored this transaction is can to share to whole internets to use in fact
Family, forms bigger value.So, this tradition both sides' pattern can not encourage the number of activities of Knowledge Transaction to increase, thus
Nor improve the income of knowledge provision side.
With new paid knowledge sharing system, using the new technique of shared economic idea, to ensure quizmaster, to answer
Multiple person, the enthusiasm of substantial amounts of third party reader, to meet the demand of tripartite, to form more valuable Knowledge Sharings, shape
Become the multi-win of whole activity link.There is currently the demand of two kinds of subdivisions：A kind of is the special of target centered on certain expert specific
The demand that family's knowledge experience is explored, another kind is then target inquiry multidigit expert's view centered on certain certain content or event
Demand；And the present invention is the development of Mobile Internet technology of the expertise experience exploration of target centered on certain expert specific.
In addition, with smart mobile phone development, people can carry out sound anywhere or anytime by the powerful disposal ability of mobile phone
Frequently, in addition video link up, therefore under current communication conditions, brainstrust can more easily provide at any time voice formed audio frequency know
Knowledge is shared, and provides video to form video Knowledge Sharing, thus forming audio frequency knowledge base, and video knowledge base.But at present also
Do not have a kind of convenient to be convenient for people to build, and from the audio frequency knowledge base of editor or the shared technology of video knowledge base.In reference
In patent document 2, mention a kind of remote speech typing end of passing through and gather voice, allow server process form language material, and in client
End re-forms to be revised the professional method marking to form corpus, although the method is accurately, gatherer process is complicated, commonly non-
Professional, is difficult to use, leads to common cellphone user, does not have wish to provide material to form corpus as requested.
Therefore, in the paid knowledge sharing field of Audio Reply, need a kind of technology at present, efficiently to carry out audio frequency conjunction
Collect so that individual can convey knowledge faster in shared economy era, and obtain paid material and harvest.
Content of the invention
The main object of the present invention is to provide excavation expert under a kind of shared economic model based on mobile Internet to know with compensation
Know the method and device forming expert's audio frequency topic album, allow expert to be made with language is convenient with oneself at any time by the method and device
Answer, set up simultaneously and form audio frequency knowledge base, allow paid knowledge question can obtain more fully Knowledge retrieval system.
As shown in figure 1, paid knowledge the method that forms expert's audio frequency topic album, it is patrolling based on following five kinds definition
Collect mobile's formation of role：
Knowledge puts question to body 11, is responsible for puing question to whenever and wherever possible by mobile terminal in systems, is cleverly asked by reasonable in design
Topic, carrys out Extracting Knowledge and answers body.
Knowledge answers body 12, is responsible for carrying out within the long short time of regulation whenever and wherever possible by mobile terminal in systems
Audio Reply, is set with the data mining object of rich experiences knowledge in systems, and according to system record knowledge altogether
Enjoy track, to form the knowledge personal value of system evaluation.
Question and answer carrier 13, responsible reading, and carry out petit average, when the value of knowledge approving this enquirement and question and answer
Afterwards, can carry out sharing the system activity of popularization.
Question and answer read body 14, only responsible reading, and carry out petit average, and enquirement body and problem body can be carried out
Scoring, but does not transmit, and question and answer read body 14 if what carrier 13 recommended to come, and it is little that this carrier can participate in this
What volume was paid is divided into.
Knowledge precipitation can record all knowledge and put question to bodies 11 with push platform 15, platform, and knowledge answers body 12, and this is asked
The carrier 13 answered, the participation knowledge and experience that this question and answer reads body 14 shares activity.Carry out the knowledge precipitation of whole issue,
And the forwarding request of reception question and answer carrier, carry out Knowledge delivery.In one wheel question and answer, there is a knowledge to put question to body 11, have one
Individual knowledge answers body 12, can have zero or more question and answer carrier 13, it is possibility to have zero or more question and answer read body 14.
The appraisal procedure often taken turns：Knowledge is answered body 12 and first can be worth according to the knowledge experience of itself, provides in systems
Carry out a marked price action in one zone of reasonableness, obtain an initial personal value value, knowledge is answered body and passed through systematic search to sense
The ken of interest, and meet the desired knowledge of price and answer body and putd question to, and the on-line payment amount of money gives platform, simultaneously
Paid knowledge is replied and is encrypted upload and provides access entrance by platform, and question and answer are read body 14 and paid small amount gold by platform
Volume, obtains consulting authority, thinks that this question and answer is valuable, can be forwarded by platform after consulting, and propagates, and changes into question and answer transmission
Body 13, carrier can carry out certain ratio with platform to this small amount amount of money and be divided into.Consult body and forward body, can be right
This time paid knowledge provides the scoring feedback carrying out that a knowledge answers body 12, provides a positive or negative value of feedback.System meeting
Body is consulted to each and forwards body scoring to count, thus after calculating this question and answer, this knowledge is answered body 12 and is being
The up-to-date personal value in system, this value also can become the initial value before next round question and answer.
Paid knowledge question through excessive wheel, knowledge precipitates and pushes platform 15 can answer body 12 for main rope using knowledge
Draw, to add up to generate this expert's audio frequency album corresponding, because audio database is differed with plain text database, in appeal
In model, expert is needed to be that knowledge answers body 12, in mobile client（I.e. smart mobile phone, panel computer etc.）, increase audio frequency and adopt
Egf block carries out taking turns simple sampled voice more, improves phonetic recognization rate, forms experience sampling, then answers body 12 in knowledge more every
Before wheel is answered, knowledge is answered body 12 and is carried out automatic speech recognition in mobile client startup audio sample module, is uploading audio frequency
While, corresponding Similar Text is also uploaded onto the server.
When editing album, knowledge is answered body 12 and can be scanned for by voice, to arrange layout, thus forming conjunction
Volume.
The present invention provides a kind of sharing based on mobile Internet to excavate the paid knowledge of expert under economic model and form expert
The mobile client end device of audio frequency album, as shown in Figure 2；
Including：Indigenous knowledge regulation engine unit 21, content display control unit 22, input block 23, electronic payment unit 24,
Network communication unit 25, audio sample unit 26, the database 27 of audio frequency training.
The present invention provides a kind of sharing based on mobile Internet to excavate the paid knowledge of expert under economic model and form expert
The server-side device of audio frequency album, as shown in Figure 3；Including：Knowledge server regulation engine unit 31, knowledge precipitation process list
Unit 32, payment processes unit 33, propagate communication unit 34, knowledge base 35, audio index processing unit 36.
The present invention provides one kind to be applied under the shared economic model of mobile Internet excavate the paid knowledge of expert and formed specially
The system of family's audio frequency album, as shown in Figure 4.Including：
Mobile client end device 41, server unit 42.
Bibliography explanation
Bibliography 1 patent application publication（Bulletin）Number CN104599167A《A kind of knowledge centered on knowledge services is handed over
Easily system and method》
Referenced patent document 2, Authorization Notice No. CN 103198828 B,《The construction method of corpus and system》,
Brief description
Fig. 1 is a kind of to excavate five kinds of angles that single human expert individual knowledge experience forms expert's audio frequency album based on mobile Internet
Color defines
Fig. 2 shares, based on mobile Internet, the mobile visitor excavating the paid knowledge of expert under economic model and forming expert's audio frequency album
Family end device
Fig. 3 is applied under the shared economic model of mobile Internet excavate the paid knowledge of expert and form the service of expert's audio frequency album
Device end device
Fig. 4 is applied under the shared economic model of mobile Internet excavate the paid knowledge of expert and form the system of expert's audio frequency album
Fig. 5 is based under the shared economic model of mobile Internet and excavates the paid knowledge of expert and form expert's audio frequency wall scroll Question Log
Embodiment schematic flow sheet
Fig. 6 is based on mobile Internet and shares under economic networks pattern, and a plurality of audio frequency Question Log is formed the enforcement of album by expert
Example schematic flow sheet
Specific embodiment
The embodiment of the present invention be will be described in connection with the drawings being shared based on mobile Internet of present invention offer and digs under network schemer
The pick paid knowledge of expert the method forming expert's audio frequency album, and fill in order to build the mobile client of paid knowledge sharing
Put, server unit, to realize whole system.
Hereinafter to introduce the embodiment of the present invention from every side.
Referring to Fig. 5, mobile Internet is shared and is excavated the paid knowledge of expert under economic networks pattern and form expert's audio frequency list
The embodiment flow process of bar Question Log is illustrated.
Step S01, knowledge is answered body 12 and is loaded activation audio sample unit 26, audio frequency in system mobile client end device 41
Sampling unit 26 exports audio frequency training text, allows knowledge answer body 12 and carries out recording sampling by input block 23, sets up oneself
Chinese acoustic model.Audio sample unit 26 is by using the voice unit (VU) of one group of voice signal sample, i.e. audio frequency training number
According to storehouse 27, to obtain the parameter of model, audio training data storehouse 27 is predominantly carried in the form of acoustic model from voice signal
The statistics taken out.
Step S02, audio sample unit 26 will check the dictionary file of the reflection mapping to voice unit (VU) sequence for the word,
Thus finding the corresponding Chinese language words signal related to voice unit (VU), provide transfer sequence and the word in audio training data storehouse 27
Allusion quotation file, realizes legal word to the mapping of voice unit (VU) sequence；And realize Unvoiced sounds fragment to accordingly noiseless or
The approximately sound sound tuple of body.Mainly precursor is used as the language dictionary file of system, Yi Jiyong during training
Accordingly noiseless or sound tuple that body is approximately sound is as excessively supplementary.After having trained, audio sample unit 26 runs decoder
The result of inspection training.Decoder first obtains an acoustic model, then tests a part of training data and reference model, finally
Estimate the model quality of training.
Step S03, knowledge is answered body 12 and is confirmed the model quality training out to set up the acoustic model of oneself, and is moving
Dynamic client terminal device 41 stores acoustic model parameters, in order to when subsequent voice carries out audio recording, audio sample unit is permissible
According to acoustic model reversal of identification voice and switch to text message.Meanwhile, mobile terminal answers body 12 to server update knowledge
There is audio search and set up the ability of album in the label of knowledge base 35 it is determined that knowledge answers body 12.
Step S04, puts question to body system to choose knowledge and answers body 12.Knowledge puts question to body 11 in mobile client end device 41,
By Indigenous knowledge regulation engine unit 21 and input block 23, search for ken interested, according to knowledge server
Regulation engine unit 31, searches for from knowledge base 35, provides speciality recommended information to give knowledge and puts question to body 11 to make reference, allows knowledge to carry
Asking that body 11 determines selects corresponding knowledge to answer body 12.
Step S05, knowledge puts question to body 11 to pass through the electronic payment unit 24 of mobile client end device 41, is answered according to knowledge
The amount of money of the corresponding personal value of body 12, is paid to platform by network communication unit 25 and puts question to the amount of money, and submit asked problem, system to
Knowledge server regulation engine unit 31 call knowledge precipitation process unit 32 confirm finish after, propagate communication unit 34 orient to
Expert reminds, and urges knowledge to answer body 12 and replies, and knowledge precipitation also can allow other in server refresh data with pushing platform 15
The Indigenous knowledge regulation engine unit 21 of the mobile client end device 41 of user, by calling content display control unit 22, shows
Knowledge puts question to the problem of body 11, and the expert info of consulting.
Step S06, knowledge answers the input block 23 that body 12 passes through system mobile client end device 41, carries out audio frequency and answers
Multiple, time control is carried out to local audio file, and adopts compression mechanism, reduced the upper cultivation time of mobile data, and adopted simultaneously
Audio sample unit 26 reversal of identification voice simultaneously switchs to text message, by network communication unit 25, by audio frequency and accordingly
Content of text is uploaded onto the server device 42, is encrypted priority assignation by knowledge precipitation process unit 32 to audio file,
It is used for distinguishing charges paid user and non-paid user, and file permission is set in the starting stage, only put question to body 11 to listen to for knowledge.
When knowledge puts question to body 11 to pass through mobile client end device 41, when accessing data, Indigenous knowledge regulation engine unit 21 is according to authority
Checking, is proved to be successful and can listen to this answer.
Step S07, question and answer read body 14, to specialist interest, read this problem exercise question it is believed that having important value,
It is ready to carry out the answer content that the small amount payment of platform sign answers body 12 to explore knowledge, by system mobile client end device
41 network communication unit 25 access server device 42, and after electronic payment unit 24 carries out paying success, refreshing service
This problem of device device 42 replies authority, server returning right limit marker, the Indigenous knowledge rule of mobile client end device 41
Engine unit 21 obtains authority so that content display control unit 22 can carry out content output, and that is, question and answer reading body 14 is permissible
Read answer.
Step S08, if question and answer are read body 14 and thought this topic and reply very valuable, can promote and share, that is, move
Dynamic client terminal device 41 passes through network communication unit 25 and submits push request to, allows server unit 42 to pass through to propagate communication unit 34
This question and answer is pushed, and stamps the label that question and answer read body 14, follow the tracks of the push result reading body 14.Server unit
42 knowledge precipitation process unit 32 also can record label, and question and answer are read body 14 and are also converted to question and answer carrier 13, in order to ask
Answer the statistics of carrier 13 revenue sharing below.
Step S09, knowledge precipitation can be retrieved knowledge and answer body 12 according to retrieval topic with the user pushing platform 15
Band authority reply, user is entered with server payment processes unit 33 by the electronic payment unit 24 of mobile client end device 41
Row interaction, after knowledge precipitation and the server unit 42 pushing platform 15 acknowledge receipt of the little amount of money of payment, that is, is converted into question and answer
Read body 14, you can obtain authority.Then mobile client end device 41 can be passed through, when accessing data, Indigenous knowledge regulation engine
Unit 21, according to Authority Verification, is proved to be successful, you can listen to this answer.
Step S010, reads body 14, the feedback of question and answer carrier 13 by calculating accumulative several question and answer, knowledge precipitates and pushes away
Send platform 15 to refresh the personal value that knowledge answers body 12, encourage knowledge to answer the shared knowledge of body 12 continuation and improve the personal value；I.e. according to this
Wheel question and answer, count the scoring that all up-to-date question and answer read body 14 and question and answer carrier 13 again, using appraisal procedure algorithm, right
The currently personal value, and knowledge puts question to body 11 to feed back, and the question and answer of a upper topic read the 1st positive and negative feedback power of body 14
Value, the 2nd positive and negative feedback weights, until n-th positive and negative feedback weights, carry out weights summation, give tacit consent to question and answer carrier simultaneously
Give for positive feedback, also all question and answer carriers 13 are carried out with weights and add up, added up, refresh knowledge and answer body 12
Personal value value.Evaluation result and knowledge are putd question to body 11, knowledge to answer body 12, question and answer carrier 13, question and answer reading body 14 by server
After their Attribute Association, and the audio frequency replying and identification text, it is stored in knowledge base 35, form a note of this question and answer
Record.The evaluation result of this question and answer can also as other users to this expert search when the sort by that carries out.
Referring to Fig. 5, mobile Internet is shared under economic networks pattern, and a plurality of audio frequency Question Log is formed album by expert
Embodiment flow process illustrate.
Step S11, knowledge answer body 12 sends speech audio by mobile client input block 23 and scans for, audio frequency
The acoustic model that the database 27 that sampling unit 26 invocation step S03 is trained in audio frequency is formed, carries out speech recognition, and passes through
Network communication unit 25 sends identified Similar Text and scans for server.
Step S12, after the audio index processing unit 36 of server obtains the Similar Text of identification, enters line retrieval, will know
Know the associated audio frequency in storehouse 35 the inside and text recalls, process through propagating communication unit 34, feed back to mobile client.
Step S13, the result that the association obtaining is fed back by mobile client content display control unit 22, at audio index
Reason unit 36 calculates text similarity, similar value highest is shown and carries out arrangement display up front.
Step S14, according to the display result of content display control unit 22, knowledge answers the newly-built special edition of body 12, shape
Become editor's demand, carry out the edit selection of album.Merging the multiple records chosen can be pre- by content display control unit 22
Look at, wherein every record comprises, add up the feedback that several question and answer read body 14 and question and answer carrier 13, knowledge precipitation and push
Platform 15 refreshes the attribute that knowledge answers multiple roles such as the personal value of body 12, and knowledge server regulation engine unit 31 calculates album
Price.
Step S15, knowledge answers body 12 by album edit validation, feeds back to server.Server audio index processes single
Unit 36, by the attribute data of the multiple roles in daily Question Log（Accumulative multiple question and answer read body 14, question and answer carrier 13
Feedback, often in wheel record, knowledge precipitation refreshes, with pushing platform 15, the personal value that knowledge answers body 12）Merge again and collect, set up rope
Draw, form album, be stored in knowledge base 35.
Finally it should be noted that：Above example only in order to technical scheme to be described, is not intended to limit；Although
With reference to previous embodiment, the present invention is carried out describing in detail, it will be understood by those within the art that：It still may be used
To modify to the technical scheme described in foregoing embodiments, or to wherein some technical characteristics equivalent；And this
The a little spirit of essence disengaging various embodiments of the present invention technical scheme and models changed or replace, do not make appropriate technical solution
Enclose.
Claims (6)
1. excavate the paid knowledge of expert under a kind of shared economic model based on mobile Internet and form the side of expert's audio frequency album
Method is it is characterised in that methods described includes：
A kind of five kinds of role definitions and payment and appraisal procedure, set up active, a paid knowledge sharing circle that is can sharing,
Encourage more people to participate in the enquirement of paid knowledge, reply, transmission, the knowledge and experience of reading shares activity；In above-mentioned role
Definition, there are this five kinds of roles in system：
Knowledge puts question to body, is responsible for puing question to whenever and wherever possible by mobile terminal in systems, by cleverly problem reasonable in design,
Carry out Extracting Knowledge and answer body；
Knowledge answers body, is responsible for being carried out audio frequency and returning within the long short time of regulation whenever and wherever possible by mobile terminal in systems
Answer, be set with the data mining object of rich experiences knowledge in systems, and according to system record knowledge sharing track,
To form the knowledge personal value of system evaluation；
Question and answer carrier, responsible reading, and carry out petit average, after approving that this puts question to the value of knowledge with question and answer, permissible
Carry out sharing the system activity of popularization；
Question and answer read body, only responsible reading, and carry out petit average, and enquirement body and problem body can be scored, but
Do not transmit, question and answer read body if recommending to come by carrier, and this carrier can participate in this petit average
It is divided into；
Paid knowledge question through excessive wheel, knowledge precipitates and pushes platform can answer body for master index using knowledge, to tire out
Family planning becomes this expert's audio frequency album corresponding, because audio database is differed with plain text database, in appeal model,
Expert is needed to be that knowledge answers body, in mobile client（I.e. smart mobile phone, panel computer etc.）, increase audio sample module and carry out
The simple sampled voice of many wheels, improves phonetic recognization rate, forms experience sampling, then again before knowledge answer body often takes turns answer, knows
Know answer body and carry out automatic speech recognition in mobile client startup audio sample module, while uploading audio frequency, corresponding
Similar Text also upload onto the server；When editing album, knowledge is answered body and can be scanned for by voice, to arrange volume
Row, thus form album.
2. the method as described in claim 1 it is characterised in that
Step S01, knowledge is answered body and is loaded activation audio sample unit in system mobile client end device, and audio sample unit is defeated
Go out audio frequency training text, allow knowledge answer body and recording sampling is carried out by input block, set up the Chinese acoustic model of oneself；Sound
Frequency sampling unit, by using the voice unit (VU) of one group of voice signal sample, i.e. audio training data storehouse, to obtain the ginseng of model
Number, the statistics that audio training data storehouse is predominantly extracted in the form of acoustic model from voice signal；
Step S02, audio sample unit will check the dictionary file of the reflection mapping to voice unit (VU) sequence for the word, thus looking for
To the corresponding Chinese language words signal related to voice unit (VU), provide transfer sequence and the dictionary file in audio training data storehouse,
Realize legal word to the mapping of voice unit (VU) sequence；And realize Unvoiced sounds fragment and approximately have to accordingly noiseless or body
The sound tuple of sound；Mainly precursor is used as the language dictionary file of system during training, and with accordingly noiseless
Or the approximately sound sound tuple of body is as excessively supplementary；After having trained, audio sample unit runs decoder to check training
Result；
Step S03, knowledge is answered body and is confirmed the model quality training out to set up the acoustic model of oneself, and in mobile client
End device stores acoustic model parameters, in order to when subsequent voice carries out audio recording, audio sample unit can be according to acoustics
Model reversal of identification voice simultaneously switchs to text message；Meanwhile, mobile terminal answers body in knowledge base to server update knowledge
Label is it is determined that knowledge answer body has audio search and sets up the ability of album；
Step S04, knowledge puts question to body in mobile client end device, by Indigenous knowledge regulation engine unit and input block,
Search for ken interested, according to knowledge server regulation engine unit, from knowledge base searching, provide speciality recommended information
Giving knowledge puts question to body to make reference, and allows knowledge put question to body to determine and selects corresponding knowledge to answer body；
Step S05, knowledge puts question to body to pass through the electronic payment unit of mobile client end device, answers body according to knowledge and corresponds to the personal value
The amount of money, paid to platform by network communication unit and put question to the amount of money, and submit asked problem to, system server knowledge rule
After engine unit calls knowledge precipitation process unit to confirm to finish, propagate communication unit and orient to expert's prompting, urge knowledge to return
Answer body to reply, knowledge precipitation also can allow the mobile client end device of other users in server refresh data with pushing platform
Indigenous knowledge regulation engine unit is by calling content display control unit, the problem of explicit knowledge's enquirement body, and consulting
Expert info；
Step S06, knowledge answers the input block that body passes through system mobile client end device, carries out Audio Reply, to this rock noise
Frequency file carries out time control, and adopts compression mechanism, reduces the upper cultivation time of mobile data, adopts audio sample unit simultaneously
Reversal of identification voice simultaneously switchs to text message, by network communication unit, audio frequency and corresponding content of text is uploaded to clothes
Business device device, priority assignation is encrypted to audio file by knowledge precipitation process unit, for distinguish charges paid user and
Non-paid user, and file permission is set in the starting stage, only put question to body to listen to for knowledge；When knowledge puts question to body to pass through mobile visitor
Family end device, when accessing data, Indigenous knowledge regulation engine unit, according to Authority Verification, is proved to be successful and can listen to this and answer
Multiple；
Step S07, question and answer read body, reading this problem exercise question it is believed that having important value, being ready to carry out the little of platform sign
Volume pays the answer content answering body to explore knowledge, by system mobile client plant network communication unit access server
Device, and after electronic payment unit carries out paying success, this problem of Refresh Servers device replies authority, server
Returning right limit marker, the Indigenous knowledge regulation engine unit of mobile client end device obtains authority so that content display control list
Unit can carry out content output；
Step S08, if question and answer read realization for this topic and replying very valuable, can promote and sharing, that is, in mobile client
End device is submitted to by network communication unit and is pushed request, allows server unit by propagating communication unit, this question and answer be carried out
Push, and stamp the label that question and answer read body, follow the tracks of the push result reading body；
Step S09, knowledge precipitation passes through electronic payment unit and the server of mobile client end device with the user pushing platform
Payment processes unit interacts, after knowledge precipitation and the server unit pushing platform acknowledge receipt of the little amount of money of payment, that is,
It is converted into question and answer and reads body, you can obtain authority；Then mobile client end device can be passed through, when accessing data, Indigenous knowledge is advised
Then engine unit, according to Authority Verification, is proved to be successful, you can listen to this answer；
Step S010, reads body, the feedback of question and answer carrier by calculating accumulative several question and answer, and knowledge precipitates and pushes platform brush
New knowledge answers the personal value of body, encourages knowledge to answer the shared knowledge of body continuation and improves the personal value；Server is by evaluation result and knowledge
After puing question to body, knowledge answer body, question and answer carrier, question and answer to read their Attribute Association of body, and the audio frequency replying is civilian with identification
This, be stored in knowledge base, forms a record of this question and answer.
3. the method as described in claim 1 and claim 2 it is characterised in that：
Step S11, knowledge answer body sends speech audio by mobile client input block and scans for, audio sample unit
The acoustic model that the data place that invocation step S03 is trained in audio frequency is formed, carries out speech recognition, and passes through network communication unit
Send identified Similar Text to scan for server；
Step S12, after the audio index processing unit of server obtains the Similar Text of identification, enters line retrieval, by knowledge base
The audio frequency that face is associated and text recall, and process through propagating communication unit, feed back to mobile client；
Step S13, the result that the association obtaining is fed back by mobile client content display control unit, audio index processing unit
Calculate text similarity, similar value highest is shown and carries out arrangement display up front；
Step S14, according to the display result of content display control unit, knowledge answers the newly-built special edition of body, and forming editor needs
Ask, carry out the edit selection of album；Merge multiple records of choosing and can pass through content display control unit preview, wherein every
Record comprises, and adds up the feedback that several question and answer read body and question and answer carrier, and knowledge precipitation is returned with pushing platform refreshing knowledge
Answer the attribute of multiple role such as personal value of body, knowledge server regulation engine unit calculates the price of album；
Step S15, knowledge answers body by album edit validation, feeds back to server；Server audio index processing unit, will be every
The attribute data of the multiple roles in its Question Log（Accumulative multiple question and answer read body, the feedback of question and answer carrier, often wheel record
Middle knowledge precipitation refreshes, with pushing platform 15, the personal value that knowledge answers body）Merge again and collect, set up index, form album, deposit
Enter knowledge base.
4. the method as described in claim 3 it is characterised in that：
Mobile Internet paid knowledge sharing network platform carries out knowledge sharing process assessment expert and shared Knowledge Creation valency
Value step：
Step S11, the knowledge precipitation process unit of server unit obtains from knowledge base searching, and sets the first of knowledge answerer
The beginning personal value, that is, according to last round of question and answer, counts the scoring of all up-to-date question and answer readers and question and answer transmitter, using assessment side
Method algorithm, to the current personal value, and knowledge quizmaster feedback, and the positive and negative feedback weights of the question and answer reader of a upper topic
Plus, carry out weights summation, simultaneously acquiescence question and answer transmitter give for positive feedback, also weights are carried out to all question and answer transmitters
Cumulative, added up, refreshed and proofread the up-to-date personal value value of knowledge answerer；
Step S12, when knowledge quizmaster is paid by the electronic payment unit of mobile client end device, server unit
Pay the amount of money that processing unit pays according to knowledge quizmaster 11, first collect a small amount of service charge in proportion, remaining sum is passed through electricity
Sub- payment unit is processed, and sends confirmation by propagating communication unit, when knowledge answerer accesses content display control unit
When, knowledge answerer can be pointed out to obtain income, knowledge answerer can obtain income by electronic payment unit；
Step S13, every new question and answer reader, little in payment when being paid by the electronic payment unit of mobile client end device
After volume reads the amount of money, this amount of money is pressed certain by knowledge precipitation and the knowledge server regulation engine unit pushing platform according to rule
Pro rate, to knowledge quizmaster, is propagated communication unit and is sent confirmation to mobile client end device, as knowledge answerer, carry
When the person of asking accesses content display control unit 22, Knowledge Acquirement income can be pointed out；
Step S14, if this question and answer reader is introduced by question and answer transmitter, distributes knowledge in above-mentioned steps S13
During the amount of money of quizmaster, knowledge server regulation engine unit can split out according to rule again and distribute a part of amount of money and give to ask
Answer transmitter, for encouraging Knowledge Sharing to transmit；
Step S15, adds up several question and answer readers, the feedback of question and answer transmitter by calculating, and knowledge precipitates and pushes platform brush
The personal value of new knowledge answerer, encourages knowledge answerer to continue shared knowledge and improves the personal value；I.e. according to epicycle question and answer, count again
All up-to-date question and answer readers and the scoring of question and answer transmitter, using appraisal procedure algorithm, to the current personal value, and knowledge carries
The person of asking feeds back, and the feedback weight of all question and answer readers of a upper topic carries out weights summation, and acquiescence question and answer pass simultaneously
The person of passing give for positive feedback, also all question and answer transmitters are carried out with weights and add up, added up, refresh knowledge answerer's
Personal value value, server, by after the Attribute Association of evaluation result and each role, is stored in knowledge base, and evaluation result can also conduct
The sort by carrying out when other users are to this expert search.
5. the method as described in claim 4 it is characterised in that：The present invention provides one kind to share economic mould based on mobile Internet
Excavate the paid knowledge of expert under formula and form the mobile client end device of expert's audio frequency album,
Including：Indigenous knowledge regulation engine unit, content display control unit, input block, electronic payment unit, network service
Unit, audio sample unit, the database of audio frequency training.
6. the method as described in claim 4 it is characterised in that：The present invention provides one kind to share economic mould based on mobile Internet
Excavate the paid knowledge of expert under formula and form the server-side device of expert's audio frequency album, including：Knowledge server regulation engine
Unit, knowledge precipitation process unit, payment processes unit, propagate communication unit, knowledge base, audio index processing unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610870493.3A CN106446205A (en) | 2016-10-01 | 2016-10-01 | Internet-based method and device for forming paid knowledgea of expert into audio compilation album |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610870493.3A CN106446205A (en) | 2016-10-01 | 2016-10-01 | Internet-based method and device for forming paid knowledgea of expert into audio compilation album |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106446205A true CN106446205A (en) | 2017-02-22 |
Family
ID=58171409
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610870493.3A Pending CN106446205A (en) | 2016-10-01 | 2016-10-01 | Internet-based method and device for forming paid knowledgea of expert into audio compilation album |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106446205A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107528822A (en) * | 2017-07-03 | 2017-12-29 | 阿里巴巴集团控股有限公司 | A kind of business performs method and device |
| CN111095305A (en) * | 2017-05-21 | 2020-05-01 | 舒辅医疗 | System and method for providing internet-based expert service |
| CN111192060A (en) * | 2019-12-23 | 2020-05-22 | 广州供电局有限公司 | Electric power IT service-based full-channel self-service response implementation method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050143999A1 (en) * | 2003-12-25 | 2005-06-30 | Yumi Ichimura | Question-answering method, system, and program for answering question input by speech |
| US20070168358A1 (en) * | 2005-12-15 | 2007-07-19 | Yong Chen | Method of rewarding value |
| CN101178718A (en) * | 2007-05-17 | 2008-05-14 | 腾讯科技（深圳）有限公司 | Knowledge sharing system, problem searching method and problem publish method |
| CN101620716A (en) * | 2009-08-07 | 2010-01-06 | 丘雷 | Payment real-time question answering system and method |
| CN101630397A (en) * | 2009-08-21 | 2010-01-20 | 蓝建栋 | Information exchange system |
| JP2012141782A (en) * | 2010-12-28 | 2012-07-26 | Yahoo Japan Corp | Knowledge search system and reward application method |
| CN103208092A (en) * | 2013-05-03 | 2013-07-17 | 闻静波 | Internet-based system for providing consulting information |
-
2016
- 2016-10-01 CN CN201610870493.3A patent/CN106446205A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050143999A1 (en) * | 2003-12-25 | 2005-06-30 | Yumi Ichimura | Question-answering method, system, and program for answering question input by speech |
| US20070168358A1 (en) * | 2005-12-15 | 2007-07-19 | Yong Chen | Method of rewarding value |
| CN101178718A (en) * | 2007-05-17 | 2008-05-14 | 腾讯科技（深圳）有限公司 | Knowledge sharing system, problem searching method and problem publish method |
| CN101620716A (en) * | 2009-08-07 | 2010-01-06 | 丘雷 | Payment real-time question answering system and method |
| CN101630397A (en) * | 2009-08-21 | 2010-01-20 | 蓝建栋 | Information exchange system |
| JP2012141782A (en) * | 2010-12-28 | 2012-07-26 | Yahoo Japan Corp | Knowledge search system and reward application method |
| CN103208092A (en) * | 2013-05-03 | 2013-07-17 | 闻静波 | Internet-based system for providing consulting information |
Non-Patent Citations (1)
| Title |
|---|
| 许森: ""知识零售平台"分答"的问题与思考"", 《新闻研究导刊》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111095305A (en) * | 2017-05-21 | 2020-05-01 | 舒辅医疗 | System and method for providing internet-based expert service |
| CN111095305B (en) * | 2017-05-21 | 2023-11-03 | 舒辅医疗 | System and method for providing internet-based expert service |
| CN107528822A (en) * | 2017-07-03 | 2017-12-29 | 阿里巴巴集团控股有限公司 | A kind of business performs method and device |
| CN111192060A (en) * | 2019-12-23 | 2020-05-22 | 广州供电局有限公司 | Electric power IT service-based full-channel self-service response implementation method |
| CN111192060B (en) * | 2019-12-23 | 2024-01-02 | 广州供电局有限公司 | Full-channel self-service response implementation method based on power IT service |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9202219B2 (en) | System and method to merge pay-for-performance advertising models | |
| Callegaro et al. | Online panel research | |
| CN107066466A (en) | Conversation type problem and answer | |
| US20180316636A1 (en) | Context-aware conversational assistant | |
| US7606726B2 (en) | Interactive survey and data management method and apparatus | |
| US10204343B2 (en) | Multi-channel customer identification | |
| US20120265574A1 (en) | Creating incentive hierarchies to enable groups to accomplish goals | |
| CN106651329A (en) | Method and device for mining individual paid knowledge based on mobile Internet for sharing | |
| CN101292282A (en) | Mobile systems and methods of supporting natural language human-machine interactions | |
| US20090006083A1 (en) | Systems And Methods For Spoken Information | |
| CN106446205A (en) | Internet-based method and device for forming paid knowledgea of expert into audio compilation album | |
| CN109684532A (en) | Legal advice service system based on casebook | |
| CN109670947A (en) | A kind of patent operation plateform system and application method based on patent application installment business | |
| CN113190671A (en) | Intelligent on-duty system for legal problems of sharing law institute | |
| US20200081915A1 (en) | System and method for automatically collecting opinions | |
| KR20190115398A (en) | Messenger-based call center service method and system using cube chain | |
| KR20010018633A (en) | Method For Opinion Survey Through Internet &CTI | |
| CN112395323A (en) | Interaction method based on user experience data | |
| CN109274899A (en) | The method and device of the paid knowledge video album of expert is formed based on internet | |
| Garfield et al. | Champion networks in federated interorganizational systems: Case studies in telemedicine | |
| CN112200606B (en) | Popularization information design method, device, equipment and medium | |
| Meeks | Small Business Market Share Growth: Case Study | |
| TWI744620B (en) | Pooling prediction system | |
| CN102947823B (en) | Conversation type problem and answer | |
| CN110264370B (en) | Mutual-help project expense settlement system, method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |
Application publication date: 20170222 |
|
| WD01 | Invention patent application deemed withdrawn after publication |