CN107230142A - Method and device, method of commerce and system based on speech production order - Google Patents
Method and device, method of commerce and system based on speech production order Download PDFInfo
- Publication number
- CN107230142A CN107230142A CN201710566431.8A CN201710566431A CN107230142A CN 107230142 A CN107230142 A CN 107230142A CN 201710566431 A CN201710566431 A CN 201710566431A CN 107230142 A CN107230142 A CN 107230142A
- Authority
- CN
- China
- Prior art keywords
- voice
- commodity
- word
- rule
- order
- 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
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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0633—Lists, e.g. purchase orders, compilation or processing
- G06Q30/0635—Processing of requisition or of purchase orders
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Abstract
The invention discloses the method and device based on speech production order, method of commerce and system, electronic equipment, it is related to speech recognition, technical field of intelligent equipment, wherein, based on the method for speech production order, including：Obtain voice and be used as target voice；The target voice is converted into target text；At least one demand for commodity information is extracted from the target text；Every the demand for commodity information includes：One or more of the unique appellation of commodity, commodity amount, amount of money；The present invention can realize light and flexible, removable, low cost, liberate the both hands of cashier, even without artificial cashier, can shorten exchange hour, transaction efficiently and accurately is carried out.
Description
Technical field
The present invention relates to speech recognition, technical field of intelligent equipment, more particularly to a kind of side based on speech production order
Method and device, method of commerce and system.
Background technology
The industries such as retail, service in the prior art, user and businessman have to pass through requirement validation, place an order, pay, making
Making the links such as sorting packing could complete once to merchandise.By taking Xian Xia eating and drinking establishment as an example, specifically need to perform following steps：
Buyer tells the demand of businessman oneself；Businessman operates cashier's machine, typing demand, and forms order；Cashier is oral to explain to buyer
Order amount of money request buyer is stated to confirm, or simultaneously by the electronic curtain display order amount of money；After the oral confirmation of buyer is errorless, business is selected
A kind of means of payment that family provides is paid the bill (such as cash, brush bank card, Alipay, wechat)；Businessman, which is manually entered, to pay
The money amount of money (paying in cash) checks cashier's machine (payment of swiping the card) or checks that buyer confirms phone whether pay successfully (wechat, branch
Pay the Third-party payments such as treasured), if paying successfully, sorting packing, or in the process by another people sorting packing.
More than there are multiple problems in this process, for example：
User and businessman have to pass through requirement validation, place an order, pay, make sorting packing etc. multiple links could complete one
Secondary transaction, each link is required for the artificial interaction of user and cashier, very slow due to manually performing relative computer execution
Slowly, therefore cause process of exchange to expend the time long, buyer queues up length, the problems such as purchase experiences are poor；
Because existing transaction system is generally desk-top and fixed, therefore buyer must reach specified cash register position ability in fact
Now pay, be unfavorable for being traded in mobile scene；
Requirement validation, place an order, payment process occupies the hand of cashier, causes cashier to make the works such as sorting packing
Make, it is necessary to which the independent position for opening up a sorter, adds manpower and time cost；
When buyer selects when being paid by the way of scanning the Quick Response Code that businessman provides, when buyer is paid successfully, to need
Successfully interface is paid to cashier's displaying, cashier pays successfully by manually checking to pay successfully interface and just can confirm that, this
One process expends the time of cashier, but does not check the problems such as businessman will undertake user's non-honest loss；
Retailer, service provider, especially small businessman is to cost extreme sensitivity, although some existing transaction systems can be at certain
Realized in a little links and accelerate speed, but transaction system cost is higher, and also small businessman provides the space place an order, pay, delivering
It is very limited, it is impossible to which that huge, complicated transaction system is installed；
In environment, source of sound is more, noise is more, and buyer's speech habits are widely different, cause existing voice to receive
Silver-colored scheme disposably can not clearly know buyer's demand, it is necessary to the problems such as confirming repeatedly.
The content of the invention
(1) goal of the invention
It is an object of the invention to provide a kind of method and device based on speech production order, method of commerce and system, electricity
Sub- equipment, the present invention can realize light and flexible, removable, low cost, liberate the both hands of cashier, even without artificial receipts
Silver-colored member, can shorten exchange hour, transaction efficiently and accurately is carried out.
(2) technical scheme
To solve the above problems, the first aspect of the present invention provides a kind of method based on speech production order, including：
Obtain voice and be used as target voice；The target voice is converted into target text；At least one is extracted from the target text
Bar demand for commodity information；Every the demand for commodity information includes：One in the unique appellation of commodity, commodity amount, the amount of money or
It is multiple；Based on the cargo price that prestores and at least one demand for commodity information extracted and counted and calculated, to generate
Order.
According to another aspect of the present invention there is provided a kind of device based on speech production order, including：Voice collecting
Module, is transmitted to sound identification module for gathering voice as target voice, and by the target voice；Speech recognition mould
Block, for the target voice to be converted into target text；Extraction module, for extracting at least one from the target text
Demand for commodity information；Every the demand for commodity information includes：One or many in the unique appellation of commodity, commodity amount, the amount of money
It is individual；Order generation module, for based on the cargo price that prestores and at least one demand for commodity information extracted and being united
Meter and calculating, to generate order.
According to another aspect of the invention there is provided a kind of method of commerce, including：Using described in claim based on language
The method generation order of sound generation order；Control message reminding module play and/or display reminding paying party displaying one-dimension code and/
Or the prompting message of Quick Response Code；Obtain one-dimension code and/or Quick Response Code, and by the one-dimension code and/or Quick Response Code and the order
Send to payment system；Receive payment system feedback payment success message after, control message reminding module play and/or
Display completes to pay prompting message.
According to another aspect of the invention there is provided a kind of transaction system, including：It is described based on speech production order
Device；Also include control module and message notifying module；The control module, for after the voice is waken up or order is generated
The first hint instructions are sent to message notifying module；It is additionally operable to after the payment success message of payment system feedback is received to disappearing
Cease reminding module and send the second hint instructions；
The message notifying module, for the broadcasting after first hint instructions are received and/or display reminding payment
The prompting message of side's displaying payment code；It is additionally operable to play and/or show after second hint instructions are received to complete payment
Prompting message.
According to another aspect of the invention there is provided a kind of method of commerce, including：It is described based on speech production order
Method generates order；Bill is generated based on the order, and Quick Response Code and/or one-dimension code are generated based on the bill, and by institute
State one-dimension code and/or Quick Response Code sends to display module and shown；The display module display reminding paying party is controlled to scan
The prompting message of the one-dimension code and/or Quick Response Code；And/or control playing module present the bill for payment side scan the one-dimension code and/
Or the prompting message of Quick Response Code；After the payment success message of payment system feedback is received, control display module display is completed
Pay prompting message；And/or control playing module completes to pay prompting message.
According to another aspect of the invention there is provided a kind of transaction system, including it is described based on speech production order
Device；Also include control module and display module；The control module, for generating bill based on the order, and based on institute
Bill generation Quick Response Code and/or Quick Response Code are stated, and the Quick Response Code and/or Quick Response Code are exported to display module；It is additionally operable to aobvious
Show that module and/or playing module send the first control instruction；It is additionally operable to receiving the payment success message of payment system feedback
Afterwards, the second control instruction is sent to display module and/or playing module；The display module, for show the Quick Response Code and/
Or Quick Response Code；It is additionally operable to scan the one-dimension code and/or Quick Response Code according to the first control instruction display reminding paying party
Prompting message；It is additionally operable to complete to pay prompting message according to second control instruction display；Playing module, for according to described
First control instruction play cuing paying party scans the prompting message of the one-dimension code and/or Quick Response Code；It is additionally operable to according to described
Second control instruction finishes playing payment prompting message.
According to another aspect of the invention there is provided a kind of electronic equipment, including：At least one processor, and with institute
State memory, microphone and the earphone of at least one processor communication connection；Wherein, have can be by described one for the memory storage
The instruction of individual computing device, the instruction is by least one described computing device, to cause at least one described processor
It is able to carry out the described method based on speech production order.
(3) beneficial effect
The above-mentioned technical proposal of the present invention has following beneficial technique effect：
Businessman or user is allowed to say demand for commodity information with predetermined rule, during analysis voice, using default corpus
And identification model, so that the process of speech recognition is more precisely and quick, it can quickly be automatically based upon speech production and order
It is single, it is to avoid each link of transaction is required for the artificial interaction of user and cashier, trading efficiency is improved, when having saved transaction
Between, it is to avoid queuing up long causes the problems such as purchase experiences are poor.
One subband has the earphone or microphone and a smart mobile phone that can be networked of microphone and its application to complete whole
Individual process.Specifically, voice is gathered by earphone or microphone and transmits voice to mobile phone, at mobile phone localization or high in the clouds
Speech production order is managed, can be to branch after the payment license for the camera acquisition Quick Response Code/one-dimension code having by mobile phone in itself
The system of paying is initiated to pay, and completes transaction.Order and transaction results can be played as the loudspeaker built in mobile phone or earphone, can also be by
The screen display of mobile phone.So as to realize the removable words of cash register, without fixed cashier, and it is cheap, it is special without buying
The hardware device of industry.Cost has been saved while mobility is realized.
One cashier can complete requirement validation simultaneously, place an order, pays, the work such as sorting packing be made, without needing list
The position of a sorter is solely opened up, human cost has been saved；
When paying successfully, system shows or played automatically payment successful result, without cashier by manually checking payment
Success interface, which just can confirm that, to be paid successfully, saves program and time；
Personalized identification model and corpus in system, renewal can be trained automatically according to the language habits, it is to avoid language
It is not high, it is necessary to the problems such as confirming repeatedly that sound is accustomed to the widely different language identification accuracy rate caused.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the method based on speech production order disclosed in the embodiment of the present invention；
Fig. 2 is the module relationship schematic diagram of the device based on speech production order disclosed in the embodiment of the present invention；
Fig. 3 is the schematic flow sheet of method of commerce provided in an embodiment of the present invention；
Fig. 4 is the schematic flow sheet of transaction system provided in an embodiment of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention of greater clarity, with reference to embodiment and join
According to accompanying drawing, the present invention is described in more detail.It should be understood that these descriptions are merely illustrative, and it is not intended to limit this hair
Bright scope.In addition, in the following description, the description to known features and technology is eliminated, to avoid unnecessarily obscuring this
The concept of invention.
The embodiment of the invention discloses a kind of method and device based on speech production order, one kind includes described based on language
The method of commerce and system of the method for sound generation order, and a kind of electronic equipment.In embodiments of the present invention, order is generated
Process is that the voice based on user or businessman carries out automatic identification generation order, and without businessman, to commodity, scanning record is merchandised one by one
System, substantially reduces exchange hour and saves transaction cost.It is described in detail individually below：
Referring to Fig. 1, Fig. 1 is a kind of flow signal of method based on speech production order disclosed in the embodiment of the present invention
Figure.Wherein, the method based on speech production order shown in Fig. 1 goes for smart mobile phone (such as Android phone, iOS hands
Machine etc.), tablet personal computer, palm PC, desktop computer, cashier's machine etc. be mobile or non-mobile electronic equipment.The electronic equipment includes
At least one processor and the memory being connected with least one processor communication, in addition to it is logical with least one described processor
Believe the sound input-output equipment of connection.Audio input device can use wire microphone, wireless microphone, wire microphone
The equipment such as array, wireless microphone array.Audio output device can be using equipment such as wired earphone, wireless headset, sound equipments.Institute
State memory storage have can by the instruction of one computing device, the instruction by least one described computing device,
To cause at least one described processor to be able to carry out following steps：
S11, obtains voice and is used as target voice.
The voice that the voice of acquisition can say for buyer, or the voice that cashier says.Buyer or seller say
During voice, to contain pre-defined rule.The pre-defined rule includes：The unique appellation of commodity or the amount of money；The unique appellation of commodity, commodity amount
The two phrase constituted with any permutation and combination method；The unique appellation of commodity, commodity amount, unit three are with any permutation and combination
The phrase that mode is constituted.Wherein, the unique appellation of commodity can be trade name, for example：The conducts such as apple, ham sausage, * board milk
The unique appellation of commodity.Default trade name code name can also be used, such as the unique appellation of commodity.For example, pre-setting
1011 be the code name of apple, then 1011 be apple this commodity the unique appellation of commodity.It should be noted that businessman can basis
Itself needs, and self-defined use trade name is as the unique appellation of commodity, or using code name as the unique appellation of commodity, can be with
Both of which is used.Commodity amount refers to the natural numbers for being used to count article number such as 1,2,3.Unit includes：Bottle, it is individual, only,
, the Chinese unit noun such as bar, jin.The example that one group given below of (can also be English) says voice with pre-defined rule：
Please pay 30 yuan of the amount of money (the single amount of money)；(the unique appellation of commodity, the default amount of the unique appellation of commodity without quantity is apple
1)；No. 1011 commodity (the only unique appellation of particular commodity)；Apple 1, or 1 apple (appoint by the unique appellation of commodity, both commodity amounts
Meaning arrangement)；1 jin of apple, or 1 jin of apple (the unique appellation of commodity, commodity amount, unit three arbitrary arrangement).
Further, the amount of money, quantity and unit can also use pre-defined code name, such as apple 11, represent apple
One jin.1 represents 10 pieces.
S12, target text is converted to by target voice.
After target voice is obtained, analysis identification is carried out to the target voice to be translated into text.By voice
The process for being translated into text is related to wave sample, Fourier transformation, input neural network recognization etc. and obtains single character.It is many
Individual character passes through language model estimation output text.This analysis recognition method is highly developed, and analysis identification process
Itself it is not the inventive point of the present invention, therefore will not be repeated here.But it should be recognized that relative to prior art, the present invention exists
To in the analysis identification process of voice, in addition to needing to use public corpus of the prior art, in order that must recognize more
Plus accurately and quickly, in addition it is also necessary to use default proprietary corpus.The proprietary corpus have collected what is commonly used under shopping environment
Trade name, commodity amount and unit, and other modification words.
It is preferred that, except proprietary corpus, present invention additionally comprises to public identification model and corpus based on further changing
Enter the personalized identification model and corpus to be formed.Personalized identification model and corpus are used to overcome because of dialect, tell word habit
Used, working environment limitation causes the difference of pronunciation.Businessman and user can test or train the system before using the system.Example
After the unique appellation of businessman's typing commodity, quantity, unit, wake-up word, closing catalogue, the rule given according to system, or
Customized rule, repeats multiple voice contents, system can according to recitations therein redjustment and modification identification model and corpus, until
The speech recognition of businessman reaches certain threshold value.The use of businessman each time all can further change its personalized identification model and
Corpus.
Corpus, identification model, text analyzing algorithm etc. can be locally stored in electronic equipment, so as to by target voice
The step of being converted to target text can be performed by the electronic equipment of front end.Further, it is also possible that electronic equipment energy
Enough have access to the corpus that is stored with, identification model, the high in the clouds of text analyzing algorithm so that target voice is converted into target text
The step of can be performed by high in the clouds, and implementing result is fed back to the electronic equipment of front end.
S13, extracts at least one demand for commodity information from target text.Every demand for commodity information at least includes：Business
One or more of the unique appellation of product, commodity amount, amount of money, can also include unit.
A demand for commodity information is potentially included in one target text, it is also possible to including a plurality of demand for commodity information.Often
Bar merchandise news can be：The unique appellation of commodity；Can also be the amount of money；Can also be the unique appellation of commodity, both commodity amounts
The phrase constituted with any permutation and combination method；It can also be the unique appellation of commodity, commodity amount, unit three with any arrangement
The phrase that combination is constituted.For example, a demand for commodity information can be：5 bottles of milk, or：Four bread；May be used also
Think：5 yuan of strawberries.
S14, based on the cargo price that prestores and at least one demand for commodity information extracted and is counted and is calculated,
To generate order.
Cargo price is by the advance typing of businessman and stores in systems, is needed having extracted all commodity from target text
Ask after information, the amount of money of every demand for commodity information is calculated based on cargo price, all demand for commodity information are counted afterwards
Total amount, obtains order.
In another embodiment of the present invention, on the basis of the foregoing method based on speech production order, wherein,
Step S11 obtains voice as target voice：Target voice is used as based on Rule voice is obtained.Obtain rule bag
Include：Wake up word+closing rule, wake-up word+end rules, the one kind waken up in word+regular three kinds of rules of payment request in limited time
Or a variety of combinations.Wherein, it is the default word of setting to wake up word and closing.For example, waking up word can be：Start, open, open
The words such as dynamic, start；Can also be trade name, for example：Apple, milk, mango.Closing can be：Terminate, stop, closing
The word such as close, terminate.
Waking up word+closing rule is：Using get wake up word at the time of as one section of voice start node.It will call out
The end node waken up at the time of getting closing after word as this section of voice.The voice waken up between word and closing is made
For the target voice of an order.
Wake up word+end rules of prescribing a time limit：Using get wake up word when as one section of voice start node.Wake up word it
Start to detect whether the time for not getting voice reaches predetermined threshold afterwards, will current time conduct if reaching predetermined threshold
The end node of this section of voice.Using wake up word and reach predetermined threshold at the time of between voice as an order target language
Sound.
Waking up word+payment request rule is：Using get wake up word at the time of as one section of voice start node.Will be
End node at the time of paying request as this section of voice is got after wake-up word.Word will be waken up and payment request is got
At the time of between voice as an order target voice.It can be one-dimension code, Quick Response Code or triggering command to pay request.
It is preferred that, obtaining rule also includes：After wake-up word is got first, if obtaining wake-up word again, judging should
Wake up word invalid.Waking up word can wake up from mobile phone dormancy, be directly entered using speech recognition is started, can be with mobile phone in work
In work, wake up application and start speech recognition.
It should be noted that in the method that this inventive embodiments is provided, sound identification module is not to be constantly in work
Make state, but in a dormant state.Working condition is switched to from resting state after wake-up word is listened to, starts to obtain language
Sound；Get closing or pay request or do not get time of voice and reach and switched to from working condition after predetermined threshold
Resting state, and stop obtaining voice.It is to reduce the energy consumption of equipment using the beneficial effect of working condition+resting state, prolongs
Long stand-by time.It will be appreciated by those skilled in the art that utilizing manual palpation, the mode of interaction, it is clear that application can also be waken up
And equipment.
In another embodiment of the present invention, on the basis of the foregoing method based on speech production order, wherein,
Step S11 obtains voice to be included as target voice：In the analysis identification process of text is converted speech into, based on a point single gauge
Then using the part or all of target voice as an order in target voice.Single gauge is divided then to include：Start mark rule, knot
One or more of beam identification rule and the mark that starts+end of identification three kinds of rules of rule are combined.Wherein, start to identify and tie
Beam identification is setting preset characters or word.For example, starting mark can be：The words such as beginning, unlatching, startup, start；
Can be trade name, for example：Apple, milk, mango.End of identification can be：The words such as end, stopping, closing, termination.
Starting mark rule is：In the analysis identification process of text is converted speech into, two are started between mark
Target voice as an order target voice.
End of identification rule is：In the analysis identification process of text is converted speech into, by between two end of identification
Target voice as an order target voice.
Starting mark+end of identification rule is：In the analysis identification process of text is converted speech into, mark will be started
Target voice between end of identification as an order target voice.
It should be noted that in the method that this inventive embodiments is provided, sound identification module is constantly in working condition,
There is no resting state.I.e. sound identification module is always in ceaselessly typing and identification voice.In text analyzing identification process,
According to start mark and/or end of identification divide order.
In another embodiment of the present invention, on the basis of the foregoing method based on speech production order, wherein,
Step S12 switchs to target text from target voice to be included：Wave sample, Fourier transformation, input neural network recognization etc. are obtained
Single character.Multiple characters pass through language model estimation output text.The text of target voice output is target text.
In another embodiment of the present invention, on the basis of the foregoing method based on speech production order, wherein,
Step S13 extracts at least one demand for commodity information from target text：Carried based on extracting rule from target text
Take at least one demand for commodity information.Extracting rule comprises the following steps：
S131, participle is carried out to target text and part of speech is demarcated, the word segmentation result with part of speech label is obtained.
S132, the word and/or phrase for meeting default typing rule are extracted based on default typing rule from word segmentation result
It is used as extraction result.Wherein, typing rule includes：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts with
The phrase that any permutation and combination method is constituted；The unique appellation of commodity, commodity amount, unit three are with any permutation and combination method structure
Into phrase.
In this inventive embodiments, if having mixed the content unrelated with demand for commodity information in target voice, only record
Relevant content is not processed as demand for commodity information, other irrelevant contents.Due to being related to the phonological component of demand for commodity information
There is clear and definite rule, i.e.,：Unique appellation+commodity amount+the unit of commodity is according to the coherent input of typing rule.For example, " opening
Begin, one jin of apple, that smacker, inside, one jin of mango." in text-processing, carry out analytic statistics in strict accordance with rule,
I.e. " one jin of apple " " one jin of mango " meet typing rule be designated as useful text, other texts do not keep a record and handled.
In another embodiment of the present invention, extracting rule can also be：In default commodity information database retrieval with
The word and/or phrase matched in target text, if the match is successful, regard the language and/or phrase that the match is successful as extraction
As a result.Wherein, default commodity information database is included：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts with
The phrase that any permutation and combination method is constituted；The unique appellation of commodity, commodity amount, unit three are with any permutation and combination method structure
Into phrase.
In another embodiment of the present invention, wherein, extracting rule also includes：Obtain a demand for commodity information it
Afterwards, judge to whether there is modification word after current commodity demand information, if there is modification word, perform following steps：
S133, the word and/or phrase for meeting default typing rule are extracted based on default typing rule from word segmentation result
It is used as demand for commodity modification information.
S134, is modified based on demand for commodity modification information to demand for commodity information.
Wherein, modification word includes：Increase word, reduce word, the one or more cancelled in word.Increase word can include：Also
, to increase again, add, add, plus etc. word；Reducing word can include：It the word such as, should not remove, reduce, subtracting, cutting
Language.Cancelling word can include:Cancel an order, delete the words such as order.
For example, one be related to demand for commodity information target text for " one jin of apple, then add half jin, half jin of mango should not
It ", wherein, " again plus half jin " and " shoulding not it " is the demand for commodity modification for an above nearest demand for commodity information
Information.
In another embodiment of the present invention, after an order is generated, in addition to intercept whether get modification
Word, if getting, performs following steps：
S14, modification order voice is used as based on acquisition Rule voice.
Acquisition rule can be with foregoing wake-up word+closing rule, wake-up word+, and end rules of prescribing a time limit, wake-up word+payment please
One or more combinations in regular three kinds of rules are asked, it is necessary to which explanation, wake-up word herein replaces with modification word.Modification
Word can have the function of wake-up.
S15, modification sequence information is extracted based on extracting rule from modification order voice.
S16, is modified based on modification sequence information to the order of generation, to generate new order.
Referring to Fig. 2, Fig. 2 is a kind of module relationship of the device based on speech production order disclosed in the embodiment of the present invention
Schematic diagram.
The embodiment of the present invention is used to perform the above-mentioned method based on speech production order.Wherein, the device shown in Fig. 2 can be with
Including but not limited to smart mobile phone (such as Android phone, iOS mobile phones), tablet personal computer, palm PC, desktop computer, cash register
The movement such as machine or non-moving equipment.As shown in figure 3, being somebody's turn to do the device based on speech production order includes：Voice acquisition module, language
Sound identification module, extraction module and order generation module.
Voice acquisition module, is transmitted to sound identification module for gathering voice as target voice, and by target voice.
The voice that the voice that voice acquisition module is obtained can say for buyer, or the voice that cashier says.Buyer sells
When family says voice, pre-defined rule is used.The pre-defined rule includes：The unique appellation of commodity or the amount of money；The unique appellation of commodity, commodity
The phrase that both quantity is constituted with any permutation and combination method；The unique appellation of commodity, commodity amount, unit three are with any arrangement
The phrase that combination is constituted.Wherein, the unique appellation of commodity can be trade name, such as apple, ham sausage, * board milk
It is used as the unique appellation of commodity.Default trade name code name can also be used, such as the unique appellation of commodity.For example, in advance
1011 code names for being apple are set, then 1011 be apple this commodity the unique appellation of commodity.It should be noted that businessman can be with
According to their needs, self-defined use trade name is gone back as the unique appellation of commodity, or using code name as the unique appellation of commodity
It can be used with both of which.Commodity amount refers to the natural numbers for being used to count article number such as 1,2,3.Unit includes：Bottle, it is individual,
Only, the Chinese unit noun such as bar, jin.The example that one group given below of (can also be English) says voice with pre-defined rule
Son：Please pay 30 yuan of the amount of money (the single amount of money)；Apple (the unique appellation of commodity, the default amount of the unique appellation of commodity without quantity
For 1)；No. 1011 commodity (the only unique appellation of particular commodity)；Apple 1, or 1 apple (the unique appellation of commodity, both commodity amounts
Arbitrary arrangement)；1 jin of apple, or 1 jin of apple (the unique appellation of commodity, commodity amount, unit three arbitrary arrangement).
Further, the amount of money, quantity and unit can also use pre-defined code name, such as apple 11, represent apple
One jin.1 represents 10 pieces.
Sound identification module, for target voice to be converted into target text.Voice acquisition module is obtaining target voice
Afterwards, target voice is sent to sound identification module, sound identification module carries out analysis identification to the target voice to incite somebody to action
It is converted into text.Speech translation is related to wave sample, Fourier transformation, input neural network recognization into the process of text
Etc. obtaining single character.Multiple characters pass through language model estimation output text.This analysis recognition method very into
It is ripe, and analysis identification process is not in itself the inventive point of the present invention, therefore will not be repeated here.But it should be recognized that relative
In prior art, the present invention is in the analysis identification process to voice, except needing to use public corpus of the prior art
Outside, in order that must recognize more accurately and quickly, in addition it is also necessary to use default proprietary corpus.The proprietary corpus have collected
Trade name, commodity amount and the unit commonly used under shopping environment, and other modification words.
It is preferred that, except proprietary corpus, present invention additionally comprises to public identification model and corpus based on further changing
Enter the personalized identification model and corpus to be formed.Personalized identification model and corpus are used to overcome because of dialect, tell word habit
Used, working environment limitation causes the difference of pronunciation.Businessman and user can test or train the system before using the system.Example
After the unique appellation of businessman's typing commodity, quantity, unit, wake-up word, closing catalogue, the rule given according to system, or
Customized rule, repeats multiple voice contents, system can according to recitations therein redjustment and modification identification model and corpus, until
The speech recognition of businessman reaches certain threshold value.The use of businessman each time all can further change its personalized identification model and
Corpus.
Extraction module, for extracting at least one demand for commodity information from target text.Every demand for commodity packet
Include：One or more of the unique appellation of commodity, commodity amount, amount of money.A commodity are potentially included in one target text to be needed
Seek information, it is also possible to including a plurality of demand for commodity information.Every merchandise news can be：The unique appellation of commodity；It can also be gold
Volume；It can also be the phrase that the unique appellation of commodity, both commodity amounts are constituted with any permutation and combination method；It can also be commodity
The phrase that unique appellation, commodity amount, unit three are constituted with any permutation and combination method.For example, a demand for commodity information
Can be：5 bottles of milk, or：Four bread；Can also be 5 yuan of strawberries.
Order generation module, for based on the cargo price that prestores and at least one demand for commodity information extracted and entering
Row statistics and calculating, to generate order.Cargo price is by the advance typing of businessman and stores in systems, is being carried from target text
Take after all demand for commodity information, the amount of money of every demand for commodity information is calculated based on cargo price, institute is counted afterwards
There is the total amount of demand for commodity information, obtain order.
In another embodiment of the present invention, on the basis of the foregoing device based on speech production order, wherein,
Voice acquisition module is based on acquisition Rule voice and is used as target voice.Obtaining rule includes：Wake-up word+closing rule,
Wake up word+end rules, one or more combinations waken up in word+regular three kinds of rules of payment request in limited time.Wherein, wake up
Word and closing are the default word of setting.For example, waking up word can be：The words such as beginning, unlatching, startup, start；Can also
For trade name, for example：Apple, milk, mango.Closing can be：The words such as end, stopping, closing, termination.
Waking up word+closing rule is：Using get wake up word at the time of as one section of voice start node.It will call out
The end node waken up at the time of getting closing after word as this section of voice.The voice waken up between word and closing is made
For the target voice of an order.
Wake up word+end rules of prescribing a time limit：Using get wake up word when as one section of voice start node.Wake up word it
Start to detect whether the time for not getting voice reaches predetermined threshold afterwards, will current time conduct if reaching predetermined threshold
The end node of this section of voice.Using wake up word and reach predetermined threshold at the time of between voice as an order target language
Sound.
Waking up word+payment request rule is：Using get wake up word at the time of as one section of voice start node.Will be
End node at the time of paying request as this section of voice is got after wake-up word.Word will be waken up and payment request is got
At the time of between voice as an order target voice.
It is preferred that, obtaining rule also includes：After wake-up word is got first, if obtaining wake-up word again, judging should
Wake up word invalid.
It should be noted that in the device that this inventive embodiments is provided, voice collecting and sound identification module are not
Working condition is constantly in, but usually in a dormant state.After wake-up word is listened to work shape is switched to from resting state
State, starts to obtain voice；Get closing or pay request or do not get voice time reach after predetermined threshold from
Working condition switches to resting state, and stops obtaining voice.It is to reduce using the beneficial effect of working condition+resting state
The energy consumption of equipment, extends stand-by time.It will be appreciated by those skilled in the art that utilizing manual palpation, the mode of interaction, it is clear that
Application and equipment can be waken up.
In another embodiment of the present invention, on the basis of the foregoing device based on speech production order, wherein,
Sound identification module, is additionally operable in the analysis identification process of text is converted speech into, based on point single gauge then by target voice
In the part or all of target voice as an order.Single gauge is divided then to include：Start mark rule, end of identification rule and
Start one or more of mark+end of identification three kinds of rules of rule to combine.Wherein, it is setting to start mark and end of identification
Preset characters or word.For example, starting mark can be：The words such as beginning, unlatching, startup, start；It can also be trade name
Claim, for example：Apple, milk, mango.End of identification can be：The words such as end, stopping, closing, termination.
Starting mark rule is：In the analysis identification process of text is converted speech into, two are started between mark
Target voice as an order target voice.
End of identification rule is：In the analysis identification process of text is converted speech into, by between two end of identification
Target voice as an order target voice.
Starting mark+end of identification rule is：In the analysis identification process of text is converted speech into, mark will be started
Target voice between end of identification as an order target voice.
It should be noted that in the device that this inventive embodiments is provided, voice acquisition module and sound identification module one
It is straight in running order, without resting state.I.e. voice acquisition module is always in ceaselessly collection voice, sound identification module one
Directly ceaselessly recognizing voice.In text analyzing identification process, just according to start mark and/or end of identification divide order.
In another embodiment of the present invention, on the basis of the foregoing method based on speech production order, wherein,
Extraction module extracts at least one demand for commodity information based on extracting rule from target text.Extracting rule includes：
S131, participle is carried out to target text and part of speech is demarcated, the word segmentation result with part of speech label is obtained.
S132, is extracted from word segmentation result based on default typing rule and meets default typing rule and in pre-set text language material
Word present in storehouse and/or phrase are used as extraction result.Typing rule includes：The unique appellation of commodity or the amount of money；Commodity are unique
The phrase that appellation, both commodity amounts are constituted with any permutation and combination method；The unique appellation of commodity, commodity amount, unit three
The phrase constituted with any permutation and combination method.
In this inventive embodiments, if having mixed the content unrelated with demand for commodity information in target voice, only record
Relevant content is not processed as demand for commodity information, other irrelevant contents.Due to being related to the phonological component of demand for commodity information
There is clear and definite rule, i.e.,：Unique appellation+commodity amount+the unit of commodity is according to the coherent input of typing rule.For example, " opening
Begin, one jin of apple, that smacker, inside, one jin of mango." in text-processing, carry out analytic statistics in strict accordance with rule,
I.e. " one jin of apple " " one jin of mango " meet typing rule be designated as useful text, other texts do not keep a record and handled.
In another embodiment of the present invention, extracting rule can also be：In default commodity information database retrieval with
The word and/or phrase matched in target text, if the match is successful, regard the language and/or phrase that the match is successful as extraction
As a result.Default commodity information database is included：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts are with any
The phrase that permutation and combination method is constituted；The unique appellation of commodity, commodity amount, unit three are constituted with any permutation and combination method
Phrase.
In another embodiment of the present invention, extracting rule also includes：After a demand for commodity information is obtained, sentence
With the presence or absence of modification word after disconnected current commodity demand information, if there is modification word, following steps are performed：Based on default typing
Rule extracts the word and/or phrase that meet default typing rule as demand for commodity modification information from word segmentation result.It is based on
Demand for commodity modification information is modified to demand for commodity information.Wherein, modification word includes：Increase word, reduce word, cancel in word
One or more.Increase word can include：Also, to increase again, add, add, plus etc. word；Reducing word can wrap
Include：It the word such as, should not remove, reduce, subtracting, cutting.Cancelling word can include:Cancel an order, delete the words such as order.Example
Such as, one be related to demand for commodity information target text for " one jin of apple, then add half jin, half jin of mango should not it ", its
In, " again plus half jin " and " shoulding not it " is the demand for commodity modification information for an above nearest demand for commodity information.
In another embodiment of the present invention, on the foregoing device basic based on speech production order, in addition to：
Editing order module 14：For after generation order, in addition to intercepting whether get modification word, if getting, being based on
Rule voice is obtained as modification order voice, and modification order letter is extracted from modification order voice based on extracting rule
Breath, is modified based on modification sequence information to the order of generation, to generate new order afterwards.
Referring to Fig. 3, Fig. 3 is a kind of schematic flow sheet of method of commerce provided in an embodiment of the present invention.
Wherein, the method for commerce shown in Fig. 3 goes for smart mobile phone (such as Android phone, iOS mobile phones), put down
The movement such as plate computer, palm PC, desktop computer, cashier's machine or non-mobile electronic equipment.The electronic equipment includes at least one
Processor, and the memory being connected with least one processor communication, and be connected with least one described processor communication
The input-output equipment such as earphone, microphone, sound equipment, display, scanner.Wherein, it is stored with the memory of electronic equipment
With computer instruction form exist transaction system and payment system, computer instruction by least one computing device so that
Obtain at least one processor and be able to carry out method of commerce as described below：
S1, order is generated using the foregoing method based on speech production order.
S2, control message reminding module is played and/or display reminding paying party shows the prompting of one-dimension code and/or Quick Response Code
Message.
Specifically, transaction system sends control instruction to the message notifying module communicated to connect with electronic equipment, to control
Message notifying module is played and/or display reminding paying party shows the prompting message of one-dimension code and/or Quick Response Code.Message notifying mould
Block can show the prompting message of one-dimension code and/or Quick Response Code using sound equipment come play cuing paying party, it would however also be possible to employ display
Device carrys out the prompting message of display reminding paying party displaying one-dimension code and/or Quick Response Code.Can also jointly it be reminded using two kinds of equipment.
Prompting message can be presented in forms such as sound, word, pictures.
S3, obtains one-dimension code and/or Quick Response Code, and one-dimension code and/or Quick Response Code and order are sent to payment system.
Buyer is after the prompting message shown by message notifying module and/or played is heard or seen, to seller's (cash register
Member) the payment code for being shown in its mobile phone terminal is shown, payment code can be one-dimension code or Quick Response Code.Seller is using sweeping
Retouch device and be directed at payment code, payment code is obtained to scan.Scanner sends payment code to the transaction system for communicating connection
System, so that transaction system obtains payment code.
S4, after the payment success message of payment system feedback is received, control message reminding module is played and/or shown
Complete to pay prompting message.
Whether transaction system will obtain payment code and send to payment system, succeeded by payment system validation of payment.If paying
Successful, then payment system sends to transaction system and pays success message.Transaction system is after receiving and paying success message, to disappearing
Cease reminding module and send control instruction, play and/or show that completion pays prompting message with control message reminding module.
In a kind of method of commerce that another embodiment of the present invention is provided, method of commerce comprises the following steps：
S01, order is generated using the foregoing method based on speech production order.
S02, bill is generated based on order, and generates Quick Response Code and/or one-dimension code based on bill, and by one-dimension code and/or
Quick Response Code sends to display module and shown, is paid for scanning input.
Specifically, transaction system is based on order and generates bill, and the bill includes the account address of businessman.Generate bill
Afterwards, Quick Response Code and/or one-dimension code are generated based on bill, and referred to control is sent with the display module that electronic equipment is communicated to connect
Order, to control display module to show the one-dimension code and/or Quick Response Code.User, which opens, pays application scanning Quick Response Code and/or one
Code is tieed up, is paid to businessman.
The prompting message of S03, control display module display reminding paying party scanning one-dimension code and/or Quick Response Code.And/or control
Scan the prompting message of one-dimension code and/or Quick Response Code in record and broadcast amplification module present the bill for payment side.
Transaction system sends control instruction to the display module that electronic equipment is communicated to connect, to control display module to show
Scan the prompting message of one-dimension code and/or Quick Response Code in present the bill for payment side.Or
Transaction system sends control instruction to the playing module that electronic equipment is communicated to connect, to control playing module to play
Scan the prompting message of one-dimension code and/or Quick Response Code in present the bill for payment side.
S04, after the payment success message of payment system feedback is received, control display module display completes to pay prompting
Message, and/or control playing module complete to pay prompting message.
Buyer starts the one-dimension code and/or two shown by mobile phone scanning display module after prompting message is heard or seen
Tie up code, mobile phone is in scanning to being connected to corresponding payment system after the one-dimension code and/or Quick Response Code with to businessman's Pay Bill.Branch
Pay whether system validation of payment succeeds.If paying successfully, payment system sends to transaction system and pays success message.Transaction system
System receive pay success message after, to message notifying module send control instruction, with control message reminding module play and/
Or display completes to pay prompting message.
Referring to Fig. 4, Fig. 4 is a kind of schematic flow sheet of transaction system provided in an embodiment of the present invention.
Wherein, the transaction system shown in Fig. 4 with computer instruction form be stored in smart mobile phone (such as Android phone,
IOS mobile phones etc.), tablet personal computer, palm PC, desktop computer, cashier's machine etc. be mobile or memory of non-mobile electronic equipment in.
The electronic equipment includes at least one processor and the memory, the memory and processor communication connection, the electricity
Sub- equipment is also including the earphone being connected with least one described processor communication, microphone, sound equipment, display, camera, scanning
The input-output equipment such as device.Wherein, be also stored with the payment system existed in computer instruction form in the memory of electronic equipment
System.
Transaction system includes：The foregoing device based on speech production order.Also include control module and message notifying mould
Block.Wherein, control module, for sending the first hint instructions to message notifying module after voice is waken up or order is generated.Also
For sending the second hint instructions to message notifying module after the payment success message of payment system feedback is received.Message is carried
Show module, for the broadcasting after the first hint instructions are received and/or the prompting message of display reminding paying party displaying payment code.
It is additionally operable to play and/or show that completion pays prompting message after the second hint instructions are received.
In the transaction system that another embodiment of the present invention is provided, transaction system includes foregoing based on speech production order
Device, in addition to control module and display module.Wherein, control module is used to generate bill based on order, and based on bill
Quick Response Code and/or Quick Response Code are generated, and Quick Response Code and/or Quick Response Code are exported to display module；Be additionally operable to display module and/or
Playing module sends the first control instruction；It is additionally operable to after the payment success message of payment system feedback is received, to display mould
Block and/or playing module send the second control instruction.Display module is used for two-dimensional code display and/or Quick Response Code；It is additionally operable to basis
First control instruction display reminding paying party scans the prompting message of one-dimension code and/or Quick Response Code；It is additionally operable to according to the second control
Instruction display completes to pay prompting message.Playing module is used to scan one-dimension code according to the first control instruction play cuing paying party
And/or the prompting message of Quick Response Code；It is additionally operable to be finished playing payment prompting message according to the second control instruction.
It should be appreciated that the above-mentioned embodiment of the present invention is used only for exemplary illustration or explains the present invention's
Principle, without being construed as limiting the invention.Therefore, that is done without departing from the spirit and scope of the present invention is any
Modification, equivalent substitution, improvement etc., should be included in the scope of the protection.In addition, appended claims purport of the present invention
Covering the whole changes fallen into scope and border or this scope and the equivalents on border and repairing
Change example.
Claims (19)
1. a kind of method based on speech production order, it is characterised in that including：
Obtain voice and be used as target voice；
The target voice is converted into target text；
At least one demand for commodity information is extracted from the target text；Every the demand for commodity information includes：Commodity are only
One or more of one appellation, commodity amount, unit, amount of money；
Based on the cargo price that prestores and at least one demand for commodity information extracted and counted and calculated, ordered with generating
It is single.
2. according to the method described in claim 1, wherein, based on obtain Rule voice be used as target voice；It is described to obtain
Rule includes：Wake up word+closing rule, wake-up word+end rules of prescribing a time limit, wake up in word+regular three kinds of rules of payment request
One or more combinations；
It is described wake-up word+closing rule be：Using get wake up word at the time of as one section of voice start node；Will institute
State and wake up end node at the time of getting closing after word as this section of voice；By between the wake-up word and closing
Voice as an order target voice；
The wake-up word+end rules of prescribing a time limit：Using get wake up word when as one section of voice start node；Waken up described
Start to detect whether the time for not getting voice reaches predetermined threshold after word, if reaching predetermined threshold, by current time
It is used as the end node of this section of voice；It regard the voice between at the time of the wake-up word and arrival predetermined threshold as an order
Target voice；
Wake-up word+payment request the rule is：Using get wake up word at the time of as one section of voice start node；Will be
The end node as this section of voice at the time of paying request is got after the wake-up word；Wake up word by described and get
Pay request at the time of between voice as an order target voice.
3. according to the method described in claim 1, wherein, it is described acquisition voice include as target voice：Digitizing the speech into
For in the analysis identification process of text, based on point single gauge then using part or all of in the target voice as order
Target voice；Described point of single gauge then includes：Start mark rule, end of identification rule and start mark+end of identification rule three
One or more of rule is planted to combine；
It is described beginning mark rule be：In the analysis identification process of text is converted speech into, two are started between mark
Target voice as an order target voice；
The end of identification rule is：In the analysis identification process of text is converted speech into, by between two end of identification
Target voice as an order target voice；
Mark+end of identification the rule that starts is：In the analysis identification process of text is converted speech into, mark will be started
Target voice between end of identification as an order target voice.
4. according to the method described in claim 1, wherein, at least one business is extracted from the target text based on extracting rule
Product demand information；The extracting rule includes：
Participle is carried out to the target text and part of speech is demarcated, the word segmentation result with part of speech label is obtained；
The default typing regular word and/or phrase are met based on default typing rule middle extract described in word segmentation result
It is used as extraction result；
The typing rule includes：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts are with any arrangement group
The phrase that conjunction mode is constituted；The phrase that the unique appellation of commodity, commodity amount, unit three are constituted with any permutation and combination method.
5. according to the method described in claim 1, wherein, at least one business is extracted from the target text based on extracting rule
Product demand information；The extracting rule includes：
Retrieval and the word and/or phrase that match in the target text in default commodity information database, if the match is successful,
Then it regard the language and/or phrase that the match is successful as extraction result；
The default commodity information database is included：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts with appoint
The phrase that one permutation and combination method is constituted；The unique appellation of commodity, commodity amount, unit three are constituted with any permutation and combination method
Phrase.
6. the method according to claim 4 or 5, wherein, the extracting rule also includes：Obtaining a demand for commodity letter
After breath, judge to whether there is modification word after current commodity demand information, if there is modification word, perform following steps：
Extracted based on the default typing rule from the word segmentation result and meet the regular word of the default typing and/or short
Language is used as demand for commodity modification information；
The demand for commodity information is modified based on the demand for commodity modification information.
7. the method according to claim any one of 1-6, wherein, after an order is generated, in addition to intercept whether
Modification word is got, if getting, following steps are performed：
Modification order voice is used as based on acquisition Rule voice；
Modification sequence information is extracted from the modification order voice based on the extracting rule；
The order of the generation is modified based on the modification sequence information, to generate new order.
8. a kind of device based on speech production order, it is characterised in that including：
Voice acquisition module, is transmitted to sound identification module for gathering voice as target voice, and by the target voice；
Sound identification module, for the target voice to be converted into target text；
Extraction module, for extracting at least one demand for commodity information from the target text；The every demand for commodity letter
Breath includes：One or more of the unique appellation of commodity, commodity amount, unit, amount of money；
Order generation module, for based on the cargo price that prestores and at least one demand for commodity information extracted and being united
Meter and calculating, to generate order.
9. device according to claim 8, wherein,
The voice acquisition module is based on acquisition Rule voice and is used as target voice；
The acquisition rule includes：Wake up word+closing rule, wake-up word+end rules, wake-up word+payment request rule in limited time
Then one or more combinations in three kinds of rules；
It is described wake-up word+closing rule be：Using get wake up word at the time of as one section of voice start node；Will institute
State and wake up end node at the time of getting closing after word as this section of voice；By between the wake-up word and closing
Voice as an order target voice；
The wake-up word+end rules of prescribing a time limit：Using get wake up word when as one section of voice start node；Waken up described
Start to detect whether the time for not getting voice reaches predetermined threshold after word, if reaching predetermined threshold, by current time
It is used as the end node of this section of voice；It regard the voice between at the time of the wake-up word and arrival predetermined threshold as an order
Target voice；
Wake-up word+payment request the rule is：Using get wake up word at the time of as one section of voice start node；Will be
The end node as this section of voice at the time of paying request is got after the wake-up word；Wake up word by described and get
Pay request at the time of between voice as an order target voice.
10. device according to claim 8, wherein,
The sound identification module, is additionally operable in the analysis identification process of text is converted speech into, and then will based on point single gauge
The part or all of target voice as an order in the target voice；Described point of single gauge then includes：Start mark rule
Then, end of identification is regular and starts the combination of one or more of mark+end of identification three kinds of rules of rule；
It is described beginning mark rule be：In the analysis identification process of text is converted speech into, two are started between mark
Target voice as an order target voice；
The end of identification rule is：In the analysis identification process of text is converted speech into, by between two end of identification
Target voice as an order target voice；
Mark+end of identification the rule that starts is：In the analysis identification process of text is converted speech into, mark will be started
Target voice between end of identification as an order target voice.
11. device according to claim 8, wherein, the extraction module is based on extracting rule from the target text
Extract at least one demand for commodity information；The extracting rule includes：
Participle is carried out to the text and part of speech is demarcated, the word segmentation result with part of speech label is obtained；
The default typing rule is met and in the default text based on default typing rule middle extract described in word segmentation result
Word present in this corpus and/or phrase are used as extraction result；
The typing rule includes：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts are with any arrangement group
The phrase that conjunction mode is constituted；The phrase that the unique appellation of commodity, commodity amount, unit three are constituted with any permutation and combination method.
12. device according to claim 8, wherein, the extraction module is based on extracting rule from the target text
Extract at least one demand for commodity information；The extracting rule includes：
Retrieval and the word and/or phrase that match in the target text in default commodity information database, if the match is successful,
Then it regard the language and/or phrase that the match is successful as extraction result；
The default commodity information database is included：The unique appellation of commodity or the amount of money；The unique appellation of commodity, both commodity amounts with appoint
The phrase that one permutation and combination method is constituted；The unique appellation of commodity, commodity amount, unit three are constituted with any permutation and combination method
Phrase.
13. the device according to claim 11 or 12, wherein, the extracting rule also includes：Needed obtaining a commodity
Ask after information, judge to whether there is modification word after current commodity demand information, walked if existing below modification word, execution
Suddenly：
Extracted based on the default typing rule from the word segmentation result and meet the regular word of the default typing and/or short
Language is used as demand for commodity modification information；
The demand for commodity information is modified based on the demand for commodity modification information.
14. the device according to claim any one of 8-13, in addition to：
Editing order module：For after generation order, in addition to intercepting whether get modification word, if getting, base
In acquisition Rule voice as modification order voice, and extracted based on the extracting rule from the modification order voice
Change sequence information；The order of the generation is modified based on the modification sequence information, to generate new order.
15. a kind of method of commerce, it is characterised in that including：
Order is generated using the method based on speech production order described in claim any one of 1-7；
Control message reminding module is played and/or display reminding paying party shows the prompting message of one-dimension code and/or Quick Response Code；
One-dimension code and/or Quick Response Code are obtained, and the one-dimension code and/or Quick Response Code and the order are sent to payment system；
After the payment success message of payment system feedback is received, control message reminding module is played and/or display completes branch
Pay prompting message.
16. a kind of transaction system, it is characterised in that including：Described in claim any one of 8-14 based on speech production order
Device；Also include control module and message notifying module；
The control module, refers to for sending the first prompting to message notifying module after the voice is waken up or order is generated
Order；It is additionally operable to after the payment success message of payment system feedback is received send the second hint instructions to message notifying module；
The message notifying module, for being played and/or display reminding paying party exhibition after first hint instructions are received
Show the prompting message of payment code；It is additionally operable to play and/or show that completing payment points out after second hint instructions are received
Message.
17. a kind of method of commerce, it is characterised in that including：
Order is generated using the method based on speech production order described in claim any one of 1-7；
Bill is generated based on the order, and Quick Response Code and/or one-dimension code are generated based on the bill, and by the one-dimension code
And/or Quick Response Code sends to display module and shown；
The display module display reminding paying party is controlled to scan the prompting message of the one-dimension code and/or Quick Response Code；And/or control
Scan the prompting message of the one-dimension code and/or Quick Response Code in record and broadcast amplification module present the bill for payment side；
After the payment success message of payment system feedback is received, control display module display completes to pay prompting message；With/
Or control playing module completes to pay prompting message.
18. a kind of transaction system, it is characterised in that including described in claim any one of 8-14 based on speech production order
Device；Also include control module and display module；
The control module, for generating bill based on the order, and generates Quick Response Code and/or two dimension based on the bill
Code, and export the Quick Response Code and/or Quick Response Code to display module；It is additionally operable to send the to display module and/or playing module
One control instruction；It is additionally operable to after the payment success message of payment system feedback is received, to display module and/or playing module
Send the second control instruction；
The display module, for showing the Quick Response Code and/or Quick Response Code；It is additionally operable to be shown according to first control instruction
Scan the prompting message of the one-dimension code and/or Quick Response Code in present the bill for payment side；It is additionally operable to be shown according to second control instruction
Complete to pay prompting message；
Playing module, for scanning the one-dimension code and/or Quick Response Code according to the first control instruction play cuing paying party
Prompting message；It is additionally operable to be finished playing payment prompting message according to second control instruction.
19. a kind of electronic equipment, it is characterised in that including：At least one processor, and it is logical with least one described processor
Believe memory, microphone and the earphone of connection；Wherein,
The memory storage has can be by the instruction of one computing device, and the instruction is by least one described processor
Perform, with cause at least one described processor be able to carry out described in claim any one of 1-7 based on speech production order
Method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710566431.8A CN107230142A (en) | 2017-07-12 | 2017-07-12 | Method and device, method of commerce and system based on speech production order |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710566431.8A CN107230142A (en) | 2017-07-12 | 2017-07-12 | Method and device, method of commerce and system based on speech production order |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107230142A true CN107230142A (en) | 2017-10-03 |
Family
ID=59957465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710566431.8A Pending CN107230142A (en) | 2017-07-12 | 2017-07-12 | Method and device, method of commerce and system based on speech production order |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107230142A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108564357A (en) * | 2018-03-29 | 2018-09-21 | 上海掌门科技有限公司 | A kind of method of payment, equipment, system and storage medium based on voice control |
| CN109102824A (en) * | 2018-07-06 | 2018-12-28 | 北京比特智学科技有限公司 | Voice error correction method and device based on human-computer interaction |
| CN109191225A (en) * | 2018-07-02 | 2019-01-11 | 阿里巴巴集团控股有限公司 | Order generation method, device, order processing method and server |
| CN109191269A (en) * | 2018-10-23 | 2019-01-11 | 杭州引灯网络科技有限公司 | A kind of commodity trading system based on interactive voice |
| CN109726372A (en) * | 2017-10-31 | 2019-05-07 | 百度在线网络技术（北京）有限公司 | Generation method, device and the computer-readable medium of work order based on message registration |
| CN109801137A (en) * | 2019-01-18 | 2019-05-24 | 泰康保险集团股份有限公司 | Commercial articles ordering method, product ordering apparatus, storage medium and electronic equipment |
| CN109817217A (en) * | 2019-01-17 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Self-service based on speech recognition peddles method, apparatus, equipment and medium |
| CN109948974A (en) * | 2019-02-27 | 2019-06-28 | Oppo广东移动通信有限公司 | A kind of material management system and its order generation method, server, storage medium |
| CN109993634A (en) * | 2019-04-12 | 2019-07-09 | 睿驰达新能源汽车科技（北京）有限公司 | A kind of method and device based on speech production order |
| WO2019242519A1 (en) * | 2018-06-22 | 2019-12-26 | 阿里巴巴集团控股有限公司 | Commodity check and apparatus, and device |
| CN111063340A (en) * | 2019-12-09 | 2020-04-24 | 用友网络科技股份有限公司 | Service processing method and device of terminal, terminal and computer readable storage medium |
| CN111326154A (en) * | 2020-03-02 | 2020-06-23 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and electronic equipment |
| CN111401878A (en) * | 2020-03-16 | 2020-07-10 | 虫极科技（北京）有限公司 | Checkout method, checkout system and computer readable storage medium |
| CN111652611A (en) * | 2020-05-27 | 2020-09-11 | 维沃移动通信有限公司 | Payment method and electronic equipment |
| CN111738707A (en) * | 2020-05-29 | 2020-10-02 | 陈旺新 | Transaction amount identification method, payment method and computer device |
| WO2020232617A1 (en) * | 2019-05-20 | 2020-11-26 | 深圳市欢太科技有限公司 | Voice information processing method and apparatus, and electronic device and storage medium |
| CN112735063A (en) * | 2019-10-28 | 2021-04-30 | 虫极科技（北京）有限公司 | Checkout method, system and storage medium |
| CN113344682A (en) * | 2017-12-26 | 2021-09-03 | 北京星选科技有限公司 | Order processing method and device, electronic equipment and computer readable storage medium |
| CN111738707B (en) * | 2020-05-29 | 2024-04-30 | 陈旺新 | Transaction amount identification method, payment method and computer device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120284130A1 (en) * | 2011-05-05 | 2012-11-08 | Ebay, Inc. | Barcode checkout at point of sale |
| CN103943105A (en) * | 2014-04-18 | 2014-07-23 | 安徽科大讯飞信息科技股份有限公司 | Voice interaction method and system |
| WO2016037415A1 (en) * | 2014-09-11 | 2016-03-17 | 中兴通讯股份有限公司 | Mobile payment method, system and device, and computer storage medium |
| CN106056424A (en) * | 2015-12-21 | 2016-10-26 | 小菜儿成都信息科技有限公司 | Agricultural product voice rapid ordering system |
| CN106611306A (en) * | 2015-10-22 | 2017-05-03 | 乐点行动支付股份有限公司 | Trading method for inputting commodity price by utilizing voice to perform payment |
| CN106653021A (en) * | 2016-12-27 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Voice wake-up control method and device and terminal |
| CN106782554A (en) * | 2016-12-19 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Voice awakening method and device based on artificial intelligence |
-
2017
- 2017-07-12 CN CN201710566431.8A patent/CN107230142A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120284130A1 (en) * | 2011-05-05 | 2012-11-08 | Ebay, Inc. | Barcode checkout at point of sale |
| CN103943105A (en) * | 2014-04-18 | 2014-07-23 | 安徽科大讯飞信息科技股份有限公司 | Voice interaction method and system |
| WO2016037415A1 (en) * | 2014-09-11 | 2016-03-17 | 中兴通讯股份有限公司 | Mobile payment method, system and device, and computer storage medium |
| CN106611306A (en) * | 2015-10-22 | 2017-05-03 | 乐点行动支付股份有限公司 | Trading method for inputting commodity price by utilizing voice to perform payment |
| CN106056424A (en) * | 2015-12-21 | 2016-10-26 | 小菜儿成都信息科技有限公司 | Agricultural product voice rapid ordering system |
| CN106782554A (en) * | 2016-12-19 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Voice awakening method and device based on artificial intelligence |
| CN106653021A (en) * | 2016-12-27 | 2017-05-10 | 上海智臻智能网络科技股份有限公司 | Voice wake-up control method and device and terminal |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109726372A (en) * | 2017-10-31 | 2019-05-07 | 百度在线网络技术（北京）有限公司 | Generation method, device and the computer-readable medium of work order based on message registration |
| CN109726372B (en) * | 2017-10-31 | 2023-06-30 | 上海优扬新媒信息技术有限公司 | Method and device for generating work order based on call records and computer readable medium |
| CN113344682A (en) * | 2017-12-26 | 2021-09-03 | 北京星选科技有限公司 | Order processing method and device, electronic equipment and computer readable storage medium |
| CN108564357A (en) * | 2018-03-29 | 2018-09-21 | 上海掌门科技有限公司 | A kind of method of payment, equipment, system and storage medium based on voice control |
| WO2019242519A1 (en) * | 2018-06-22 | 2019-12-26 | 阿里巴巴集团控股有限公司 | Commodity check and apparatus, and device |
| CN109191225A (en) * | 2018-07-02 | 2019-01-11 | 阿里巴巴集团控股有限公司 | Order generation method, device, order processing method and server |
| CN109102824A (en) * | 2018-07-06 | 2018-12-28 | 北京比特智学科技有限公司 | Voice error correction method and device based on human-computer interaction |
| CN109102824B (en) * | 2018-07-06 | 2021-04-09 | 北京比特智学科技有限公司 | Voice error correction method and device based on man-machine interaction |
| CN109191269A (en) * | 2018-10-23 | 2019-01-11 | 杭州引灯网络科技有限公司 | A kind of commodity trading system based on interactive voice |
| CN109817217A (en) * | 2019-01-17 | 2019-05-28 | 深圳壹账通智能科技有限公司 | Self-service based on speech recognition peddles method, apparatus, equipment and medium |
| CN109801137A (en) * | 2019-01-18 | 2019-05-24 | 泰康保险集团股份有限公司 | Commercial articles ordering method, product ordering apparatus, storage medium and electronic equipment |
| CN109948974A (en) * | 2019-02-27 | 2019-06-28 | Oppo广东移动通信有限公司 | A kind of material management system and its order generation method, server, storage medium |
| CN109993634A (en) * | 2019-04-12 | 2019-07-09 | 睿驰达新能源汽车科技（北京）有限公司 | A kind of method and device based on speech production order |
| WO2020232617A1 (en) * | 2019-05-20 | 2020-11-26 | 深圳市欢太科技有限公司 | Voice information processing method and apparatus, and electronic device and storage medium |
| CN112735063A (en) * | 2019-10-28 | 2021-04-30 | 虫极科技（北京）有限公司 | Checkout method, system and storage medium |
| CN111063340A (en) * | 2019-12-09 | 2020-04-24 | 用友网络科技股份有限公司 | Service processing method and device of terminal, terminal and computer readable storage medium |
| CN111326154A (en) * | 2020-03-02 | 2020-06-23 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and electronic equipment |
| CN111326154B (en) * | 2020-03-02 | 2022-11-22 | 珠海格力电器股份有限公司 | Voice interaction method and device, storage medium and electronic equipment |
| CN111401878A (en) * | 2020-03-16 | 2020-07-10 | 虫极科技（北京）有限公司 | Checkout method, checkout system and computer readable storage medium |
| CN111401878B (en) * | 2020-03-16 | 2023-12-01 | 虫极科技（北京）有限公司 | Checkout method, checkout system and computer readable storage medium |
| CN111652611A (en) * | 2020-05-27 | 2020-09-11 | 维沃移动通信有限公司 | Payment method and electronic equipment |
| CN111738707A (en) * | 2020-05-29 | 2020-10-02 | 陈旺新 | Transaction amount identification method, payment method and computer device |
| CN111738707B (en) * | 2020-05-29 | 2024-04-30 | 陈旺新 | Transaction amount identification method, payment method and computer device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230142A (en) | Method and device, method of commerce and system based on speech production order | |
| CN109360550B (en) | Testing method, device, equipment and storage medium of voice interaction system | |
| CN108305626A (en) | The sound control method and device of application program | |
| US8001469B2 (en) | Automatic generation of interactive systems from a formalized description language | |
| US7110963B2 (en) | Point-of-sale customer order system utilizing an unobtrusive transmitter/receiver and voice recognition software | |
| CN107451893A (en) | A kind of method, high in the clouds information processor and the system of voice shopping | |
| CN109389971A (en) | Insurance recording quality detecting method, device, equipment and medium based on speech recognition | |
| CN109658191A (en) | A kind of member's purchase system and method based on language identification and recognition of face | |
| CN108460591A (en) | Payment processing method and device, method of commerce and mobile device | |
| CN107357787A (en) | Semantic interaction method, apparatus and electronic equipment | |
| CN109086026A (en) | Broadcast the determination method, apparatus and equipment of voice | |
| CN112233690B (en) | Double recording method, device, terminal and storage medium | |
| CN110335596A (en) | Products Show method, apparatus, equipment and storage medium based on speech recognition | |
| CN110020908A (en) | Server, electronic equipment, control device, control method and storage medium | |
| CN107104994A (en) | Audio recognition method, electronic installation and speech recognition system | |
| CN106611306A (en) | Trading method for inputting commodity price by utilizing voice to perform payment | |
| CN110933225A (en) | Call information acquisition method and device, storage medium and electronic equipment | |
| JP5532781B2 (en) | Waiting service server, waiting service system using the server, and expected end time calculation method for waiting service | |
| CN107688935A (en) | Payment devices, method and device, system, medium and movement, electronic equipment | |
| KR20190084033A (en) | Information processing apparatus and information processing method | |
| CN111161729A (en) | Voice interaction method and device for intelligent self-service equipment | |
| KR20210003639A (en) | Customer management device and customer management method | |
| JP6550111B2 (en) | Display device, display method and program | |
| CN115271932A (en) | Outbound risk identification method and device | |
| CN111508142A (en) | Sight voice interaction automatic cigarette vending machine |
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 |