CN110659970A

CN110659970A - Account information processing method and device based on voice recognition and electronic equipment

Info

Publication number: CN110659970A
Application number: CN201810601038.2A
Authority: CN
Inventors: 傅星雅; 吴天昊; 刘玉忠
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Baidu Online Network Technology Beijing Co Ltd; Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2018-06-12
Filing date: 2018-06-12
Publication date: 2020-01-07

Abstract

The invention discloses an accounting information processing method and device based on voice recognition and electronic equipment, wherein the method comprises the following steps: when the vacant necessary slot filling field exists in the task template corresponding to the type of the financial intention according to the text information corresponding to the voice information input by the user, the user is prompted by voice to supplement the vacant necessary slot filling field, the content of the vacant necessary slot filling field is supplemented according to the supplementary voice information input by the user, target information is generated according to the content of all slot filling fields filled in the task template, and corresponding operation is executed according to the target information and the type of the financial intention. Therefore, the financial information is managed in a voice mode, and a user can manage the financial information conveniently.

Description

Account information processing method and device based on voice recognition and electronic equipment

Technical Field

The invention relates to the technical field of information, in particular to an accounting information processing method and device based on voice recognition and electronic equipment.

Background

With the advent of the intelligent era, electronic devices such as mobile terminals are more and more powerful in configuration and more complete in functions. Meanwhile, due to the improvement of the consumption level and the increase of consumption types of the user, the user can record the consumption amount of the user through the accounting application program in the electronic equipment so as to clearly know the consumption condition of the user.

In the related art, in the process of using the accounting application by the user, the user is usually required to manually record the bill or manually set the filter bar to query the bill, and since there are many manual operations in the process of using the accounting application, the financial management experience of the user is not ideal.

Disclosure of Invention

The present invention is directed to solving, at least to some extent, one of the technical problems in the related art.

Therefore, a first objective of the present invention is to provide an accounting information processing method based on voice recognition, which obtains all necessary slot field contents in a task template through multiple rounds of voice interaction with a user, and generates target information according to all slot field contents filled in the task template, so as to manage accounting information in a voice manner, thereby facilitating the user to manage accounting information.

A second object of the present invention is to provide an accounting information processing apparatus based on voice recognition.

A third object of the invention is to propose an electronic device.

A fourth object of the invention is to propose a non-transitory computer-readable storage medium.

A fifth object of the invention is to propose a computer program product.

In order to achieve the above object, a first embodiment of the present invention provides an accounting information processing method based on voice recognition, including: acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, determining an accounting intention type according to a recognition result, and filling slot position field contents matched with a task template corresponding to the accounting intention type according to key information in the text information; detecting whether all necessary filling slot position fields in the task template are completely filled, and if the necessary filling slot position fields with the vacancy are obtained, prompting the user to supplement the necessary filling slot position fields with the vacancy through voice; acquiring supplementary voice information input by the user, performing voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary text information, performing natural language recognition, and further supplementing the content of the matched vacant necessary filling slot position field; and generating target information according to all the slot field contents filled in the task template, and executing corresponding operation according to the target information and the accounting intention type.

The accounting information processing method based on voice recognition of the embodiment of the invention prompts the user to supplement the vacant necessary filling slot position field by voice when determining that the task template corresponding to the accounting intention type has the vacant necessary filling slot position field according to the character information corresponding to the voice information input by the user, and supplements the content of the vacant necessary slot filling field according to the supplementary voice information input by the user, generating target information according to all slot field contents filled in the task template, executing corresponding operation according to the target information and the accounting intention type, therefore, all necessary slot field contents in the task template are acquired through multi-round voice interaction with the user, and target information is generated according to the contents of all slot fields filled in the task template, so that the financial information is managed in a voice mode, and convenience is brought to users for managing the financial information.

To achieve the above object, a second embodiment of the present invention provides an accounting information processing apparatus based on voice recognition, including: the filling module is used for acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, determining an accounting intention type according to a recognition result, and filling slot position field contents matched with a task template corresponding to the accounting intention type according to key information in the text information; the detection processing module is used for detecting whether all necessary slot filling field in the task template are completely filled, and if the necessary slot filling field with a vacancy is obtained, the user is prompted by voice to supplement the necessary slot filling field with the vacancy; the supplementary module is used for acquiring supplementary voice information input by the user, carrying out voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary text information and carrying out natural language recognition, and further supplementing the content of the matched vacant necessary filling slot position field; and the processing module is used for generating target information according to all the slot field contents filled in the task template and executing corresponding operation according to the target information and the type of the accounting intention.

The accounting information processing device based on voice recognition of the embodiment of the invention prompts the user to supplement the vacant necessary filling slot position field by voice when determining that the task template corresponding to the accounting intention type has the vacant necessary filling slot position field according to the character information corresponding to the voice information input by the user, and supplements the content of the vacant necessary slot filling field according to the supplementary voice information input by the user, generating target information according to all slot field contents filled in the task template, executing corresponding operation according to the target information and the accounting intention type, therefore, all necessary slot field contents in the task template are acquired through multi-round voice interaction with the user, and target information is generated according to the contents of all slot fields filled in the task template, so that the financial information is managed in a voice mode, and convenience is brought to users for managing the financial information.

To achieve the above object, a third aspect of the present invention provides an electronic device, including: a processor and a memory; wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, for implementing the accounting information processing method based on voice recognition as described in the above embodiments.

In order to achieve the above object, a fourth aspect embodiment of the present invention proposes a non-transitory computer-readable storage medium, which when executed by a processor, implements the accounting information processing method based on voice recognition as described in the above embodiments.

In order to achieve the above object, a fifth aspect embodiment of the present invention proposes a computer program product, which when executed by an instruction processor in the computer program product, executes the accounting information processing method based on voice recognition as described in the above embodiments.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The foregoing and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:

fig. 1 is a flowchart of an accounting information processing method based on voice recognition according to one embodiment of the present invention;

fig. 2 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention;

fig. 3 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention;

fig. 4 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention;

fig. 5 is a schematic structural diagram of an accounting information processing apparatus based on voice recognition according to another embodiment of the present invention;

fig. 6 is a schematic configuration diagram of an accounting information processing apparatus based on voice recognition according to another embodiment of the present invention;

fig. 7 is a schematic configuration diagram of an accounting information processing apparatus based on voice recognition according to another embodiment of the present invention;

fig. 8 is a schematic structural diagram of an accounting information processing apparatus based on voice recognition according to an embodiment of the present invention;

fig. 9 is a schematic structural diagram of an accounting information processing apparatus based on voice recognition according to another embodiment of the present invention;

fig. 10 is a schematic configuration diagram of an accounting information processing apparatus based on voice recognition according to another embodiment of the present invention;

fig. 11 is a schematic diagram of a hardware structure of an electronic device for executing an accounting information processing method based on voice recognition according to an embodiment of the present invention.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.

The following describes an accounting information processing method, an accounting information processing device and an electronic device based on voice recognition according to embodiments of the present invention with reference to the accompanying drawings.

Fig. 1 is a flowchart of an accounting information processing method based on voice recognition according to one embodiment of the present invention.

As shown in fig. 1, the account information processing method may include:

step 101, acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, determining an accounting intention type according to a recognition result, and filling slot field contents matched with a task template corresponding to the accounting intention type according to key information in the text information.

It should be noted that the accounting information processing method based on voice recognition may be configured in an accounting information processing apparatus based on voice recognition, and the accounting information processing apparatus based on voice recognition may be disposed in an electronic device, and the electronic device exists in various forms, including but not limited to a mobile terminal, a personal computer device, a server, and the like, which is not limited in this embodiment of the present invention.

Specifically, when the user has an accounting requirement (e.g., accounting, checking account or modifying accounting), the user may input voice information related to the requirement by voice. Correspondingly, after the accounting information processing device based on voice recognition receives voice information input by a user, voice recognition is carried out on the voice information so as to convert the voice information into character information, and then natural semantic analysis is carried out on the character information, and when the intention of the voice information input by the user is determined to be an accounting intention type according to an analysis result, the content of the slot position field matched with the task template corresponding to the accounting intention type is filled according to key information in the text information.

The accounting intent types may include, but are not limited to, accounting intent, checking intent, intent to modify accounting information, and the like.

Wherein the task template includes a plurality of slot fields, and each slot field may include: slot name, slot attributes, and slot field candidate content.

Wherein the slot attributes include a mandatory fill attribute and an unnecessary fill attribute.

It should be understood that the slot attributes of each slot field in the task template may be all filling-necessary attributes, or the slot attributes of some slot fields may be filling-necessary attributes. That is, the slot attribute of a partial slot bit field or all slot bit fields in the task template is a padding attribute.

The slot fields in the task template may include, but are not limited to, a billing date field, a billing subject field, a billing category field, and an amount field, among others.

The schematic of the slot fields in the task template corresponding to the accounting intent is shown in table 1.

It should be understood that the slot names of the necessary slot fields in the task template corresponding to the billing intention in table 1 include the billing date, the billing subject and the amount, and the slot names of the unnecessary slot fields are referred to as the billing category.

Wherein, it should be understood that the above table 1 is only an example of the task template corresponding to the billing intention.

The indication of the plurality of slot fields in the task template corresponding to the intention of modifying the accounting information is shown in table 2.

Each slot field in the task template may be set by default, or may be preset by a user according to requirements. For example, as shown in table 1, in the process of billing by the user, in order to clearly understand the income and expense condition of the user, the user may open a setting page corresponding to the task template as required, and set the slot attribute of the bill category field in the setting page as a mandatory filling attribute, so that the user can not confirm whether the bill that the user needs to record is income or place according to the voice information input by the user, and the user can supplement the bill category of the bill by voice through voice interaction with the user, and determine whether the bill is income or expense through the supplementary voice information input by the user. Therefore, the requirement of the user for setting the task template in a personalized manner is met.

For example, assume that the slot field in the task template is designed as shown in table 1, assume that the user inputs 25 yuan for "now ceiling", convert the voice information into text information, determine that the intent of the user to input the voice information is a billing intent according to the text information, fill the content corresponding to the bill date field to 5, 12 and 12 days in 2018 (i.e., the date corresponding to today) according to the key information "today" in the text information "now ceiling 25 yuan", determine that the notebook account is a charge according to the key information "spent" in the text information, fill the content corresponding to only the bill large-class field in the task template to "charge", and fill the content corresponding to the amount field to 25 yuan according to the key information "25 yuan".

102, detecting whether all necessary slot filling field in the task template are completely filled, and executing 105 if all necessary slot filling field are completely filled; if it is known that there is a vacant field for the slot to be filled, step 103 is executed.

And 103, prompting the user to supplement the vacant necessary slot filling field by voice.

And step 104, acquiring the supplementary voice information input by the user, carrying out voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary accounting text information, carrying out natural language recognition, and further supplementing the matched vacant necessary filling slot position field content.

For example, after the content of the slot field matched with the preset task template is filled according to the key information in the text information "25 yuan for now ceiling", whether all the necessary filling slot fields in the task template are completely filled is detected, and it can be determined that the necessary filling slot field "bill subject" in the task template is a vacant necessary filling slot field. At this time, the voice information of 'what bought takes 25 yuan' can be output to prompt the user to input the supplementary billing information, assuming that the supplementary voice information input by the user is 'clothes', the supplementary voice information is subjected to voice recognition and converted into supplementary billing text information, the corresponding supplementary billing text information is 'clothes', and the key information 'clothes' in the supplementary billing text information can be determined to determine that the content corresponding to the necessary filling slot field 'billing subject' in the task template is 'clothes' by performing natural language recognition on the supplementary billing text information.

And 105, generating target information according to all the slot field contents filled in the task template, and executing corresponding operation according to the target information and the type of the accounting intention.

It should be understood that, the accounting intent types are different in corresponding to different executed operations, for example, the accounting intent types are accounting intentions, the target information may be output to the user by voice for confirmation, and when it is known that the user confirms the target information for storage. For another example, the type of the accounting intention is an accounting intention, and the billing is retrieved according to the target information to generate an accounting result which is fed back to the user. For another example, the accounting intent type is a modified accounting intent, and the bill query can be performed according to the target information and the accounting intent type, and the content of the corresponding slot field in the searched bill is modified, and the modified bill information is stored.

Fig. 2 is a flowchart of a billing processing method based on voice recognition according to another embodiment of the present invention.

As shown in fig. 2, the accounting processing method may include:

step 201, acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, and if the voice information is known as the bookkeeping intention, filling the slot field content matched with the bookkeeping intention in the bookkeeping template according to the bookkeeping key information in the text information.

Note that, for convenience of description, the billing template corresponding to the billing intention is referred to as a billing template.

Specifically, when the user has a billing demand, the user can input voice information related to billing in a voice mode. Correspondingly, after the voice information input by the user is received by the accounting processing device based on the voice recognition, the voice information is subjected to the voice recognition so as to be converted into the character information, and then, natural semantic analysis is carried out on the character information, and when the intention of the voice information input by the user is determined to be the accounting intention according to the analysis result, the content of the slot position field matched with the content of the preset accounting template is filled according to the accounting key information in the text information.

For example, it is assumed that the slot field in the billing template corresponding to the type of the accounting intention is designed as shown in table 1, it is assumed that the voice information input by the user is "25 yuan in smallpox," the voice information is converted into text information, it is determined that the voice information input by the user is the billing intention according to the text information, it is possible to fill the content corresponding to the billing date field as 2018, 5, 12 months (i.e., the date corresponding to today) according to the billing key information "today" in the text information "25 yuan in the text information," it is possible to determine that the notebook account is a payment "according to the" spending "of the billing key information in the text information, and fill the content corresponding to only the large category field of the bill in the billing template as" payment, "and fill the content corresponding to the amount field as 25 yuan" according to the "25 yuan" of the billing key information.

Step 202, detecting whether all necessary slot filling fields in the accounting template are completely filled, and if all necessary slot filling fields are completely filled, executing step 205; if it is known that there is a vacant field for the slot to fill, step 203 is executed.

Step 203, the user is prompted by voice to replenish the vacant necessary slot filling field.

And step 204, acquiring supplementary accounting voice information input by the user, performing voice recognition on the supplementary accounting voice information to convert the supplementary accounting voice information into supplementary accounting text information, performing natural language recognition, and further supplementing the matched vacant necessary slot position field content.

For example, after filling the content of the slot field matched with the preset billing template according to the billing key information in the text information "25 yuan for now smallpox", detecting whether all necessary slot fields in the billing template are completely filled, and determining that the necessary slot field "billing subject" in the billing template is a vacant necessary slot field. At this time, the voice information of 'what bought takes 25 yuan' can be output to prompt the user to input the supplementary billing information, assuming that the supplementary billing voice information input by the user is 'clothes', the supplementary billing voice information is subjected to voice recognition and converted into supplementary billing text information, the corresponding supplementary billing text information is 'clothes', and the content corresponding to the necessary filling slot position field 'billing subject' in the supplementary billing text information can be determined as 'clothes' by performing natural language recognition on the supplementary billing text information.

And step 205, generating accounting information according to all the slot field contents filled in the accounting template, outputting the accounting information to the user by voice for confirmation, and storing the accounting information when the user confirms the accounting information.

In order to accurately record the consumption condition of the user, after accounting information is generated according to all the slot field contents filled in the accounting template, the accounting information can be broadcasted in a voice mode, so that the user can confirm the accounting information.

For example, after the content corresponding to the necessary slot field "billing subject" in the billing template is supplemented according to the supplementary billing voice information input by the user, the billing information can be generated according to the content of all slot fields filled in the billing template, and the billing information "5 month and 12 days, clothes, expenses, 25 yuan, whether there is a mistake?" can be broadcasted by voice.

As a possible implementation manner, after the billing information is broadcasted to the user by voice, whether the feedback information of the user is received within the preset time can be detected, and if the feedback information of the user is not received within the preset time, the user is informed to confirm the billing information.

The preset time is a preset time value, for example, the preset time is 40 seconds, and after the accounting information is broadcasted to the user in a voice mode, if it is determined that the feedback information of the user is not received within 40 seconds, the user is informed to confirm the accounting information.

In addition, if the information fed back by the user is received within the preset time without error response, the user is informed to confirm the accounting information.

If the user confirms the accounting information, the recorded accounting information is accurate, and at the moment, the accounting information can be stored, so that the user can conveniently and accurately know the consumption condition of the user according to the stored accounting information.

For example, the slot field of the billing template is designed as shown in table 1, when the user inputs the voice information of "car-driving took 15" and voice recognition is performed on the voice information, the text information corresponding to the voice information of "car-driving took 15", the text information is naturally analyzed to determine that the text information is intended for billing, that is, the user inputs the voice information as a bill-keeping intention, the bill is determined as a payment according to the key information "spent" in the text information of "car-driving took 15", correspondingly, the content corresponding to the general field in the billing template is filled as a payment, the bill is determined as traffic according to the key information "car-driving" in the text information, correspondingly, the content corresponding to the bill subject field in the billing template is filled as "traffic", the amount of consumption of the bill is determined as 15 yuan according to the key information "15" in the billing template, the amount of consumption of the bill is determined as "traffic" today ", the content corresponding to the bill-keeping text information of the slot field in the billing template is determined as" traffic ", the post-keeping date is determined as" the post-keeping date ", the post-keeping date is determined as if the voice information of the post-keeping text information of the post-keeping track, the content of the post-keeping track, the track of the track, the track of.

When the vacant necessary slot filling field is determined to be arranged in the accounting template corresponding to the accounting intention according to the character information corresponding to the voice information input by the user, the voice prompts the user to supplement the vacant necessary slot filling field, the vacant necessary slot filling field is supplemented according to the supplementary accounting voice information input by the user, accounting information is generated according to the contents of all slot filling fields filled in the accounting template and is output to the user for confirmation in a voice mode, and the user is informed to confirm the accounting information and store the accounting information. Therefore, all necessary slot field contents in the bookkeeping template are obtained through multi-round voice interaction with the user, and the bookkeeping information is generated according to all slot field contents filled in the bookkeeping template, so that intelligent bookkeeping in a voice mode is realized, the recorded bill is matched with the consumption condition of the user, the integrity and the accuracy of the recorded bill are improved, and the subsequent user can conveniently know the consumption condition of the user according to the recorded bill.

When the intelligent billing is carried out, the recorded bill is matched with the consumption condition of the user, the completeness and the accuracy of the recorded bill are improved, and the follow-up user can conveniently know the consumption condition of the user according to the recorded bill.

In practical application, in the process of voice billing, sometimes the voice information input by the user is wrong, so that the user can modify the billing information in a voice interaction manner conveniently, and the recorded billing information is reported to the user after the complete billing information is acquired. If the user finds that the billing information is wrong or the user expresses that the billing information is wrong, the user can modify the billing information through voice interaction. The procedure of correcting the billing information by voice according to this embodiment is described below with reference to fig. 3.

Fig. 3 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention.

As shown in fig. 3, the account information processing method may include:

step 301, acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, and if the voice information is known as the bookkeeping intention, filling the slot field content matched with the bookkeeping template according to the key information in the text information.

Step 302, detecting whether all necessary slot filling fields in the accounting template are completely filled, and if all necessary slot filling fields are completely filled, executing step 305; if it is known that there is a vacant field for the slot to be filled, step 303 is performed.

Step 303, the user is prompted by voice to replenish the vacant necessary slot filling field.

And step 304, acquiring the supplementary voice information input by the user, performing voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary accounting text information, performing natural language recognition, and further supplementing the matched vacant necessary filling slot position field content.

And 305, generating accounting information according to the contents of all the slot fields filled in the accounting template, and outputting the accounting information to the user for confirmation in a voice mode.

And step 306, acquiring the corrected voice information input by the user, and performing voice recognition on the corrected voice information to convert the corrected voice information into corrected accounting text information.

And 307, performing natural language identification on the corrected accounting text information, further replacing the matched wrong slot field content, and storing the corrected accounting information.

For example, the user says "buy vegetables and flower 25 yuan today", then fill out and report after the information: "day 4, 27, diet, expenditure, 25 yuan. ". If the user finds that the amount is misspoken, the user may enter a "wrong, 24-yuan" correction voice message. And performing voice recognition on the corrected voice information to convert the corrected voice information into corrected accounting text information, wherein the corrected accounting text information is 'wrong, namely 24 yuan', performing natural language recognition on the corrected accounting text information, determining that the intention of the user for inputting the voice information is a corrected intention, determining that the matched wrong slot position field is a money amount field by analyzing the corrected text information, further correcting the content of the matched bill money amount field into 24 yuan, and storing the corrected accounting information.

According to the accounting information processing method based on voice recognition, after accounting information is generated according to all slot field contents filled in an accounting template, voice is output to a user for confirmation, corrected voice information input by the user is obtained, the corrected voice information is subjected to voice recognition and converted into corrected accounting text information, natural language recognition is carried out on the corrected accounting text information, the matched wrong slot field contents are replaced, and the corrected accounting information is stored.

In order to make the accounting information processing method based on voice recognition of the embodiment of the present invention clearer, the following introduces a specific process of the accounting information processing method based on voice recognition of the embodiment with reference to fig. 4, where the embodiment takes the accounting information processing method based on voice recognition applied to a server as an example to describe, and the specific process is as follows:

step 1, a user inputs voice information describing bill contents through terminal equipment and uploads the voice information to a server.

It should be understood that there is an accounting application program in the terminal device, and when the user has a recording demand, the user may open the accounting application program in the terminal device and input the voice information in the accounting application program, and correspondingly, the accounting application program uploads the voice information input by the user to the server through the mobile terminal.

And 2, the server receives the voice information and converts the voice information into text information through voice recognition.

And 3, the server carries out natural language recognition on the text information, and recognizes that the intention of the user for inputting the voice information is an accounting intention.

Step 4, the server fills the slot position field content matched with the preset accounting template according to the accounting key information in the text information, and the slot position filling condition is as follows: billing date-air, billing subject-shopping; bill category-point out; the bill amount is-258 yuan.

And 5, the server detects that the accounting date field of the necessary filling slot position field in the accounting template is missing, and sends voice information 'please explain the accounting time' related to the necessary filling slot position field for supplementing the missing position to the terminal equipment.

And 6, the terminal equipment plays the voice information in a voice mode.

And 7, the terminal equipment receives the supplementary accounting voice information today input by the user and sends the supplementary accounting voice information today to the server.

And 8, the server performs voice recognition on the supplementary accounting voice information to convert the supplementary accounting voice information into supplementary accounting text information and performs natural language recognition, so as to supplement the matched necessary slot filling field content.

And 9, the server generates accounting information according to the contents of all the slot field filled in the accounting template.

And step 10, the server sends the voice message '4 months and 27 days for shopping, indicating 258 yuan, and judging whether the billing information has errors?' to the terminal equipment.

And 11, the terminal equipment broadcasts the voice information corresponding to the accounting information in a voice mode.

In step 12, the terminal device receives the corrected billing voice message "not 27 days, but 26 days" input by the user, and generates the corrected billing voice message to the server.

And step 13, the server determines that the intention of the user is 'correction intention' according to the corrected accounting voice information, and modifies the content of the accounting date field to 4 months and 26 days.

Specifically, the server performs voice recognition on the corrected bookkeeping voice information to convert the corrected bookkeeping voice information into corrected bookkeeping text information, performs natural language recognition on the corrected bookkeeping text information, further replaces the content of the matched wrong slot position field, and stores the corrected bookkeeping information.

Step 14, the server does not feed back the user within a preset time, or the feedback is "no problem".

Step 15, storing the accounting information.

According to the accounting information processing method based on voice recognition, after accounting information is generated according to all slot field contents filled in the task template, voice is output to a user for confirmation, corrected voice information input by the user is obtained, voice recognition is carried out on the corrected voice information to be converted into corrected accounting text information, natural language recognition is carried out on the corrected accounting text information, the matched wrong slot field contents are replaced, and the corrected accounting information is stored.

Fig. 5 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention.

As shown in fig. 5, the billing processing method may include:

and step 501, acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, and if the voice information is found to be an account checking intention, filling the content of the slot position field matched with the account checking intention in the task template corresponding to the account checking intention according to account checking key information in the text information.

Wherein, the task template that the checking intention corresponds includes a plurality of slot position fields, and every slot position field can include: slot name, slot attributes, and slot field candidate content.

For convenience of description, the task template corresponding to the checking intention is referred to as a bill query template.

It should be understood that the slot attributes of each slot field in the bill inquiry template may be all filling-necessary attributes, or the slot attributes of some slot fields may be filling-necessary attributes. That is, the slot attribute of a partial slot bit field or all slot bit fields in the bill query template is a padding attribute.

The slot fields in the billing query template may include, but are not limited to, time range, billing subject, and billing category.

In practical applications, the related electronic billing application can collect and classify bills of the user for a period of time (such as a month and a year), and present the total amount of expenses of the user for a certain period of time and the amount of expenses of each category (such as catering, shopping, and the like) respectively. However, after the user finishes watching the statistics, the user sometimes needs to know what the maximum expenditure/expenditures are respectively during the period of time, or what the maximum expenditure is for shopping class, so as to know what the "big money" is spent. In order to meet the requirement of the user for finding the bill with the maximum amount, the minimum amount, and the like recently, in an embodiment of the present invention, the slot attribute of the slot with the slot name being the time range, the bill subject, and the bill feature may be set as the mandatory fill attribute, that is, the slot name of the mandatory fill slot field includes the time range, the bill subject, and the bill feature.

It should be noted that the slot field candidate content corresponding to the bill feature may include a money amount feature and/or a time feature.

Wherein the monetary features may include, but are not limited to, monetary maximum, monetary minimum, monetary second maximum, monetary third maximum, and the like.

Wherein temporal characteristics may include, but are not limited to, most recent in time, least recent in time, and the like.

The schematic form of the slot field of the bill query template is shown in table 3.

The slot names of the necessary slot fields in the bill query template in table 3 include a bill date field, a bill subject and a bill feature, and the slot names of the unnecessary slot fields include a bill category.

All values of the billing subject field can include but are not limited to catering, transportation, entertainment, clothing, house payment, communication, daily life, red envelope, and the like.

The bill query template comprises a bill query template, a bill characteristic slot and a bill characteristic slot, wherein the bill characteristic slot is arranged in the bill query template, bills with non-original information, such as the bill with the largest amount of money, the bill with the second largest amount of money, the bill with the earliest time and the like, can be searched, the requirements of a user on knowing the bills with the largest amount of money, the smallest amount of money and the like are met, and the personalized check requirement of the user is met.

Each slot field in the bill inquiry template can be set by default or can be preset by a user according to requirements.

For example, assuming that the slot field in the bill query template is designed as shown in table 3, assuming that the voice information input by the user is "what the largest meal expense is", the voice information is converted into the text information, the intention of the user to input the voice information is determined as the checking intention according to the text information, the checking key information "the largest" and the "expense" in the text information "what the largest meal expense is" can determine that the bill with the largest bill amount is searched by the user, and correspondingly, the content corresponding to the bill characteristics can be filled into the bill with the largest amount according to the text information. According to the 'expense' in the text information, the fact that the bill is paid and searched by the user can be determined, and correspondingly, the content corresponding to the large class of the bill in the bill query template can be filled as the bill according to the text information. According to the account checking key information 'catering' in the text information, the bill subject of the bill searched by the user can be determined as catering, and correspondingly, the content corresponding to the slot bit field with the slot bit name as the account subject can be filled as 'catering' according to the text information.

Step 502, detecting whether all necessary slot filling fields in the bill inquiry template are completely filled, and if all necessary slot filling fields are completely filled, executing step 505; if it is known that there is a free must fill slot field, step 503 is performed.

Step 503, the user is prompted by voice to replenish the vacant necessary slot filling field.

And 504, acquiring supplementary account checking voice information input by a user, performing voice recognition on the supplementary account checking voice information to convert the supplementary account checking voice information into supplementary account checking text information, performing natural language recognition, and further supplementing the matched vacant necessary slot filling field contents.

For the above example, after filling the content of the slot position field matched in the bill inquiry template according to the account checking key information in the text information "what is the largest meal expense", detecting whether all the necessary slot position fields in the bill inquiry template are completely filled, and determining that the necessary slot position field "time range" in the bill inquiry template is the vacant necessary slot position field. At this time, voice information of "please explain the time range" may be output to prompt the user to input the supplementary accounting voice information, assuming that the supplementary accounting voice information input by the user is "yesterday", voice recognition is performed on the supplementary accounting voice information to convert into supplementary accounting text information, the corresponding supplementary accounting text information is "yesterday", natural language recognition is performed on the supplementary accounting text information to determine date information (for example, date information is 2018, 4, 15, and) corresponding to the accounting key information "yesterday", and content corresponding to the necessary slot filling field "time range" in the bill inquiry template is filled in to 2018, 4, 15, and according to the determined date.

And 505, generating account checking information according to all the slot position field contents filled in the bill inquiry template, and performing bill retrieval according to the account checking information to generate an account checking result to be fed back to the user.

Wherein, it should be understood that, after obtaining the checking result, the checking result can be fed back to the user in various ways, and the checking result can be displayed through voice broadcast and/or a front-end interface.

In the practical application process, the user can set a mode for displaying the checking result according to the requirement, for example, the user can set a mode for displaying the checking result through a front-end interface, and can also set a mode for displaying the query result through voice broadcast. In addition, the user can also set a mode of displaying the checking result through voice broadcasting and a front-end interface, and the user can set a mode of displaying the checking result according to personalized requirements, which is not limited in the embodiment.

As an example, the form of the audit result of voice broadcast and front-end interface display may be: billing time + subject + income/expense + amount.

For example, after supplementing the content corresponding to the slot field "time range" in the bill query template according to the supplementary audit voice information input by the user, it may be determined that all the slot fields that must be filled in the supplementary bill query template are complete, at this time, the audit information may be generated according to the content of all the slot fields filled in the bill query template, and the generated audit information is: 15 of 2018, 4 months, catering, expenditure and the largest amount, and according to the checking information, the bill information is retrieved to acquire the checking result corresponding to the checking information, and if the user selects the checking result through voice broadcast, after the checking result is acquired, the voice broadcast is performed: the meal cost is 160 yuan after 4 months and 15 days.

According to the accounting information processing method based on voice recognition, when the vacant necessary filling slot position fields are determined to be arranged in the bill inquiry template corresponding to the inquiry intention according to the voice information input by the user, the user is prompted by voice to supplement the vacant necessary filling slot position fields, the supplementary inquiry voice information input by the user is obtained, the supplementary inquiry voice information is subjected to voice recognition and converted into supplementary inquiry text information, natural language recognition is carried out, the matched vacant necessary filling slot position field contents are supplemented, the inquiry information is generated according to all the slot position field contents filled in the bill inquiry template, and the bill retrieval is carried out according to the inquiry information to generate the inquiry result which is fed back to the user. From this for the user can realize checking account through voice mode, has avoided the manual trouble of setting for the screening condition one by one of user, promotes the simple operation nature greatly, has made things convenient for user checking account, has reduced user's cost, has improved user's checking account and has experienced. In addition, the accounting information processing method based on voice recognition of the embodiment is used by electronic equipment without a manual interaction mode.

Fig. 6 is a flowchart of an accounting information processing method based on voice recognition according to another embodiment of the present invention.

As shown in fig. 6, the billing processing method may include:

step 601, acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, and if the voice information is found to be an account checking intention, filling the slot position field content matched with the slot position field content in the preset bill query template according to account checking key information in the text information.

Step 602, detecting whether all necessary slot filling fields in the bill inquiry template are completely filled, and if all necessary slot filling fields are completely filled, executing step 605; if it is known that there is a vacant necessary slot position field, step 603 is executed.

Step 603, the user is prompted by voice to replenish the vacant necessary slot filling field.

And step 604, acquiring the supplementary audit trail voice information input by the user, performing voice recognition on the supplementary audit trail voice information to convert the supplementary audit trail voice information into supplementary audit trail text information, performing natural language recognition, and further supplementing the matched vacant necessary slot filling field contents.

And 605, generating account checking information according to all the slot position field contents filled in the bill inquiry template, and performing bill retrieval according to the account checking information to generate an account checking result.

And 606, displaying the account checking result through voice broadcasting and/or a front-end interface.

And step 607, calling accounting information corresponding to the checking result and feeding back the accounting information to the user.

In order to clearly record the consumption condition of the user and facilitate the subsequent determination of the specific use of the corresponding bill, as an exemplary embodiment, after the checking result is obtained, the accounting information corresponding to the checking result may also be called, that is, the accounting information input when the user records the bill is obtained.

It should be noted that the billing information input by the user when recording the bill may be input by the user in a text manner or a voice manner, and the embodiment takes the example of inputting the billing information in a voice manner as an example for description.

As an exemplary embodiment, the form of the audit result of the voice broadcast is: billing time + subject + income/expense + amount of money + voice information corresponding to billing information corresponding to the query result. The form of the account checking result displayed on the front-end interface can be bill time, subject, income/expense, amount and text information of the account checking information corresponding to the inquiry result.

For example, the slot fields corresponding to the bill query template are shown in table 3, and it is assumed that the voice information input by the user during the account check is "what the earliest red packet income in the last month is", the voice information is converted into the text information by voice recognition, the text information is "what the earliest red packet income in the last month is", the intention analysis is performed on the text information, the intention corresponding to the text information can be determined as the account check intention, the time range that can be checked is "the last month" according to the key information "the last month" of the account check in the text information, and correspondingly, the content corresponding to the time range word in the bill query template is filled in as "the last month"; according to the earliest checking key information in the text information, filling the content corresponding to the bill features in the bill inquiry template into the earliest time; according to the key information 'red packet' of checking account in the text information, filling the contents of the bill subjects in the bill inquiry template into 'red packet'; and according to the checking key information income in the text information, filling the contents of the large types of bills in the bill inquiry template into income. Whether all necessary slot filling fields in the bill inquiry template are completely filled or not is detected, so that the fact that all the necessary slot filling fields in the bill inquiry template are completely filled can be determined, at the moment, account checking information is generated according to all the slot filling field contents filled in the bill inquiry template, bill retrieval is carried out according to the account checking information, an account checking result is obtained, and accounting information corresponding to the account checking result is obtained, and the voice broadcasting is carried out: day 3 and 21, the red envelope is received in 200 yuan, and the recorded text is the 200 red envelope given by the small red envelope today. Therefore, the user can inquire the bill in a voice mode, the trouble that the user manually sets screening conditions is avoided, the voice information of the user in recording the bill is broadcasted, the user can know the specific purpose of the corresponding bill according to the voice information, and the user is helped to better carry out balance management and cognition.

When the account information processing method based on voice recognition determines that the bill inquiry template corresponding to the inquiry intention has the vacant necessary filling slot field according to the voice information input by the user, the voice prompts the user to replenish the vacant necessary filling slot field, obtains the supplementary inquiry voice information input by the user, performs voice recognition on the supplementary inquiry voice information to convert the supplementary inquiry voice information into supplementary inquiry text information and performs natural language recognition, further supplements the content of the vacant necessary filling slot field, generates the inquiry information according to the content of all slot field filled in the bill inquiry template, performs bill retrieval according to the inquiry information to generate an inquiry result, and displays the inquiry result through voice broadcast and/or a front-end interface, and calls the information corresponding to the inquiry result to feed back to the user, so that the user can realize the inquiry through voice accounting, the trouble that the user manually sets the screening conditions item by item is avoided, the operation convenience is greatly improved, the user can check accounts conveniently, the accounting information corresponding to the account checking result is fed back to the user, the user can know the specific purpose of the corresponding bill according to the accounting information, and the user can conveniently and accurately manage the balance.

In order to make the accounting information processing method based on voice recognition more clear in the embodiment of the present invention, the following describes the accounting information processing method based on voice recognition in this embodiment with reference to a specific embodiment:

the embodiment is described by taking an example that the accounting information processing method based on voice recognition is applied to a service and a user inputs voice information for inquiring a bill as "what the maximum shopping expense is", and a specific process of the accounting information processing method based on voice recognition of the embodiment is described below with reference to fig. 7.

Step 1, a user inputs voice information of 'what the maximum shopping expense is in the last month' through terminal equipment, and uploads the voice information to a server.

Specifically, in the process of using the accounting application program in the terminal device, when the user has an account checking requirement, the user can open the accounting application program in the terminal device, and input the voice information of "what the maximum shopping expense is" in the accounting application program, and correspondingly, the accounting application program uploads the voice information input by the user to the server through the terminal device.

And 3, the server carries out natural language identification on the text information, and identifies the intention of the user for inputting the voice information as an account checking intention.

Step 4, the server fills the matched slot field content in the preset bill query template according to the account checking key information in the text information, and determines the slot filling condition: time range-null; billing subject-shopping; billing characteristics-amount is greatest.

Specifically, the server may determine, according to the text information "what the maximum shopping expense is" the maximum "and" expense ", that the purpose of checking by the user is to search for a bill with the maximum amount, and correspondingly, fill the content corresponding to the bill characteristic field in the bill query template as" maximum amount ", and fill the bill subject in the bill query template as" shopping "according to the checking key information" shopping "in the text information.

And 5, the server detects that the necessary slot filling field 'time range' in the bill inquiry template is absent, and sends voice information 'please explain the bill time' related to the necessary slot filling field for supplementing the vacancy to the terminal equipment.

It should be understood that fig. 7 is only an example of the voice information related to filling the empty mandatory fill slot field, and the prompt voice information of this embodiment is not limited thereto.

And 6, the terminal equipment broadcasts the voice information 'please explain the bill time' by voice.

And 7, the terminal equipment receives the supplementary account checking voice information 'last month' input by the user and sends the supplementary account checking voice information 'last month' to the server.

And 8, the server performs voice recognition on the supplementary account checking voice information to convert the supplementary account checking voice information into supplementary account checking text information and performs natural language recognition, so as to supplement the content of the time range of the matched vacant necessary slot filling field.

And 9, the server generates account checking information according to all slot field contents filled in the bill inquiry template, the server performs bill retrieval according to the bill time (last month) and the bill subjects (shopping), and sorts the results according to bills to obtain a bill with the largest last month, shopping and money.

And step 10, acquiring the accounting information corresponding to the bill, and sending the bill and the corresponding accounting information to the terminal equipment.

Step 11, the terminal device displays the result in a voice and text mode: x month x day, shopping cost xx yuan, accounting information is "buy xi wear and spend xx yuan for old man today".

The accounting information processing method based on voice recognition of the embodiment of the invention prompts a user to supplement the vacant necessary filling slot position field by voice when the vacant necessary filling slot position field is arranged in the bill inquiry template according to the voice information input by the user, obtains the supplementary checking voice information input by the user, carries out voice recognition on the supplementary checking voice information, converts the supplementary checking voice information into supplementary checking text information and carries out natural language recognition, further supplements the content of the matched vacant necessary filling slot position field, generates checking information according to the content of all slot position fields filled in the bill inquiry template, carries out bill retrieval according to the checking information to generate a checking result, displays the checking result through voice broadcasting and/or a front end interface, and calls the information corresponding to the checking result and feeds back the checking result to the user, so that the user can realize checking through a voice mode, the trouble that the user manually sets the screening conditions item by item is avoided, the operation convenience is greatly improved, the user can check accounts conveniently, the accounting information corresponding to the account checking result is fed back to the user, the user can know the specific purpose of the corresponding bill according to the accounting information, and the user can conveniently and accurately manage the balance.

In order to implement the above embodiments, the present invention further provides an accounting information processing apparatus based on voice recognition.

Fig. 8 is a schematic structural diagram of an accounting information processing apparatus based on voice recognition according to an embodiment of the present invention.

As shown in fig. 8, the accounting information processing apparatus based on voice recognition includes a padding module 110, a detection processing module 120, a supplementary module 130, and a processing module 140, wherein:

the filling module 110 is configured to obtain voice information input by a user, perform voice recognition on the voice information to convert the voice information into text information, perform natural language recognition, determine an accounting intent type according to a recognition result, and fill slot field content matched with a task template corresponding to the accounting intent type according to key information in the text information.

And the detection processing module 120 is configured to detect whether all the necessary slot filling fields in the task template are completely filled, and if it is known that the necessary slot filling fields have a vacancy, prompt the user to supplement the vacant necessary slot filling fields by voice.

And the supplement module 130 is configured to obtain the supplementary voice information input by the user, perform voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary text information, perform natural language recognition, and further supplement the content of the matched vacant necessary slot filling field.

And the processing module 140 is configured to generate target information according to all the slot field contents filled in the task template, and perform corresponding operations according to the target information and the accounting intent type.

In an embodiment of the present invention, as shown in fig. 9, the apparatus may further include:

a setting module 150, configured to set each slot field in the task template, where each slot field includes: the slot device comprises a slot name, a slot attribute and slot field candidate content, wherein the slot attribute comprises: a mandatory property and an unnecessary property.

In an embodiment of the present invention, the processing module 140 is further configured to: and when all necessary slot field are completely filled, generating target information according to all slot field contents filled in the task template, and executing corresponding operation according to the target information and the accounting intention type.

In an embodiment of the present invention, when the accounting intent type is an accounting intent, the processing module 140 is specifically configured to: and outputting the target information voice to the user for confirmation, and storing the target information when the user is known to confirm.

In an implementation of the present invention, the processing module 140 is specifically configured to: and if the feedback information of the user is not received within the preset time, or the information fed back by the user is received without error response, the user is informed that the target information is confirmed to be stored.

In an embodiment of the present invention, when the accounting intent type is an accounting intent type, as shown in fig. 10, the apparatus may further include:

a correction processing module 160, configured to obtain corrected voice information input by the user, perform voice recognition on the corrected voice information, and convert the voice information into corrected accounting text information; and performing natural language identification on the corrected bookkeeping text information, further replacing the matched wrong slot field content, and storing the corrected target information.

In an embodiment of the present invention, when the accounting intent type is an accounting intent, the processing module 140 is specifically configured to: and according to the target information, carrying out bill retrieval to generate an account checking result and feeding back the account checking result to the user.

In one embodiment of the present invention, when the accounting intent type is an accounting intent, the slot name of the mandatory slot field includes: time range, billing subject, and billing characteristics, wherein,

the slot field candidate content corresponding to the bill feature comprises: a monetary characteristic, and/or a temporal characteristic.

In an embodiment of the present invention, the processing module 140 is specifically configured to: displaying the account checking result through voice broadcasting and/or a front-end interface; and calling accounting information corresponding to the account checking result and feeding back the accounting information to the user.

It should be noted that the explanation of the embodiment of the accounting information processing method based on voice recognition is also applicable to the accounting information processing apparatus based on voice recognition in this embodiment, and is not repeated here.

In order to achieve the above embodiments, the present invention also proposes a non-transitory computer-readable storage medium, which when instructions in the storage medium are executed by a processor, enables the voice recognition-based accounting information processing method shown in the above embodiments to be performed.

In order to implement the above embodiments, the present invention also provides a computer program product, which when executed by an instruction processor in the computer program product, executes the accounting information processing method based on voice recognition shown in the above embodiments.

Fig. 11 is a schematic diagram of a hardware structure of an electronic device for executing an accounting information processing method based on voice recognition according to an embodiment of the present invention, and as shown in fig. 11, the electronic device includes:

one or more processors 1110 and a memory 1120, with one processor 1110 being an example in fig. 11.

The electronic device may further include: an input device 1130 and an output device 1140.

The processor 1110, the memory 1120, the input device 1130, and the output device 1140 may be connected by a bus or other means, and the bus connection is exemplified in fig. 11.

The memory 1120, which is a non-transitory computer readable storage medium, may be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the accounting information processing method based on speech recognition in the embodiment of the present application (for example, the populating module 110, the detecting processing module 120, the supplementing module 130, and the processing module 140 shown in fig. 8). The processor 1110 executes various functional applications of the server and data processing by running non-transitory software programs, instructions, and modules stored in the memory 1120, that is, implements the accounting information processing method based on voice recognition in the above method embodiments.

The memory 1120 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created from use of the accounting information processing apparatus based on voice recognition, and the like. Further, the memory 1120 may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 1120 optionally includes memory located remotely from processor 1110, and these remote memories may be connected to a speech recognition based accounting information processing apparatus over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The input device 1130 may receive input numeric or character information and generate key signal inputs related to user settings and function control of the accounting information processing device based on voice recognition. The output device 1140 may include a display device such as a display screen.

One or more modules are stored in memory 1120 and, when executed by one or more processors 1110, perform a method of speech recognition-based accounting information processing in any of the method embodiments described above.

The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of the present application.

The electronic device of embodiments of the present invention exists in a variety of forms, including but not limited to:

(1) mobile communication devices, which are characterized by mobile communication capabilities and are primarily targeted at providing voice and data communications. Such terminals include smart phones (e.g., iphones), multimedia phones, functional phones, and low-end phones, among others.

(2) The ultra-mobile personal computer equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include PDA, MID, and UMPC devices, such as ipads.

(3) Portable entertainment devices such devices may display and play multimedia content. Such devices include audio and video players (e.g., ipods), handheld game players, electronic books, and smart toys and portable car-mounted devices.

(4) The server is similar to a general computer architecture, but has higher requirements on processing capability, stability, reliability, safety, expandability, manageability and the like because of the need of providing highly reliable services.

(5) And other electronic devices with data interaction functions.

In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.

The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.

In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.

Claims

1. An accounting information processing method based on voice recognition is characterized by comprising the following steps:

acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, determining an accounting intention type according to a recognition result, and filling slot position field contents matched with a task template corresponding to the accounting intention type according to key information in the text information;

detecting whether all necessary filling slot position fields in the task template are completely filled, and if the necessary filling slot position fields with the vacancy are obtained, prompting the user to supplement the necessary filling slot position fields with the vacancy through voice;

acquiring supplementary voice information input by the user, performing voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary text information, performing natural language recognition, and further supplementing the content of the matched vacant necessary filling slot position field;

and generating target information according to all the slot field contents filled in the task template, and executing corresponding operation according to the target information and the accounting intention type.

2. The method of claim 1, further comprising:

setting each slot position field in the task template, wherein each slot position field comprises: the slot device comprises a slot name, a slot attribute and slot field candidate content, wherein the slot attribute comprises: a mandatory property and an unnecessary property.

3. The method of claim 1, wherein after said detecting whether all of the mandatory fill slot fields in the task template are filled completely, further comprising:

and if all the necessary slot field is completely filled, generating target information according to all the slot field contents filled in the task template, and executing corresponding operation according to the target information and the accounting intention type.

4. The method of claim 1, wherein when the accounting intent type is accounting intent, the performing respective operations in accordance with the target information and accounting intent type comprises:

and outputting the target information voice to the user for confirmation, and storing the target information when the user is known to confirm.

5. The method of claim 4, wherein learning that the user confirms the target information for storage comprises:

and if the feedback information of the user is not received within the preset time, or the information fed back by the user is received without error response, the user is informed that the target information is confirmed to be stored.

6. The method of claim 4 or 5, after said confirming of said voice output of said target information to said user, further comprising:

acquiring corrected accounting voice information input by the user, and performing voice recognition on the corrected accounting voice information to convert the corrected accounting voice information into corrected accounting text information;

and performing natural language identification on the corrected bookkeeping text information, further replacing the matched wrong slot field content, and storing the corrected target information.

7. The method of claim 1, wherein when the accounting intent type is accounting intent, the performing respective operations according to the target information and accounting intent type comprises:

and according to the target information, carrying out bill retrieval to generate an account checking result and feeding back the account checking result to the user.

8. The method of claim 7, wherein when the accounting intent type is an accounting intent, the slot name of the mandatory slot field comprises:

time range, billing subject, and billing characteristics, wherein,

9. The method of claim 7 or 8, wherein the generating of the audit result by the bill retrieval according to the target information and the feedback of the audit result to the user comprises:

displaying the account checking result through voice broadcasting and/or a front-end interface;

and calling accounting information corresponding to the account checking result and feeding back the accounting information to the user.

10. An accounting information processing apparatus based on voice recognition, comprising:

the filling module is used for acquiring voice information input by a user, performing voice recognition on the voice information to convert the voice information into text information, performing natural language recognition, determining an accounting intention type according to a recognition result, and filling slot position field contents matched with a task template corresponding to the accounting intention type according to key information in the text information;

the detection processing module is used for detecting whether all necessary slot filling field in the task template are completely filled, and if the necessary slot filling field with a vacancy is obtained, the user is prompted by voice to supplement the necessary slot filling field with the vacancy;

the supplementary module is used for acquiring supplementary voice information input by the user, carrying out voice recognition on the supplementary voice information to convert the supplementary voice information into supplementary text information and carrying out natural language recognition, and further supplementing the content of the matched vacant necessary filling slot position field;

and the processing module is used for generating target information according to all the slot field contents filled in the task template and executing corresponding operation according to the target information and the type of the accounting intention.

11. The apparatus of claim 10, further comprising:

a setting module, configured to set each slot field in the task template, where each slot field includes: the slot device comprises a slot name, a slot attribute and slot field candidate content, wherein the slot attribute comprises: a mandatory property and an unnecessary property.

12. The apparatus of claim 10, wherein the processing module is further configured to:

and when all necessary slot field are completely filled, generating target information according to all slot field contents filled in the task template, and executing corresponding operation according to the target information and the accounting intention type.

13. The apparatus of claim 10, wherein when the accounting intent type is accounting intent, the processing module is specifically configured to:

14. The apparatus of claim 13, wherein the processing module is specifically configured to:

15. The apparatus of claim 13 or 14, further comprising:

the correction processing module is used for acquiring correction voice information input by the user, and performing voice recognition on the correction voice information to convert the correction voice information into correction accounting text information; and performing natural language identification on the corrected bookkeeping text information, further replacing the matched wrong slot field content, and storing the corrected target information.

16. The apparatus of claim 10, wherein when the accounting intent type is an accounting intent, the processing module is specifically configured to:

17. The apparatus of claim 16, wherein when the accounting intent type is an accounting intent, the slot name of the mandatory slot field comprises:

time range, billing subject, and billing characteristics, wherein,

18. The apparatus according to claim 16 or 17, wherein the processing module is specifically configured to:

19. An electronic device comprising a processor and a memory;

wherein the processor executes a program corresponding to the executable program code by reading the executable program code stored in the memory, for implementing the voice recognition-based accounting information processing method of any one of claims 1 to 9.

20. A computer program product, characterized in that when an instruction processor in the computer program product executes, it implements the voice recognition-based accounting information processing method of any one of claims 1 to 9.

21. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the voice recognition-based accounting information processing method according to any one of claims 1 to 9.