WO2018133723A1 - 语音购物方法、装置和计算机可读存储介质 - Google Patents

语音购物方法、装置和计算机可读存储介质 Download PDF

Info

Publication number
WO2018133723A1
WO2018133723A1 PCT/CN2018/072138 CN2018072138W WO2018133723A1 WO 2018133723 A1 WO2018133723 A1 WO 2018133723A1 CN 2018072138 W CN2018072138 W CN 2018072138W WO 2018133723 A1 WO2018133723 A1 WO 2018133723A1
Authority
WO
WIPO (PCT)
Prior art keywords
shopping
user
voice
order
keyword
Prior art date
Application number
PCT/CN2018/072138
Other languages
English (en)
French (fr)
Inventor
袁晓春
王林丽
赖南华
于治武
王晓亮
Original Assignee
北京京东尚科信息技术有限公司
北京京东世纪贸易有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京京东尚科信息技术有限公司, 北京京东世纪贸易有限公司 filed Critical 北京京东尚科信息技术有限公司
Priority to US16/479,841 priority Critical patent/US11631123B2/en
Publication of WO2018133723A1 publication Critical patent/WO2018133723A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0641—Shopping interfaces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G06F3/167—Audio in a user interface, e.g. using voice commands for navigating, audio feedback
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/18—Speech classification or search using natural language modelling
    • G10L15/1815—Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L2015/088—Word spotting
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G10L2015/223—Execution procedure of a spoken command

Definitions

  • the present disclosure relates to the field of voice recognition technologies, and in particular, to a voice shopping method, a voice shopping device, and a computer readable storage medium.
  • Intelligent voice interaction is a new generation of interactive technology based on voice input.
  • voice interaction devices have emerged based on this technology.
  • Related technologies include smart speakers, SIRI assistants, and Xunfei language points. Users can use these devices to make basic interactions with the e-commerce platform through voice input without the need for cumbersome manual input, which provides convenience for users.
  • the inventors of the present disclosure have found that the related art has the following problems: most of the related technologies stay on the network connection between the user and the platform by means of some simple interconnections, and there is no personalized voice processing for the user, resulting in the recognition efficiency of the voice input. Lower, user experience is poor.
  • the present disclosure proposes a voice shopping technology solution that can process voice input according to a user's personal shopping habits, thereby improving voice recognition efficiency and improving user experience.
  • a voice shopping method including: receiving a query instruction issued by a user by voice, performing semantic recognition on the query instruction to determine a keyword of the user's query content; Recording a shopping behavior of the user, determining a search range of the keyword; searching the keyword within the search range to obtain product information related to the query content, and performing voice output; and receiving the user
  • the ordering instruction issued by the voice performs semantic identification on the ordering instruction to determine whether to place an order.
  • the query instruction is semantically identified to determine a shopping scenario in which the user is located; and determining the search scope of the keyword includes: determining, according to the shopping scenario and the user's shopping behavior record The search scope of the keyword.
  • each historical shopping data system of the user is determined as a search range of the keyword
  • the shopping behavior record includes a historical order, a shopping cart, a recent attention, and a search result
  • the historical shopping data system is The shopping behavior records a corresponding data system, the historical shopping data system including a historical order data system, a shopping cart data system, a historical browsing data system, and a search data system.
  • the historical shopping data system corresponding to the shopping behavior record is determined as the keyword
  • the search range determines the entire network as the search range of the keyword.
  • the voice output prompts the shopping website system to be abnormal.
  • the voice output prompts the user to correctly input the voice.
  • the voice outputs the source, the name, the promotion information, and the real-time price of the related item of the query content, and prompts the user to confirm the purchase.
  • the voice output prompts that the related product of the query content does not support purchase.
  • the automatic order is placed and the voice output prompts the order to be successful, and the semantic recognition result in response to the order instruction is not purchased, and the order is not automatically performed and the voice is not automatically
  • the output prompt is not placed, the semantic recognition result in response to the order instruction is that the user ends the conversation or the user times out no response, does not automatically place an order and the voice output prompts the order to fail, ends the dialogue, in response to the
  • the semantic recognition result of the order instruction is that the user wants to change a product purchase or query, and continues to recommend other commodities, and in response to the semantic recognition result of the order instruction, a new round of inquiry is searched for, and a new query content is searched.
  • the semantic recognition result of the order instruction is still for a product purchase or inquiry, and the voice output prompts that there is no more related item temporarily, in response to the user actively withdrawing, The voice output prompts the end of the order.
  • the item information includes a self-operating status, an upper shelf status, a cash on delivery status, a price status, a preferential status, and a delivery address inventory status of the item related to the query content, where the search keyword is the user At least one of the commodity name of the inquiry and the commodity preference item inquired by the user.
  • a voice shopping apparatus including: a semantic recognition unit, configured to receive a query instruction issued by a user by voice, and perform semantic recognition on the query instruction to determine the query content of the user And a keyword for receiving the order issued by the user, and performing semantic recognition on the order instruction to determine whether to place an order; the keyword search unit is configured to determine, according to the user's shopping behavior record, Defining a search range of the keyword, and searching the keyword within the search range to obtain product information related to the query content; and a voice output unit for voice outputting the product information and related prompts information.
  • the semantic recognition unit includes: a query content determining subunit, configured to receive a query instruction issued by the user by using a voice, perform semantic recognition on the query instruction, to determine a keyword of the query content of the user, and Receiving an order instruction issued by the user by voice, determining whether to place an order according to a semantic recognition result of the order instruction; a shopping scene determining subunit, configured to perform semantic recognition on the query instruction to determine the user Shopping scene at the place.
  • the keyword search unit includes: a search range determining subunit, configured to determine a search range of the keyword according to the shopping scenario and the shopping behavior record of the user; and a commodity information determining subunit, configured to: Searching for the keyword within the search range to obtain product information related to the query content.
  • the search scope determining subunit when the shopping scenario is the product information of the user in the shopping behavior record of the user, the historical shopping data system corresponding to the shopping behavior record Determining the search range of the keyword, in other cases, determining the entire network as the search range of the keyword; the shopping behavior record includes a historical order, a shopping cart, a recent attention, and a search result; the historical shopping The data system is a data system corresponding to the shopping behavior record, the historical shopping data system including a historical order data system, a shopping cart data system, a historical browsing data system, and a search data system.
  • the voice output unit outputs, in the case that the query content has matching data in the search range, the source, the name, the promotion information, and the real-time price of the related item of the query content, and prompts the The user confirms the purchase, and if the query content does not have matching data within the search range, the voice output prompts that the related product of the query content does not support purchase.
  • the device further includes: an automatic ordering unit, configured to respond to the semantic recognition result of the order instruction as a purchase, automatically place an order and voice output prompts the order success, and respond to the semantics of the order instruction
  • the result of the recognition is that the purchase result is not purchased, the automatic order is not placed, and the voice output prompts that the order is not placed.
  • the semantic recognition result in response to the order instruction is that the user ends the conversation or the user times out no response, and does not automatically place an order and voice output. Prompting that the order fails, ending the conversation; the commodity recommendation unit is configured to respond to the semantic recognition result of the order instruction, the user wants to change a commodity purchase or query, continue to recommend other commodities, and notify the voice output unit to output The next item of information related to the query content.
  • the merchandise recommendation unit responds to the recommendation to the last item, and the semantic recognition result of the order instruction is still for a product purchase or inquiry, and the voice output prompts that there is no more related product, in response to the The user actively exits, and the voice output unit is notified to prompt the user to end the order.
  • a voice shopping device comprising: a memory; and a processor coupled to the memory, the processor configured to execute based on an instruction stored in the memory device
  • the voice shopping method in any of the embodiments.
  • a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the voice shopping method of any of the embodiments.
  • the content that the user is interested in and the shopping scene in which the user is located are identified from the voice input, and the search range of the query content is determined according to the shopping habit of the user. This makes the search results more in line with the user's real shopping intentions, thereby improving the efficiency of speech recognition and improving the user experience.
  • FIG. 1 illustrates an exemplary flow chart of a voice shopping method, in accordance with some embodiments of the present disclosure
  • FIG. 2 illustrates an exemplary flow chart of a voice shopping method in accordance with further embodiments of the present disclosure
  • FIG. 3 illustrates an exemplary flowchart of a voice shopping method in accordance with further embodiments of the present disclosure
  • FIG. 4 illustrates an exemplary block diagram of a voice shopping device, in accordance with some embodiments of the present disclosure
  • FIG. 5 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure
  • FIG. 6 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure
  • FIG. 7 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure.
  • FIG. 1 illustrates an exemplary flow chart of a voice shopping method, in accordance with some embodiments of the present disclosure.
  • the method includes: step 101, determining a keyword; step 102, determining a search range; step 103, obtaining product information; and step 104, determining whether to place an order.
  • a user's voice input is received, and the voice input is semantically identified to determine a keyword for the user to query the content.
  • the user can voice input the name of the product that he wants to purchase or inquire, such as: "I want to buy mineral water” or "I want to buy Nongfu Spring". Or you can also enter the preferential information of the goods by voice input, such as: "What price has been cut recently" or "What are the discounts of the products I are concerned about?"
  • the keyword may be a product name related to the query content or a preferential item name of the product.
  • the voice output prompts the shopping website system to be abnormal. If the result of semantic recognition is only a wake-up word, such as: "Excuse Jingdong", the voice output prompts the user to correctly input the voice.
  • a search range of keywords is determined based on the user's shopping behavior record.
  • the shopping behavior record may be historical order data, shopping cart data, historical browsing data or search data, etc.
  • the search scope is a historical shopping data system corresponding to the shopping behavior record.
  • step 103 the keyword is searched to obtain the item information related to the query content and the voice output is performed.
  • the voice output may be based on the obtained product information: "razor, has been reduced by 5 yuan, the current price is 20 yuan, confirm purchase or change one?".
  • step 104 an order instruction of the user's voice input is received, and the order instruction is semantically identified to determine whether to place an order. For example, the user can confirm purchase or not purchase by voice input.
  • the present disclosure determines the query content of the user by voice recognition, and determines the search range of the keyword according to the shopping habit of the user. This makes the search results more in line with the user's real shopping intentions, and narrows the search range, thereby improving the efficiency of speech recognition and improving the user experience.
  • FIG. 2 illustrates an exemplary flow chart of a voice shopping method in accordance with further embodiments of the present disclosure.
  • the method includes: Step 201, determining a keyword; Step 202, determining a shopping scenario; Step 203, determining a shopping scenario; Step 204 and Step 205, searching for a keyword; Step 206, obtaining product information; Step 207 , to determine whether to place an order.
  • step 201 the user's voice input is received, and the voice input is semantically identified to determine a keyword for the user to query the content.
  • step 202 the speech input is semantically identified to determine the shopping scenario in which the user is located.
  • step 203 it is determined whether the shopping scene is the product information in the user's inquiry shopping activity record. If so, the keywords are searched for in one or more historical shopping data systems corresponding to one or several shopping behavior records (step 204). If not, the keywords are searched throughout the network (step 205).
  • the user can enter: "What items are there in my shopping cart?" or "What promotions have I recently seen?"
  • the user's voice input it can be determined that the shopping scene is a discount for the user who wants to inquire about the items in the shopping cart or in the browsing record, and further can search the keyword by using the historical browsing data system or the shopping cart data system as the search range.
  • the historical shopping data system can be a historical order data system, a shopping cart data system, a historical browsing data system, or a search data system.
  • the keywords may be searched in the order of the historical order data system, the shopping cart data system, the historical browsing data system, and the search data system.
  • step 206 item information related to the query content is obtained and voice output is performed.
  • the search results may be voiced one by one in the order of the historical order data system, the shopping cart data system, the historical browsing data system, and the search data system.
  • step 207 an order instruction of the user's voice input is received, and the order instruction is semantically identified to determine whether to place an order.
  • the disclosure determines the search range of the query content according to the shopping habits and shopping intentions of the user, realizes the personalized voice shopping service for different users, improves the efficiency of the voice recognition, and improves the user experience.
  • FIG. 3 illustrates an exemplary flow chart of a voice shopping method in accordance with further embodiments of the present disclosure.
  • the method includes:
  • step 301 a voice input of the user is received, and the voice input is semantically identified to determine a keyword of the user querying the content;
  • step 302 the voice input is semantically identified to determine a shopping scene in which the user is located;
  • step 303 the search range of the keyword is determined according to the shopping scenario and the user's shopping behavior record
  • step 304 searching for a keyword, determining whether the query content has matching data in the search range;
  • the voice output query related product information of the content and prompt the user to confirm the purchase (step 305);
  • the voice output prompts that the relevant item of the query content does not support the purchase (step 306).
  • the item information includes the self-operating status, the shelf status, the cash on delivery status, the price status, the preferential status, and the receipt address inventory status of the item related to the query content.
  • the shipping address set in the user's client may be queried in turn, and the user identifies the default address and the inventory status of the initial shipping address set in the e-commerce platform. If the search result involves multiple historical shopping data systems, the voice output may be sequentially performed in the order of the historical order data system, the shopping cart data system, the historical browsing data system, and the search data system.
  • the method also includes:
  • step 307 it is determined whether the semantic recognition result of the user's voice input is a determined purchase
  • step 308 If yes, the response is an automatic order, and the voice output prompts that the order is successful (step 308);
  • the response is not to automatically place an order and end the conversation, and the voice output prompts the order to fail. If the user's voice input is that the user wants to change a product purchase or inquiry, the response is to continue to recommend other products. If the user's voice input is to re-open a round of queries, the new query content is searched.
  • the voice output prompts that there are no more related items. If the user voluntarily quits, the voice output prompts the end of the order.
  • the present disclosure performs personalized search for different users, which improves the efficiency of voice recognition.
  • the order or recommendation operation is automatically performed, which satisfies the demand for the simplest shopping operation in the whole process of shopping, and improves the user experience.
  • FIG. 4 illustrates an exemplary block diagram of a voice shopping device, in accordance with some embodiments of the present disclosure.
  • the apparatus includes a semantic recognition unit 41, a keyword search unit 42, and a voice output unit 43.
  • the semantic recognition unit 41 receives the user's voice input, and performs semantic recognition on the voice input to determine the keyword of the user's query content and the user's order instruction. For example, the semantic recognition unit 41 can translate the received voice input into a corresponding text file. Then, the text file is processed to obtain a semantic analysis result, that is, a keyword, where the keyword may be a product name or a preferential item name that the user queries.
  • the keyword search unit 42 determines the search range of the keyword according to the shopping behavior record of the user, and searches for the keyword within the search range to obtain the product information related to the query content.
  • the voice output unit 43 outputs the product information and the related prompt information by voice. For example, the keyword search unit 42 returns the self-operated state, the shelf-on state, the cash on delivery state, the price status, the preferential status, and the delivery address inventory status of the related item to the voice output unit 43 as the query result.
  • the speech output unit 43 processes the query results into a standardized text file and finally converts it into a speech output.
  • the keyword search unit of the present disclosure determines the search range of the keyword according to the shopping habit of the user, so that the search result is more in line with the real shopping intention of the user. Moreover, this narrows the search range, thereby improving the efficiency of speech recognition and improving the user experience.
  • FIG. 5 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure.
  • the apparatus includes a semantic recognition unit 51, a keyword search unit 52, and a voice output unit 43.
  • the semantic recognition unit 51 includes a query content determination sub-unit 511 and a shopping scene determination sub-unit 512.
  • the keyword search unit 52 includes a search range determination sub-unit 521 and a commodity information determination sub-unit 522.
  • the query content determining sub-unit 511 receives the user's voice input, and performs semantic recognition on the voice input to determine the keyword of the user's query content and the user's order instruction.
  • the shopping scene determination sub-unit 512 performs semantic recognition of the voice input to determine the shopping scene in which the user is located. For example, if the user voice inputs "What is the discount for the mobile phone in my car?", the query content determining sub-unit 511 recognizes the keyword as the product name: "mobile phone" and the preferential item name: "offer”.
  • the shopping scene identified by the shopping scene determination sub-unit 512 is: the user specifies to query the keyword in the shopping cart data system.
  • the search range determining sub-unit 521 determines the search range of the keyword based on the shopping scene and the shopping behavior record of the user.
  • the item information determination sub-unit 522 searches for a keyword within the search range to obtain item information related to the inquiry content. For example, according to the above example, the search range determining sub-unit 521 can determine that the search range is the shopping cart data system.
  • the product information determining sub-unit 522 searches for two keywords of "mobile phone" and "offer" in the shopping cart data system, and transmits the matched related information to the voice output unit 43 for voice output.
  • the voice output unit 43 prompts the user to confirm the purchase by voice, and the user inputs the order instruction by voice, and the semantic recognition unit 51 performs semantic recognition on the order instruction.
  • FIG. 6 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure, the device further including an automatic ordering unit 64 and a merchandise recommendation unit 65.
  • the recognition result of the order instruction by the semantic recognition unit 51 is purchase
  • the response of the automatic order unit 64 is automatic order
  • the semantic recognition unit 51 outputs the prompt order success.
  • the number of purchases can be set to 1 by default when placing an order automatically. If the user confirms that the purchase, the product information determination sub-unit 522 finds that there is an abnormality such as the goods have been removed, the delivery address is out of stock, or the commodity price changes, the voice output unit 43 prompts the user for the reason for the failure of the order and the failure to place the order.
  • the response of the automatic order unit 64 is that the order is not automatically placed, and the voice output unit 51 is notified to prompt the user that the order has not been placed.
  • the item recommendation unit 65 responds to continue to recommend other items, and notifies the voice output unit 43 to output the next piece of item information related to the inquiry content. If the item recommendation unit 65 recommends the last item, and the recognition result of the semantic recognition unit 51 is still intended to be replaced by an item purchase or inquiry, the voice output unit 43 prompts that there are no more related items temporarily. If the user voluntarily quits, the voice output unit 43 prompts the end of the order.
  • the present disclosure performs personalized search for different users according to the user historical shopping data and the shopping scene in which the user is located, thereby improving the efficiency of the voice recognition; and automatically placing an order or recommending according to the order instruction input by the user voice.
  • the operation satisfies the needs of the simplest shopping operation in the whole process of shopping, and improves the user experience.
  • FIG. 7 illustrates an exemplary block diagram of a voice shopping device in accordance with further embodiments of the present disclosure.
  • the apparatus 70 of this embodiment includes a memory 701 and a processor 702 coupled to the memory 701, the processor 702 being configured to perform any of the implementations of the present disclosure based on instructions stored in the memory 701.
  • the voice shopping method in the example is not limited to.
  • Memory 701 can include, for example, system memory, fixed non-volatile storage media, and the like.
  • the system memory stores, for example, an operating system, an application, a boot loader, a database, and other programs.
  • a computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the voice shopping method of any of the above embodiments.
  • the computer readable storage medium is a non-transitory computer readable storage medium.
  • the methods and systems of the present disclosure may be implemented in a number of ways.
  • the methods and systems of the present disclosure may be implemented in software, hardware, firmware, or any combination of software, hardware, or firmware.
  • the above-described sequence of steps for the method is for illustrative purposes only, and the steps of the method of the present disclosure are not limited to the order specifically described above unless otherwise specifically stated.
  • the present disclosure may also be embodied as programs recorded in a recording medium, the programs including machine readable instructions for implementing a method in accordance with the present disclosure.
  • the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Acoustics & Sound (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本公开涉及一种语音购物方法、装置和计算机可读存储介质，涉及语音识别技术领域。该方法包括：接收用户通过语音发出的查询指令，对查询指令进行语义识别以确定用户的查询内容的关键词；根据用户的购物行为记录，确定关键词的搜索范围；在搜索范围内对关键词进行搜索，以获得与查询内容相关的商品信息，并进行语音输出；和接收用户通过语音发出的下单指令，对下单指令进行语义识别以确定是否下单。该方法和装置能够根据用户的购物习惯对语音输入进行处理，从而提高语音识别效率，改善用户体验。

Description

语音购物方法、装置和计算机可读存储介质
相关申请的交叉引用
本申请是以CN申请号为201710050620.X，申请日为2017年1月23日的申请为基础，并主张其优先权，该CN申请的公开内容在此作为整体引入本申请中。
技术领域
本公开涉及语音识别技术领域，特别涉及一种语音购物方法、语音购物装置和计算机可读存储介质。
背景技术
近年来，互联网与人们日常生活的结合越来越紧密，尤其是在电子商务领域，人们已经习惯了通过网络购买各种商品和服务。例如，在中国市场，很多人通过京东电商平台获得日常所需的电子消费品、日用百货用和电信增值服务等商品。因此，如何为用户提供更加快速、方便的交互服务成为目前电子商务领域的一个重要技术问题。
智能语音交互是基于语音输入的新一代交互技术，目前基于这种技术出现了很多语音交互设备。相关技术包括智能音箱、SIRI助手和讯飞语点等。用户利用这些设备可以仅通过语音输入即可与电子商务平台进行基本的交互，而无需进行繁琐的手动输入，为用户提供了方便。
发明内容
本公开的发明人发现上述相关技术中存在如下问题：相关技术大都停留在依靠一些简单的互联拼凑将用户与平台进行网络连接，并没有针对用户进行个性化的语音处理，导致语音输入的识别效率较低，用户体验较差。针对所述问题中的至少一个问题，本公开提出了一种语音购物技术方案，能够根据用户的个人购物习惯对语音输入进行处理，从而提高语音识别效率，改善用户体验。
根据本公开的一些实施例，提供了一种语音购物方法，包括：接收用户通过语音发出的查询指令，对所述查询指令进行语义识别以确定所述用户的查询内容的关键词；根据所述用户的购物行为记录，确定所述关键词的搜索范围；在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息，并进行语音输出； 和接收用户通过语音发出的下单指令，对所述下单指令进行语义识别以确定是否下单。
可选地，对所述查询指令进行语义识别以确定所述用户所处的购物场景；所述确定所述关键词的搜索范围包括：根据所述购物场景和所述用户的购物行为记录确定所述关键词的搜索范围。
可选地，将所述用户的各历史购物数据系统确定为所述关键词的搜索范围，所述购物行为记录包括历史订单、购物车、最近关注和搜索结果，所述历史购物数据系统为与所述购物行为记录对应的数据系统，所述历史购物数据系统包括历史订单数据系统、购物车数据系统、历史浏览数据系统和搜索数据系统。
可选地，在所述购物场景为所述用户查询所述用户的购物行为记录中的商品信息的情况下，将所述购物行为记录对应的所述历史购物数据系统确定为所述关键词的搜索范围，在其它情况下，将全网确定为所述关键词的搜索范围。
可选地，在购物网站系统异常的情况下，语音输出提示所述购物网站系统异常，在所述查询指令的语义识别结果为唤醒词的情况下，语音输出提示所述用户正确输入的语音。
可选地，在所述查询内容在所述搜索范围内存在匹配数据的情况下，语音输出所述查询内容的相关商品的来源、名称、促销信息和实时价，并提示所述用户确认购买，在所述查询内容在所述搜索范围内不存在匹配数据的情况下，语音输出提示所述查询内容的相关商品暂不支持购买。
可选地，响应于所述下单指令的语义识别结果为购买，自动下单并语音输出提示下单成功，响应于所述下单指令的语义识别结果为不购买，不自动下单并语音输出提示未下单，响应于所述下单指令的语义识别结果为所述用户结束对话或所述用户超时不回应，不自动下单并语音输出提示下单失败，结束对话，响应于所述下单指令的语义识别结果为所述用户想换一个商品购买或查询，继续推荐其他商品，响应于所述下单指令的语义识别结果为重新开启一轮查询，搜索新的查询内容。
可选地，响应于推荐到最后一个商品，所述下单指令的语义识别结果仍为想换一个商品购买或查询，语音输出提示暂时没有更多有关的商品，响应于所述用户主动退出，语音输出提示此次下单结束。
可选地，所述商品信息包括所述查询内容相关的商品的自营状态、上架状态、货到付款状态、价格状态、优惠状态和收货地址库存状态，所述搜索关键词为所述用户 查询的商品名称和所述用户查询的商品优惠项目中的至少一种。
根据本公开的另一些实施例，提供一种语音购物装置，包括：语义识别单元，用于接收用户通过语音发出的查询指令，对所述查询指令进行语义识别，以确定所述用户的查询内容的关键词，和用于接收用户通过语音发出的下单指令，对所述下单指令进行语义识别以确定是否下单；关键词搜索单元，用于根据所述用户的购物行为记录，确定所述关键词的搜索范围，并在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息；语音输出单元，用于语音输出所述商品信息和相关的提示信息。
可选地，所述语义识别单元包括：查询内容确定子单元，用于接收用户通过语音发出的查询指令，对所述查询指令进行语义识别，以确定所述用户的查询内容的关键词，和用于接收用户通过语音发出的下单指令，根据对所述下单指令的语义识别结果确定是否下单；购物场景确定子单元，用于对所述查询指令进行语义识别以确定所述用户所处的购物场景。
可选地，所述关键词搜索单元包括：搜索范围确定子单元，用于根据所述购物场景和所述用户的购物行为记录确定所述关键词的搜索范围；商品信息确定子单元，用于在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息。
可选地，所述搜索范围确定子单元在所述购物场景为所述用户查询所述用户的购物行为记录中的商品信息的情况下，将所述购物行为记录对应的所述历史购物数据系统确定为所述关键词的搜索范围，在其它情况下，将全网确定为所述关键词的搜索范围；所述购物行为记录包括历史订单、购物车、最近关注和搜索结果；所述历史购物数据系统为与所述购物行为记录对应的数据系统，所述历史购物数据系统包括历史订单数据系统、购物车数据系统、历史浏览数据系统和搜索数据系统。
可选地，所述语音输出单元在所述查询内容在所述搜索范围内存在匹配数据的情况下，语音输出所述查询内容的相关商品的来源、名称、促销信息和实时价，并提示所述用户确认购买，在所述查询内容在所述搜索范围内不存在匹配数据的情况下，语音输出提示所述查询内容的相关商品暂不支持购买。
可选地，该装置还包括：自动下单单元，用于响应于所述下单指令的语义识别结果为购买，自动下单并语音输出提示下单成功，响应于所述下单指令的语义识别结果为不购买，不自动下单并语音输出提示未下单，响应于所述下单指令的语义识别结果 为所述用户结束对话或所述用户超时不回应，不自动下单并语音输出提示下单失败，结束对话；商品推荐单元，用于响应于所述下单指令的语义识别结果为所述用户想换一个商品购买或查询，继续推荐其他商品，并通知所述语音输出单元输出与所述查询内容相关的下一条商品信息。
可选地，所述商品推荐单元响应于推荐到最后一个商品，所述下单指令的语义识别结果仍为想换一个商品购买或查询，语音输出提示暂时没有更多有关的商品，响应于所述用户主动退出，通知所述语音输出单元提示所述用户此次下单结束。
根据本公开的又一些实施例，提供一种语音购物装置，包括：存储器；和耦接至所述存储器的处理器，所述处理器被配置为基于存储在所述存储器设备中的指令，执行所述任一个实施例中的语音购物方法。
根据本公开的再一些实施例，提供一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现所述任一个实施例中的语音购物方法。
在上述实施例中，从语音输入中识别出用户感兴趣的内容和所处的购物场景，并根据用户的购物习惯确定查询内容的搜索范围。这样使得搜索结果更符合用户的真实购物意图，从而提高了语音识别效率，改善了用户体验。
通过以下参照附图对本公开的示例性实施例的详细描述，本公开的其它特征及其优点将会变得清楚。
附图说明
此处所说明的附图用来提供对本公开的进一步理解，构成本申请的一部分，本公开的示意性实施例及其说明用于解释本公开，并不构成对本公开的不当限定。在附图中：
图1示出根据本公开的一些实施例的语音购物方法的示例性流程图；
图2示出根据本公开的另一些实施例的语音购物方法的示例性流程图；
图3示出根据本公开的又一些实施例的语音购物方法的示例性流程图；
图4示出根据本公开的一些实施例的语音购物装置的示例性框图；
图5示出根据本公开的另一些实施例的语音购物装置的示例性框图；
图6示出根据本公开的又一些实施例的语音购物装置的示例性框图；
图7示出根据本公开的再一些实施例的语音购物装置的示例性框图。
具体实施方式
现在将参照附图来详细描述本公开的各种示例性实施例。应注意到：除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对布置、数字表达式和数值不限制本公开的范围。
同时，应当明白，为了便于描述，附图中所示出的各个部分的尺寸并不是按照实际的比例关系绘制的。
以下对至少一个示例性实施例的描述实际上仅仅是说明性的，决不作为对本公开及其应用或使用的任何限制。
对于相关领域普通技术人员已知的技术、方法和设备可能不作详细讨论，但在适当情况下，所述技术、方法和设备应当被视为授权说明书的一部分。
在这里示出和讨论的所有示例中，任何具体值应被解释为仅仅是示例性的，而不是作为限制。因此，示例性实施例的其它示例可以具有不同的值。
应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步讨论。
图1示出根据本公开的一些实施例的语音购物方法的示例性流程图。
如图1所示，该方法包括：步骤101，确定关键词；步骤102，确定搜索范围；步骤103，获得商品信息；步骤104，确定是否下单。
在步骤101中，接收用户的语音输入，对语音输入进行语义识别以确定用户查询内容的关键词。用户可以语音输入想要购买或者查询的商品名称，如：“我要买矿泉水”或“我要买农夫山泉”等。或者也可以语音输入查询商品的优惠信息，如：“最近有什么降价”或“我关注的商品有什么优惠”等。关键词可以是查询内容相关的商品名称或者商品的优惠项目名称等。
在一些实施例中，如果购物网站系统异常，则语音输出提示购物网站系统异常。如果语义识别的结果仅为唤醒词，如：“请问京东”，则语音输出提示所述用户正确输入的语音。
在步骤102中，根据用户购物行为记录确定关键词的搜索范围。其中购物行为记录可以是历史订单数据、购物车数据、历史浏览数据或搜索数据等，搜索范围为购物行为记录对应的历史购物数据系统。
在步骤103中，对关键词进行搜索以获得与查询内容相关的商品信息并进行语音输出。
在一些实施例中，可以根据获得的商品信息语音输出：“剃须刀，已降价5元，目前价格是20元，确认购买还是换一个？”。
在步骤104中，接收用户的语音输入的下单指令，对下单指令进行语义识别以确定是否下单。例如，用户可以通过语音输入确认购买或者不购买。
上述实施例中，本公开通过语音识别确定用户的查询内容，并根据用户的购物习惯确定关键词的搜索范围。这就使得搜索结果更符合用户的真实购物意图，且缩小了搜索范围，从而提高了语音识别的效率，改善了用户体验。
图2示出根据本公开的另一些实施例的语音购物方法的示例性流程图。
如图2所示，该方法包括：步骤201，确定关键词；步骤202，确定购物场景；步骤203，判断购物场景；步骤204和步骤205，搜索关键词；步骤206，获得商品信息；步骤207，确定是否下单。
在步骤201中，接收用户的语音输入，对语音输入进行语义识别以确定用户查询内容的关键词。
在步骤202中，对语音输入进行语义识别以确定用户所处的购物场景。
在步骤203中，判断购物场景是否为用户查询购物行为记录中的商品信息。如果是，则在一项或几项购物行为记录对应的一个或几个历史购物数据系统中搜索关键词(步骤204)。如果否，则在全网搜索关键词(步骤205)。
在一些实施例中，用户可以输入：“我购物车里有什么商品降价？”或“我最近看过的商品有什么促销？”。根据用户的语音输入可以确定购物场景为用户想要查询其购物车中或浏览记录中的商品有什么优惠，进而还可以将历史浏览数据系统或购物车数据系统作为搜索范围对关键词进行搜索。历史购物数据系统可以为历史订单数据系统、购物车数据系统、历史浏览数据系统或搜索数据系统等。
在另一些实施例中，可以按照历史订单数据系统、购物车数据系统、历史浏览数据系统、搜索数据系统的顺序对关键词进行搜索。
在步骤206中，获得与查询内容相关的商品信息并进行语音输出。
在一些实施例中，可以按照历史订单数据系统、购物车数据系统、历史浏览数据系统、搜索数据系统的顺序逐个对搜索结果进行语音输出。
在步骤207中，接收用户的语音输入的下单指令，对下单指令进行语义识别以确定是否下单。
上述实施例中，本公开根据用户的购物习惯和购物意图确定查询内容的搜索范 围，实现了针对不同用户的个性化语音购物服务，提高了语音识别的效率，改善了用户体验。
图3示出根据本公开的又一些实施例的语音购物方法的示例性流程图。
如图3所示，该方法包括：
在步骤301中，接收用户的语音输入，对语音输入进行语义识别以确定用户查询内容的关键词；
在步骤302中，对语音输入进行语义识别以确定用户所处的购物场景；
在步骤303中，根据购物场景和用户的购物行为记录确定关键词的搜索范围；
在步骤304中，搜索关键词，判断查询内容在搜索范围内是否存在匹配数据；
如果是，则语音输出查询内容的相关商品信息，并提示用户确认购买(步骤305)；
如果否，则语音输出提示查询内容的相关商品暂不支持购买(步骤306)。
例如，商品信息包括查询内容相关的商品的自营状态、上架状态、货到付款状态、价格状态、优惠状态和收货地址库存状态等。
在一些实施例中，可以依次查询用户的客户端中设定的收货地址，用户标识默认地址和电商平台中设定的初始收货地址的库存情况。如果搜索结果涉及到多个历史购物数据系统，则可以按照历史订单数据系统、购物车数据系统、历史浏览数据系统、搜索数据系统的顺序依次进行语音输出。
该方法还包括：
在步骤307中，判断用户的语音输入的语义识别结果是否为确定购买；
如果是，则响应为自动下单，并语音输出提示下单成功(步骤308)；
如果否，则响应为不自动下单，并语音输出提示未下单(步骤309)。
在一些实施例中，如果用户超时不回应，如超过5秒未收到回应，或用户的语音输入为用户结束对话，则响应为不自动下单并结束对话，并语音输出提示下单失败。如果用户的语音输入为用户想换一个商品购买或查询，则响应为继续推荐其他商品。如果用户的语音输入为重新开启一轮查询，则搜索新的查询内容。
在另一些实施例中，如果推荐到最后一个商品，用户的语音输入仍为想换一个商品购买或查询，则语音输出提示暂时没有更多有关的商品。如果用户主动退出，则语音输出提示此次下单结束。
上述实施例中，本公开针对不同用户进行个性化搜索，提高了语音识别的效率。并根据用户语音输入的下单指令，自动进行下单或推荐操作，满足了购物全流程最简 购物操作的需求，改善了用户体验。
图4示出根据本公开的一些实施例的语音购物装置的示例性框图。
如图4所示，该装置包括：语义识别单元41、关键词搜索单元42和语音输出单元43。
语义识别单元41接收用户的语音输入，对语音输入进行语义识别以确定用户查询内容的关键词和用户的下单指令。例如，语义识别单元41可以讲接收到的语音输入转化成相应的文本文件。然后对文本文件进行处理，从而得到语义分析结果，即关键词，这里的关键词可以是用户查询的商品名称或者优惠项目名称等。
关键词搜索单元42，根据用户的购物行为记录确定关键词的搜索范围，并在搜索范围内搜索关键词以获得与查询内容相关的商品信息。语音输出单元43用语音输出商品信息和相关的提示信息。例如，关键词搜索单元42向语音输出单元43返回相关的商品的自营状态、上架状态、货到付款状态、价格状态、优惠状态和收货地址库存状态等作为查询结果。语音输出单元43将查询结果处理为标准化的文本文件，并最终转化成语音输出。
上述实施例中，本公开的关键词搜索单元根据用户的购物习惯确定关键词的搜索范围，使得搜索结果更符合用户的真实购物意图。而且这样就缩小了搜索范围，从而提高了语音识别的效率，改善了用户体验。
图5示出根据本公开的另一些实施例的语音购物装置的示例性框图。
如图5所示，该装置包括：语义识别单元51、关键词搜索单元52和语音输出单元43。语义识别单元51包括：查询内容确定子单元511和购物场景确定子单元512。关键词搜索单元52包括：搜索范围确定子单元521和商品信息确定子单元522。
查询内容确定子单元511接收用户的语音输入，对语音输入进行语义识别以确定用户查询内容的关键词和用户的下单指令。购物场景确定子单元512对语音输入进行语义识别以确定用户所处的购物场景。例如，用户语音输入“我的购车里的手机有什么优惠？”，则查询内容确定子单元511识别关键词为商品名称：“手机”和优惠项目名称：“优惠”。购物场景确定子单元512识别的购物场景为：用户指定在购物车数据系统中查询关键词。
搜索范围确定子单元521根据购物场景和用户的购物行为记录确定关键词的搜索范围。商品信息确定子单元522在搜索范围内搜索关键词以获得与查询内容相关的商品信息。例如，根据上面的例子，搜索范围确定子单元521可以确定搜索范围为购物 车数据系统。商品信息确定子单元522在购物车数据系统中搜索“手机”和“优惠”两个关键词，将查找到匹配的相关信息发送给语音输出单元43进行语音输出。同时语音输出单元43用语音提示用户确认购买，用户通过语音输入下单指令，语义识别单元51对下单指令进行语义识别。
在一些实施例中，如图6示出根据本公开的又一些实施例的语音购物装置的示例性框图，该装置还包括：自动下单单元64和商品推荐单元65。
如果语义识别单元51对下单指令的识别结果为购买，则自动下单单元64的响应为自动下单，语义识别单元51输出提示下单成功。例如，自动下单时可以将购买数量默认为1。如果用户确认购买后，商品信息确定子单元522发现存在商品已经下架、发货地址无货或者商品价格变动等异常情况时，语音输出单元43向用户提示下单失败原因和下单失败。
如果语义识别单元51对下单指令的识别结果为不购买，则自动下单单元64的响应为不自动下单，并通知语音输出单元51提示用户未下单。如果语义识别单元51对下单指令的识别结果为换一个商品购买或查询，则商品推荐单元65响应为继续推荐其他商品，并通知语音输出单元43输出与查询内容相关的下一条商品信息。如果商品推荐单元65推荐到最后一个商品，语义识别单元51的识别结果仍为想换一个商品购买或查询，则语音输出单元43提示暂时没有更多有关的商品。如果用户主动退出，则语音输出单元43提示此次下单结束。
上述实施例中，本公开根据用户历史购物数据和所处的购物场景，针对不同用户进行个性化搜索，提高了语音识别的效率；并根据用户语音输入的下单指令，自动进行下单或推荐操作，满足了购物全流程最简购物操作的需求，改善了用户体验。
图7示出根据本公开的再一些实施例的语音购物装置的示例性框图。
如图7所示，该实施例的装置70包括：存储器701以及耦接至该存储器701的处理器702，处理器702被配置为基于存储在存储器701中的指令，执行本公开中任意一个实施例中的语音购物方法。
存储器701例如可以包括系统存储器、固定非易失性存储介质等。系统存储器例如存储有操作系统、应用程序、引导装载程序(Boot Loader)、数据库以及其他程序等。
在一些实施例中，提供一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现上述任一个实施例中的语音购物方法。例如，该计算机可读存 储介质为非瞬时性计算机可读存储介质。
至此，已经详细描述了根据本公开的语音购物方法、语音购物装置和计算机可读存储介质。为了避免遮蔽本公开的构思，没有描述本领域所公知的一些细节。本领域技术人员根据上面的描述，完全可以明白如何实施这里公开的技术方案。
可能以许多方式来实现本公开的方法和系统。例如，可通过软件、硬件、固件或者软件、硬件、固件的任何组合来实现本公开的方法和系统。用于所述方法的步骤的上述顺序仅是为了进行说明，本公开的方法的步骤不限于以上具体描述的顺序，除非以其它方式特别说明。此外，在一些实施例中，还可将本公开实施为记录在记录介质中的程序，这些程序包括用于实现根据本公开的方法的机器可读指令。因而，本公开还覆盖存储用于执行根据本公开的方法的程序的记录介质。
虽然已经通过示例对本公开的一些特定实施例进行了详细说明，但是本领域的技术人员应该理解，以上示例仅是为了进行说明，而不是为了限制本公开的范围。本领域的技术人员应该理解，可在不脱离本公开的范围和精神的情况下，对以上实施例进行修改。本公开的范围由所附权利要求来限定。

Claims (17)

  1. 一种语音购物方法，包括：
    接收用户通过语音发出的查询指令，对所述查询指令进行语义识别以确定所述用户的查询内容的关键词；
    根据所述用户的购物行为记录，确定所述关键词的搜索范围；
    在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息，并进行语音输出；和
    接收用户通过语音发出的下单指令，对所述下单指令进行语义识别以确定是否下单。
  2. 根据权利要求1所述的语音购物方法，其中，所述确定所述关键词的搜索范围包括：
    对所述查询指令进行语义识别以确定所述用户所处的购物场景；
    根据所述购物场景和所述用户的购物行为记录确定所述关键词的搜索范围。
  3. 根据权利要求2所述的语音购物方法，其中，所述确定所述关键词的搜索范围包括：
    将所述用户的各历史购物数据系统确定为所述关键词的搜索范围，
    所述历史购物数据系统为与所述购物行为记录对应的数据系统，
    所述购物行为记录包括历史订单、购物车、最近关注和搜索结果；
    所述历史购物数据系统包括历史订单数据系统、购物车数据系统、历史浏览数据系统和搜索数据系统。
  4. 根据权利要求3所述的语音购物方法，其中，所述确定所述关键词的搜索范围包括：
    在所述购物场景为所述用户查询所述用户的购物行为记录中的商品信息的情况下，将所述购物行为记录对应的所述历史购物数据系统确定为所述关键词的搜索范围；
    在其它情况下，将全网确定为所述关键词的搜索范围。
  5. 根据权利要求1所述的语音购物方法，还包括：
    在购物网站系统异常的情况下，语音输出提示所述购物网站系统异常；
    在所述查询指令的语义识别结果为唤醒词的情况下，语音输出提示所述用户正确 输入的语音。
  6. 根据权利要求1所述的语音购物方法，其中，在所述搜索范围内对所述关键词进行搜索以获得与所述查询内容相关的商品信息并进行语音输出包括：
    在所述查询内容在所述搜索范围内存在匹配数据的情况下，语音输出所述查询内容的相关商品的来源、名称、促销信息和实时价，并提示所述用户确认购买；
    在所述查询内容在所述搜索范围内不存在匹配数据的情况下，语音输出提示所述查询内容的相关商品暂不支持购买。
  7. 根据权利要求1所述的语音购物方法，其中，对所述下单指令进行语义识别以确定是否下单包括：
    响应于所述下单指令的语义识别结果为购买，自动下单并语音输出提示下单成功；
    响应于所述下单指令的语义识别结果为不购买，不自动下单并语音输出提示未下单；
    响应于所述下单指令的语义识别结果为所述用户结束对话或所述用户超时不回应，不自动下单并语音输出提示下单失败，结束对话；
    响应于所述下单指令的语义识别结果为所述用户想换一个商品购买或查询，继续推荐其他商品；
    响应于所述下单指令的语义识别结果为重新开启一轮查询，搜索新的查询内容。
  8. 根据权利要求7所述的语音购物方法，还包括：
    响应于推荐到最后一个商品，所述下单指令的语义识别结果仍为想换一个商品购买或查询，语音输出提示暂时没有更多有关的商品；
    响应于所述用户主动退出，语音输出提示此次下单结束。
  9. 根据权利要求1-8任一项所述的语音购物方法，其中，
    所述商品信息包括所述查询内容相关的商品的自营状态、上架状态、货到付款状态、价格状态、优惠状态和收货地址库存状态，
    所述搜索关键词为所述用户查询的商品名称和所述用户查询的商品优惠项目中的至少一种。
  10. 一种语音购物装置，包括：
    语义识别单元，用于接收用户通过语音发出的查询指令，对所述查询指令进行语义识别，以确定所述用户的查询内容的关键词，和用于接收用户通过语音发出的下单 指令，对所述下单指令进行语义识别以确定是否下单；
    关键词搜索单元，用于根据所述用户的购物行为记录，确定所述关键词的搜索范围，并在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息；
    语音输出单元，用于语音输出所述商品信息和相关的提示信息。
  11. 根据权利要求10所述的语音购物装置，其中，
    所述语义识别单元包括：
    查询内容确定子单元，用于接收用户通过语音发出的查询指令，对所述查询指令进行语义识别，以确定所述用户的查询内容的关键词，和用于接收用户通过语音发出的下单指令，根据对所述下单指令的语义识别结果确定是否下单；
    购物场景确定子单元，用于对所述查询指令进行语义识别以确定所述用户所处的购物场景；
    所述关键词搜索单元包括：
    搜索范围确定子单元，用于根据所述购物场景和所述用户的购物行为记录确定所述关键词的搜索范围；
    商品信息确定子单元，用于在所述搜索范围内对所述关键词进行搜索，以获得与所述查询内容相关的商品信息。
  12. 根据权利要求11所述的语音购物装置，其中，
    所述搜索范围确定子单元在所述购物场景为所述用户查询所述用户的购物行为记录中的商品信息的情况下，将所述购物行为记录对应的历史购物数据系统确定为所述关键词的搜索范围，在其它情况下，将全网确定为所述关键词的搜索范围；
    所述购物行为记录包括历史订单、购物车、最近关注和搜索结果；
    所述历史购物数据系统为与所述购物行为记录对应的数据系统，所述历史购物数据系统包括历史订单数据系统、购物车数据系统、历史浏览数据系统和搜索数据系统。
  13. 根据权利要求10所述的语音购物装置，其中，
    所述语音输出单元在所述查询内容在所述搜索范围内存在匹配数据的情况下，语音输出所述查询内容的相关商品的来源、名称、促销信息和实时价，并提示所述用户确认购买，在所述查询内容在所述搜索范围内不存在匹配数据的情况下，语音输出提示所述查询内容的相关商品暂不支持购买。
  14. 根据权利要求10所述的语音购物装置，还包括：
    自动下单单元，用于响应于所述下单指令的语义识别结果为购买，自动下单并语音输出提示下单成功，响应于所述下单指令的语义识别结果为不购买，不自动下单并语音输出提示未下单，响应于所述下单指令的语义识别结果为所述用户结束对话或所述用户超时不回应，不自动下单并语音输出提示下单失败，结束对话；
    商品推荐单元，用于响应于所述下单指令的语义识别结果为所述用户想换一个商品购买或查询，继续推荐其他商品，并通知所述语音输出单元输出与所述查询内容相关的下一条商品信息。
  15. 根据权利要求14所述的语音购物装置，其中，
    所述商品推荐单元响应于推荐到最后一个商品，所述下单指令的语义识别结果仍为想换一个商品购买或查询，语音输出提示暂时没有更多有关的商品，响应于所述用户主动退出，通知所述语音输出单元提示所述用户此次下单结束。
  16. 一种语音购物装置，包括：
    存储器；以及
    耦接至所述存储器的处理器，所述处理器被配置为基于存储在所述存储器中的指令，执行如权利要求1-9中任一项所述的语音购物方法。
  17. 一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如权利要求1-9中任一项所述的语音购物方法。
PCT/CN2018/072138 2017-01-23 2018-01-10 语音购物方法、装置和计算机可读存储介质 WO2018133723A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/479,841 US11631123B2 (en) 2017-01-23 2018-01-10 Voice shopping method, device and computer readable storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710050620.X 2017-01-23
CN201710050620.XA CN108346073B (zh) 2017-01-23 2017-01-23 一种语音购物方法和装置

Publications (1)

Publication Number Publication Date
WO2018133723A1 true WO2018133723A1 (zh) 2018-07-26

Family

ID=62907734

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/072138 WO2018133723A1 (zh) 2017-01-23 2018-01-10 语音购物方法、装置和计算机可读存储介质

Country Status (3)

Country Link
US (1) US11631123B2 (zh)
CN (1) CN108346073B (zh)
WO (1) WO2018133723A1 (zh)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109993634A (zh) * 2019-04-12 2019-07-09 睿驰达新能源汽车科技（北京）有限公司 一种基于语音生成订单的方法及装置
CN110689375A (zh) * 2019-09-27 2020-01-14 广州云从人工智能技术有限公司 一种基于便携式设备的信息交互方法、系统、设备和介质
CN110767222A (zh) * 2019-06-19 2020-02-07 北京嘀嘀无限科技发展有限公司 一种接单方法及装置
CN113256378A (zh) * 2021-05-24 2021-08-13 北京小米移动软件有限公司 一种确定用户购物需求的方法

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107316645B (zh) * 2017-06-01 2021-10-12 北京京东尚科信息技术有限公司 一种语音购物的方法和系统
TWI657433B (zh) * 2017-11-01 2019-04-21 財團法人資訊工業策進會 語音互動裝置及應用其之語音互動方法
WO2019224039A1 (en) * 2018-05-21 2019-11-28 Bic Violex S.A System and method for providing a voice-activated ordering of replacement shaving cartridge
CN109087110A (zh) * 2018-08-24 2018-12-25 深圳市云之音科技有限公司 一种智能销售方法、装置及终端设备
CN110866157B (zh) * 2018-08-27 2022-07-15 北京猎户星空科技有限公司 机器人应答方法、装置及机器人
CN109410483A (zh) * 2018-10-08 2019-03-01 珠海格力电器股份有限公司 一种语音互动处理方法及装置
CN111161706A (zh) * 2018-10-22 2020-05-15 阿里巴巴集团控股有限公司 交互方法、装置、设备和系统
CN109326286A (zh) * 2018-10-23 2019-02-12 出门问问信息科技有限公司 语音信息处理方法、装置及电子设备
CN109191269A (zh) * 2018-10-23 2019-01-11 杭州引灯网络科技有限公司 一种基于语音交互的商品交易系统
CN111105258B (zh) * 2018-10-29 2023-06-02 阿里巴巴集团控股有限公司 商品定价的方法、装置和系统
CN111160817B (zh) * 2018-11-07 2024-03-05 北京京东振世信息技术有限公司 货品验收方法及系统、计算机系统和计算机可读存储介质
CN109754236A (zh) * 2018-11-13 2019-05-14 斑马网络技术有限公司 基于车载场景的购买方法、装置、系统、设备及存储介质
CN109615425A (zh) * 2018-12-04 2019-04-12 北京羽扇智信息科技有限公司 一种产品推荐方法、装置、电子设备及存储介质
CN109636525A (zh) * 2018-12-11 2019-04-16 苏州创旅天下信息技术有限公司 一种语音购票方法及系统
CN109817217A (zh) * 2019-01-17 2019-05-28 深圳壹账通智能科技有限公司 基于语音识别的自助贩卖方法、装置、设备及介质
CN109903134A (zh) * 2019-02-26 2019-06-18 北京多点在线科技有限公司 一种订单自动生成方法及系统、存储介质
CN110084660A (zh) * 2019-03-13 2019-08-02 佛山市云米电器科技有限公司 互联网家电的语音购物系统及其购物方法
WO2020232615A1 (zh) * 2019-05-20 2020-11-26 深圳市欢太科技有限公司 信息推荐方法、装置、电子设备以及存储介质
CN110377812A (zh) * 2019-06-14 2019-10-25 平安科技（深圳）有限公司 自助购物方法、装置、设备及计算机可读存储介质
CN110517665B (zh) * 2019-08-29 2021-09-03 中国银行股份有限公司 获取测试样本的方法及装置
CN110830665A (zh) * 2019-11-12 2020-02-21 德邦物流股份有限公司 一种语音交互的方法、装置及快递业务系统
CN110955764B (zh) * 2019-11-19 2021-04-06 百度在线网络技术（北京）有限公司 场景知识图谱的生成方法、人机对话方法以及相关设备
CN113409777B (zh) * 2020-03-16 2023-05-23 上海博泰悦臻网络技术服务有限公司 一种记录用户关注点的方法、车载主机及车辆
CN111475714A (zh) * 2020-03-17 2020-07-31 北京声智科技有限公司 一种信息推荐的方法、装置、设备和介质
CN113763084A (zh) * 2020-09-21 2021-12-07 北京沃东天骏信息技术有限公司 产品推荐的处理方法、装置、设备及存储介质
CN114897576B (zh) * 2022-05-05 2024-04-19 深圳市极客智能科技有限公司 基于数据分析的商品推送方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104616190A (zh) * 2015-03-05 2015-05-13 广州新节奏智能科技有限公司 基于互联网和移动互联网的多终端3d体感购物方法及系统
CN105931104A (zh) * 2016-04-25 2016-09-07 张浩楠 一种基于互联网超市购物系统

Family Cites Families (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6321208B1 (en) * 1995-04-19 2001-11-20 Brightstreet.Com, Inc. Method and system for electronic distribution of product redemption coupons
WO1996037881A2 (en) * 1995-05-26 1996-11-28 Applied Language Technologies Method and apparatus for dynamic adaptation of a large vocabulary speech recognition system and for use of constraints from a database in a large vocabulary speech recognition system
US5832212A (en) * 1996-04-19 1998-11-03 International Business Machines Corporation Censoring browser method and apparatus for internet viewing
US20020143550A1 (en) * 2001-03-27 2002-10-03 Takashi Nakatsuyama Voice recognition shopping system
WO2002089035A1 (en) * 2001-04-27 2002-11-07 United States Postal Service Item delivery service with return component
US7594246B1 (en) * 2001-08-29 2009-09-22 Vulcan Ventures, Inc. System and method for focused navigation within a user interface
US9367841B2 (en) * 2011-07-18 2016-06-14 Tiger T G Zhou Facilitating mobile device payments using product code scanning
US20050043940A1 (en) * 2003-08-20 2005-02-24 Marvin Elder Preparing a data source for a natural language query
US7387250B2 (en) * 2003-12-04 2008-06-17 Scanbuy, Inc. System and method for on the spot purchasing by scanning barcodes from screens with a mobile device
JP3923513B2 (ja) * 2004-06-08 2007-06-06 松下電器産業株式会社 音声認識装置および音声認識方法
US8191098B2 (en) * 2005-12-22 2012-05-29 Verimatrix, Inc. Multi-source bridge content distribution system and method
US20070256093A1 (en) * 2006-04-28 2007-11-01 Xanga.Com, Inc. Decentralized and fraud-resistant system and method for rating information content
US9292825B2 (en) * 2006-07-05 2016-03-22 International Business Machines Corporation Multi-tier inventory visibility
US8078502B2 (en) * 2007-02-01 2011-12-13 International Business Machines Corporation Populating an E-commerce shopping cart and other E-commerce fields based upon content extracted from natural language input
US7840409B2 (en) * 2007-02-27 2010-11-23 Nuance Communications, Inc. Ordering recognition results produced by an automatic speech recognition engine for a multimodal application
US7689473B2 (en) * 2008-03-19 2010-03-30 Ashdan Llc Method for generating a shopping list using virtual merchandising options
US20110307280A1 (en) * 2010-06-15 2011-12-15 Triptility, LLC Apparatus and method for searching and booking a complete travel itinerary
US8359642B1 (en) * 2010-06-25 2013-01-22 Sprint Communications Company L.P. Restricting mature content
US9298816B2 (en) * 2011-07-22 2016-03-29 Open Text S.A. Methods, systems, and computer-readable media for semantically enriching content and for semantic navigation
AU2012101185B4 (en) * 2011-08-19 2013-05-02 Apple Inc. Creating and viewing digital note cards
US9620111B1 (en) * 2012-05-01 2017-04-11 Amazon Technologies, Inc. Generation and maintenance of language model
US11087381B2 (en) * 2013-02-25 2021-08-10 Dvorah Hirsch Method for simultaneously one-step filling a virtual shopping cart with many items from one or multiple resources; all items of any type or characteristics from potential resources which have been embodied into a customized list which has been simultaneously generated and sourced in one-step then presented visually to user to select items; culminating and resulting acquisition to be simultaneosly placed in a single virtual shpping cart and all are acquired simultaneously from many source
US9805554B2 (en) * 2013-06-25 2017-10-31 Bally Gaming, Inc. Providing secondary wagering-game play via a mobile device
US20150212710A1 (en) * 2013-10-10 2015-07-30 Go Daddy Operating Company, LLC Card interface for managing domain search results
CN104820928A (zh) * 2015-04-30 2015-08-05 重庆宅天下商贸有限公司 一种基于移动互联网实现快速购物的系统及方法
US10438273B2 (en) * 2015-05-29 2019-10-08 Home Depot Product Authority, Llc Methods, apparatuses, and systems for online item lookup operations
US20170243107A1 (en) * 2016-02-19 2017-08-24 Jack Mobile Inc. Interactive search engine
EP3502840B1 (en) * 2016-08-16 2020-11-04 Sony Corporation Information processing device, information processing method, and program
US11010784B2 (en) * 2017-01-31 2021-05-18 Walmart Apollo, Llc Systems and methods for search query refinement
US20180232443A1 (en) * 2017-02-16 2018-08-16 Globality, Inc. Intelligent matching system with ontology-aided relation extraction
US10885091B1 (en) * 2017-12-12 2021-01-05 Amazon Technologies, Inc. System and method for content playback
US10896457B2 (en) * 2017-12-12 2021-01-19 Amazon Technologies, Inc. Synchronized audiovisual responses to user requests

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104616190A (zh) * 2015-03-05 2015-05-13 广州新节奏智能科技有限公司 基于互联网和移动互联网的多终端3d体感购物方法及系统
CN105931104A (zh) * 2016-04-25 2016-09-07 张浩楠 一种基于互联网超市购物系统

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109993634A (zh) * 2019-04-12 2019-07-09 睿驰达新能源汽车科技（北京）有限公司 一种基于语音生成订单的方法及装置
CN110767222A (zh) * 2019-06-19 2020-02-07 北京嘀嘀无限科技发展有限公司 一种接单方法及装置
CN110689375A (zh) * 2019-09-27 2020-01-14 广州云从人工智能技术有限公司 一种基于便携式设备的信息交互方法、系统、设备和介质
CN113256378A (zh) * 2021-05-24 2021-08-13 北京小米移动软件有限公司 一种确定用户购物需求的方法

Also Published As

Publication number Publication date
US20210358020A1 (en) 2021-11-18
CN108346073A (zh) 2018-07-31
CN108346073B (zh) 2021-11-02
US11631123B2 (en) 2023-04-18

Similar Documents

Publication Publication Date Title
WO2018133723A1 (zh) 语音购物方法、装置和计算机可读存储介质
CN107316645B (zh) 一种语音购物的方法和系统
US10002375B1 (en) Hashtag shopping and rating
US11238870B2 (en) Interaction method, electronic device, and server
CN106991107B (zh) 信息提供系统、信息提供方法及存储介质
US8484099B1 (en) Method, medium, and system for behavior-based recommendations of product upgrades
US9626703B2 (en) Voice commerce
US9563705B2 (en) Re-ranking results in a search
US20150339759A1 (en) Detecting product attributes associated with product upgrades based on behaviors of users
TW201804400A (zh) 資料物件推送的方法、裝置及系統
US20150154685A1 (en) Automated detection of new item features by analysis of item attribute data
US11960842B2 (en) Methods and apparatus for natural language understanding in conversational systems using machine learning processes
US11741139B2 (en) Systems and methods for determining a response to a user query
US20150178792A1 (en) Methods and systems to provide price matching offers to potential customers
US20220382794A1 (en) System and method for programmatic generation of attribute descriptors
JP2003044756A (ja) オンラインショッピング方法、ショッピングサイト、音声認識装置、音声認識サポート装置
CN111914188A (zh) 推荐目标用户的选择方法、系统、设备及存储介质
US20210073224A1 (en) Query rewrite for low performing queries based on customer behavior
US11107144B2 (en) Automatically determining in real-time a triggering model for personalized recommendations
US11210341B1 (en) Weighted behavioral signal association graphing for search engines
US20080103936A1 (en) Automated decision assistant
KR101695570B1 (ko) 키워드 추출에 의한 온라인 쇼핑 장바구니의 자동 생성 방법
CN108133011B (zh) 一种资讯推送方法及装置
US10346894B2 (en) Methods and systems to create purchase lists from customer receipts
KR101645554B1 (ko) 소비자 프로파일링을 통한 맞춤형 쇼핑정보 제공 시스템 및 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18741344

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 28.10.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 18741344

Country of ref document: EP

Kind code of ref document: A1