A quick order opening method based on voice

Abstract

The present invention discloses a method for rapid ordering based on voice, which belongs to the field of intelligent ordering of goods and solves the problem of how to rapidly order goods based on voice technology, thereby improving ordering efficiency. After defining a database and an ordering language template, the present invention adopts voice wake-up technology, voice recognition technology and semantic recognition SDK technology to convert the acquired voice into a target character string; data entities are extracted from the target character string according to the ordering language template and matched with the database; finally, the matched information in the database is integrated and displayed on the software interface; the user confirms the information through voice and completes the order. The present invention can accurately obtain the user's ordering demand information, and no longer uses the method of inputting keywords or manual selection to search and order goods, thereby realizing rapid ordering of goods based on voice technology and improving the user's ordering efficiency. At the same time, it improves the user's software interaction experience and has won praise from users.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent commodity invoicing, and specifically relates to a method for rapid voice invoicing. Background Art

[0002] With the development of the internet, the PC internet has become increasingly saturated, while the mobile internet has experienced explosive growth. Mobile internet (MI) is an emerging service that uses mobile wireless communications to access services via smart mobile terminals. The mobile internet has entered the voice era. Various information terminals, including laptops, mobile phones, learning machines, in-car navigation devices, and even smart toys, now have both listening and speaking capabilities, enabling web searches using only speech, without the need for keyboards or handwriting.

[0003] Currently, mobile inventory management software offers a variety of ways to select products. You can display a list of all products on the software interface and then select the corresponding product, or you can enter keywords to search for products. However, these methods are very inefficient and lack interest.

[0004] To this end, the present invention proposes a method for quickly opening an order based on voice. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method for quickly issuing invoices based on voice, which solves the problem of how to quickly issue invoices for goods based on voice technology, thereby improving the efficiency of invoicing.

[0006] To achieve the above-mentioned object, according to an embodiment of the first aspect of the present invention, a method for quickly issuing an order based on voice is proposed, comprising:

[0007] Step 1: Define a database; the database is provided with a commodity structure table, a commodity unit structure table, a commodity-unit association table, and a customer structure table;

[0008] Step 2: Define a billing language template; the billing language template is: customer | customer name entity | purchase | quantity entity | unit entity | product name entity;

[0009] Step 3: Obtain the target string; use voice wake-up technology, voice recognition technology, and semantic recognition SDK technology to convert the acquired voice into the target string;

[0010] Step 4: Extract data entities from the target string according to the order form language template and match them with the database; extract Chinese character strings from the target string, use Chinese character to pinyin technology to convert the extracted Chinese character strings into the corresponding pinyin strings and pinyin initials abbreviation strings of the corresponding entities, and finally use the Chinese character strings of the entities, the pinyin strings of the Chinese characters, and the pinyin initials abbreviation strings of the Chinese characters to perform an exact match in the database;

[0011] Step 5: Integrate the information matched in the database and display it on the software interface;

[0012] Step 6: The user confirms the information through voice to complete the order placement.

[0013] Further, the commodity structure table is used to store commodity information; the commodity information includes: the primary key of the commodity structure table, commodity name, pinyin of the commodity name, pinyin initials abbreviation of the commodity name, and other relevant commodity information.

[0014] Further, the commodity unit structure table is used to store the basic unit information of commodities; the basic unit information of the commodities includes: the primary key of the commodity unit structure table, unit name, pinyin of the unit name, pinyin initials abbreviation of the unit name, and other relevant basic unit information of the commodities.

[0015] Further, the commodity-unit association table is used to store information on the association relationship between commodities and units; the information on the association relationship between commodities and units includes: the primary key of the commodity-unit association table, commodity unit price, conversion relationship between units, and other relevant commodity unit association information.

[0016] Further, the customer structure table is used to store customer information; the customer information includes: the primary key of the customer structure table, customer name, pinyin of the customer name, pinyin initials abbreviation of the customer name, and other relevant customer information.

[0017] Further, the extraction logic of the customer name entity is: extract the string between the Chinese character strings "customer" and "purchase", use Chinese character to pinyin technology to obtain the pinyin and initials abbreviation of the string, and use a database query statement to perform an exact match on the corresponding fields in the customer structure table, so as to obtain the corresponding customer information stored in the customer structure table.

[0018] Further, the extraction logic of the quantity entity is: extract the continuous quantity string after the Chinese character string "purchase" and convert it into Roman numerals.

[0019] Furthermore, the extraction logic of the unit entity is: extract the first character after the quantity string, use the Chinese character to pinyin technology to obtain the pinyin and initials of the character, use the database query statement to accurately match the corresponding field in the commodity unit structure table, if the field representing the unit is not matched, continue to append the subsequent strings in sequence for matching, and accurately obtain the corresponding unit information stored in the commodity unit structure table.

[0020] Furthermore, the extraction logic of the product name entity is: extract the character string after the unit character string, use the Chinese character to pinyin technology to obtain the pinyin and initials of the character string, and use the database query statement to accurately match the corresponding field in the product structure table to accurately obtain the corresponding product information stored in the product structure table.

[0021] Furthermore, the logic for extracting the unit price of the commodity is as follows: according to the matched commodity information and the unit information of the commodity, the commodity price is queried from the commodity-unit association table.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] After defining the database and the order language template, the present invention adopts voice wake-up technology, voice recognition technology and semantic recognition SDK technology to convert the acquired voice into a target character string; extracts data entities from the target character string according to the order language template and matches them with the database; that is, extracts Chinese character character strings from the target character string, adopts Chinese character to pinyin technology to convert the extracted Chinese character character strings into pinyin character strings and pinyin initials strings corresponding to the corresponding entities, and uses the entity's Chinese character character string, Chinese character pinyin character string and Chinese character pinyin initials strings to perform precise matching in the database; finally, the matched information in the database is integrated and displayed on the software interface; the user confirms the information through voice and completes the order; the present invention can accurately obtain the user's order demand information, and no longer uses the method of entering keywords or manual selection to search and order products, realizes the rapid ordering of products based on voice technology, and improves the user's ordering efficiency; at the same time, it improves the user's software interaction experience and has won praise from users. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0025] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] like Figure 1 As shown, the voice-based quick order opening method includes the following steps:

[0027] Step 1: Define the database;

[0028] In this embodiment, a database is defined according to order opening requirements, including multiple data tables, namely: a commodity structure table, a commodity unit structure table, a commodity-unit association table, and a customer structure table;

[0029] The product structure table is used to store product information; the product information includes: the primary key of the product structure table, the product name, the pinyin of the product name, the pinyin abbreviation of the product name, and other related product information; specifically, the product structure adopts a table structure; the product structure table is named CP_Product_Info, the primary key of the product structure table is named ProductId, the product name is named ProductName, the pinyin of the product name is named ProductNamePinYing, and the pinyin abbreviation of the product name is named ProductNamePY; and some other related product information is also named for the product structure table;

[0030] The commodity unit structure table is used to store basic unit information of commodities; the basic unit information of commodities includes: the primary key of the commodity unit structure table, the unit name, the pinyin of the unit name, the pinyin abbreviation of the unit name, and other relevant basic unit information; specifically, the commodity unit structure adopts a table structure; generally speaking, each commodity can have multiple units, such as Snow Beer Draft, the unit of this commodity can be cans, boxes (1 box = 12 cans), pieces (1 piece = 6 cans), etc.; the commodity unit structure table is named CP_UnitSetting, the primary key of the commodity unit structure table is named UnitId, the unit name is named UnitName, the pinyin of the unit name is named UnitNamePinYing, and the pinyin abbreviation of the unit name is named UnitNamePY; and some other relevant basic unit information is also named for the commodity unit structure table;

[0031] The product-unit association table is used to store information about the association between products and units. The information about the association between products and units includes: the primary key of the product-unit association table, the product unit price, the conversion relationship between units, and other relevant product-unit association information. Specifically, the product-unit association table adopts a table structure; the product-unit association table is named CP_Product_Unit, the primary key of the product-unit association table is named Id, the product unit price is named SalePrice, and the conversion relationship between units is named UnitRatio. In addition, other relevant product-unit association information is also named for the product-unit association table.

[0032] The client structure table is used to store client information; the client information includes: the primary key of the client structure table, the client name, the pinyin of the client name, the initials of the pinyin of the client name, and other relevant client information; specifically, the client structure table adopts a table structure; the client structure table is named CP_Client_Info, the primary key of the client structure table is named ClientId; the client name is named ClientName, the pinyin of the client name is named ClientNamePinYing, and the initials of the pinyin of the client name is named ClientNamePY; and some other relevant client information is also named for the client structure table;

[0033] Step 2: Define the billing language template;

[0034] In this application, the billing language template is defined for product billing, specifically:

[0035] Customer | Customer Name Entity | Purchase | Quantity Entity | Unit Entity | Product Name Entity

[0036] For example: Customer Zhang San buys a can of Snow Beer; the user only needs to express it according to this language template;

[0037] Step 3: Get the target string;

[0038] In this embodiment, the existing voice wake-up technology, voice recognition technology and semantic recognition SDK technology are used to convert the acquired voice into the target character string;

[0039] Assume the following application scenario:

[0040] Merchant: Xiaomei Xiaomei, I want to place an order.

[0041] The voice assistant replied: OK, master.

[0042] Merchant: Customer Zhang San purchased a can of Snow Beer.

[0043] Voice assistant reply: Okay, master. The price of a can of Snow Beer is 6 yuan.

[0044] Merchant: Place the order.

[0045] Voice assistant reply: Okay, master. The order has been placed for you.

[0046] Step 4: Extract data entities from the target string according to the order form language template and match them with the database;

[0047] The process includes: obtaining the target string through Step 3, extracting the Chinese character string from the target string, using the Chinese character to pinyin technology to convert the extracted Chinese character string into the corresponding pinyin string and the first letter abbreviation string of the pinyin for the corresponding entity, and finally using the Chinese character string of the entity, the pinyin string of the Chinese character, and the first letter abbreviation string of the pinyin of the Chinese character to perform an exact match in the database;

[0048] According to the order form language template described in Step 3, it includes the following data entities: customer name entity, quantity entity, unit entity, and commodity name entity;

[0049] The following is the extraction logic for each entity:

[0050] The extraction logic of the customer name entity is: extract the string between the Chinese character strings "customer" and "purchase". For example, if the extracted string is "Zhang San", use the Chinese character to pinyin technology to get the pinyin "zhangsan" and the first letter abbreviation "zhs", and use the database query statement to perform an exact match in the customer structure table CP_Client_Info for the ClientName, ClientNamePinYing, and ClientNamePY fields, so as to obtain all the customer information of the customer "Zhang San" stored in the customer structure table;

[0051] The extraction logic of the quantity entity is: extract the consecutive quantity string after the Chinese character string "purchase"; for example, if "one" is obtained, it is converted into the Roman numeral "1".

[0052] The extraction logic of the unit entity is: extract the first character after the quantity string. For example, if "can" is obtained, use the Chinese character to pinyin technology to get the pinyin "ting" and the first letter abbreviation "t", and use the database query statement to perform an exact match in the commodity unit structure table CP_UnitSetting for the UnitName, UnitNamePinYing, and UnitNamePY fields. If the field representing the unit is not matched, continue to append the subsequent strings in sequence for matching, and exactly obtain all the unit information of "can" stored in the commodity unit structure table;

[0053] The extraction logic of the commodity name entity is as follows: extract the string after the unit string. For example, "Snowflake Beer" is obtained. Using the Chinese character to pinyin technology, the pinyin "xuehuapijiu" and the initials abbreviation "xhpj" are obtained, and the database query statement is used to accurately match the fields of ProductName, ProductNamePinYing, and ProductNamePY in the commodity structure table CP_Product_Info to accurately obtain all the commodity information of the commodity "Snowflake Beer" stored in the commodity structure table;

[0054] The extraction logic of the unit price of the commodity is as follows: according to the matched commodity information and the unit information of the commodity, the commodity price SalePrice is queried from the commodity-unit association table;

[0055] Step Five: Integrate the information matched in the database and display it on the software interface;

[0056] Step Six: The user confirms the information through voice and completes the order placement;

[0057] This step can specifically refer to Step Three. In addition, the extraction of the voice semantics of the user is an existing technology and will not be elaborated here.

[0058] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A voice-based quick ordering method, characterized in that: including: Step 1: Define a database; the database is set with a product structure table, a product unit structure table, a product-unit association table, and a customer structure table; Step 2: Define an invoicing language template; the invoicing language template is: Customer | Customer Name Entity | Purchase | Quantity Entity | Unit Entity | Product Name Entity; Step 3: Obtain a target string; use voice wake-up technology, speech recognition technology, and semantic recognition SDK technology to convert the obtained voice into a target string; Step 4: Extract data entities from the target string according to the invoicing language template and match them with the database; extract a Chinese character string from the target string, use the Chinese character to pinyin technology to convert the extracted Chinese character string into the corresponding pinyin string and the first letter abbreviation string of the pinyin of the corresponding entity, and finally use the Chinese character string of the entity, the pinyin string of the Chinese character, and the first letter abbreviation string of the pinyin of the Chinese character to perform an exact match in the database; Step 5: Integrate the information matched in the database and display it on the software interface; Step 6: The user confirms the information by voice to complete the order.

2. The method for rapid billing based on voice according to claim 1, characterized in that: The product structure table is used to store product information; The product information includes: the primary key of the product structure table, the product name, the pinyin of the product name, and the first letter abbreviation of the pinyin of the product name.

3. The method for rapid billing based on voice according to claim 1, characterized in that: The product unit structure table is used to store the basic information of the product unit; The basic information of the product unit includes: the primary key of the product unit structure table, the unit name, the pinyin of the unit name, and the first letter abbreviation of the pinyin of the unit name.

4. The method for quickly issuing invoices based on voice according to claim 1, characterized in that: The product-unit association table is used to store the information of the association relationship between the product and the unit; The information of the association relationship between the product and the unit includes: the primary key of the product-unit association table, the product unit price, and the conversion relationship between the units.

5. The method for rapid billing based on voice according to claim 1, characterized in that: The customer structure table is used to store customer information; the customer information includes: the primary key of the customer structure table, the customer name, the pinyin of the customer name, and the first letter abbreviation of the pinyin of the customer name.

6. The method for quickly issuing invoices based on voice according to claim 1, characterized in that: The extraction logic of the customer name entity is: extract the string between the Chinese character strings "Customer" and "Purchase", use the Chinese character to pinyin technology to obtain the pinyin and the first letter abbreviation of the string, and use a database query statement to perform an exact match on the corresponding fields in the customer structure table, so as to obtain the corresponding customer information stored in the customer structure table.

7. The method for rapid billing based on voice according to claim 1, characterized in that: The extraction logic of the quantity entity is: extract the consecutive quantity string after the Chinese character string "Purchase" and convert it into a Roman numeral.

8. The method for rapid billing based on voice according to claim 1, characterized in that: The extraction logic of the unit entity is: extract the first character after the quantity string, use the Chinese character to pinyin technology to obtain the pinyin and the first letter abbreviation of the character, use a database query statement to perform an exact match on the corresponding fields in the product unit structure table, if no field representing the unit is matched, then continue to append the subsequent strings in turn for matching, and accurately obtain the corresponding unit information stored in the product unit structure table.

9. The method for quickly issuing invoices based on voice according to claim 1, characterized in that: The extraction logic of the product name entity is as follows: extract the character string following the unit character string, use the Chinese character to pinyin technology to obtain the pinyin and initials of the character string, and use the database query statement to accurately match the corresponding fields in the product structure table to accurately obtain the corresponding product information stored in the product structure table.

10. The method for quickly issuing invoices based on voice according to claim 1, characterized in that: The logic for extracting the unit price of the commodity is as follows: according to the matched commodity information and the unit information of the commodity, the commodity price is queried from the commodity-unit association table.

Images