WO2021135317A1

WO2021135317A1 - Method and system for making call on the basis of order

Info

Publication number: WO2021135317A1
Application number: PCT/CN2020/111942
Authority: WO
Inventors: 邱会会; 倪合强; 李斌; 白云; 李宝军
Original assignee: 苏宁云计算有限公司
Priority date: 2019-12-30
Filing date: 2020-08-28
Publication date: 2021-07-08
Also published as: CN111178807A; CA3166344A1

Abstract

A method and system for making a call on the basis of an order. The method comprises the following steps: scanning an order number to acquire and store an order address and a phone number corresponding to an order, and automatically dialing the phone number on the basis of a delivery policy of the order (S1); automatically connecting the call and recognizing, by means of a machine, speech information of a consignee corresponding to the phone number, and starting a machine conversation (S2); extracting a keyword of the speech information, extracting a piece of reply speech from a reply word stock corresponding to the keyword, and sending the reply speech to the consignee (S3); and when the reply speech cannot be extracted from the reply word stock, and the consignee cannot be replied to, converting the machine conversation into a manual conversation (S4). By means of firstly performing phone communication between a machine and a consignee, during a delivery process of a deliveryman, the delivery efficiency can be significantly improved, and the delivery time of the deliveryman can be reduced; and directly making a call by means of an order improves the success rate of one delivery instance, the risk of personal information leakage of the consignee is reduced, and the professional service level of the deliveryman is improved.

Description

Method and system for dialing calls based on orders

Technical field

The present invention relates to the field of intelligent voice, in particular to a method and system for making calls based on orders.

Background technique

In the current logistics, express delivery, and takeaway industries, after a user purchases or places an order on the Internet, the courier or takeaway brother needs to contact the user to understand the delivery time, location and other issues, and the general delivery location There are supermarkets, real estate, express cabinets, rookie stations, etc. In the final delivery, contact the user generally through the following methods: first, send SMS; second, push the delivery information through the APP on the recipient’s mobile phone; third, directly call to receive the goods The person’s phone number, the courier brother enters the phone, makes the call, the user also answers the phone, and can complete the express delivery in time.

However, the above three methods will have some problems that need to be solved: when sending delivery information through the first type of SMS, the user now has too many SMS messages to pay attention to the delivery information in time, and the user has a bunch of SMS messages, such as advertising push SMS, marketing Users will not respond easily to text messages, fraudulent text messages, etc., and also reduce the frequency of checking text messages. When the second type of mobile phone APP pushes delivery information, some users activate the message reminder function, which can receive the message in time, but the user cannot see it in time; some users do not activate the message reminder, and can only see the message after the user logs in to the system , Not in time. When calling directly through the third type of courier or food delivery person, the mobile phone number is 11 digits. A courier has hundreds of orders a day. Direct manual calls will result in too slow efficiency. Some couriers still ride bicycles. When you make a call, traffic safety is more dangerous. How to quickly and efficiently deliver the last mile of express delivery requires further technological innovation.

Summary of the invention

In order to solve the problems of the prior art, the embodiments of the present invention provide a method and system for making a call based on an order, which can automatically dial the order information and realize the combination of machine and manual communication, and improve the efficiency of delivery.

In order to solve the above technical problems, the technical solution adopted by the present invention is:

In the first aspect, an embodiment of the present invention provides a method for making a call based on an order, which includes the following steps:

Scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

Automatically connect and the machine recognizes the voice message of the consignee corresponding to the telephone number, and starts the machine dialogue;

Extract the keywords of the voice information, extract the reply voice from the reply vocabulary corresponding to the keywords and send it to the consignee;

When the reply voice cannot be extracted from the reply lexicon and reply to the consignee, the machine dialogue is converted into a manual dialogue.

Further, the distribution strategy includes:

Order delivery according to the distance from the receiving place to the delivery station;

Or deliver them sequentially in the order of scanning the order number;

Or the historical call-through rate of the consignee’s answering calls is smoothly delivered sequentially;

Or according to the consignee’s user tag level for delivery.

Further, the machine recognition of the voice information is realized by a voice recognition model, and the establishment of the voice recognition model includes: collecting voice information between the consignee and the delivery person, training the voice information and storing it.

Further, the machine recognition further includes a semantic understanding of the voice information, and the semantic understanding is realized through a voice rule configuration and a voice similarity algorithm.

Further, the reply voice included in the reply vocabulary is a text conversion through a voice text, and the text of the voice text is determined according to the transcript or the end of the conversation during the communication between the consignee and the delivery person.

On the other hand, an embodiment of the present invention also provides a system for making a call based on an order, including:

The order scanning module is used to scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

The machine recognition module is used to automatically connect and machine recognize the voice information of the consignee corresponding to the telephone number, and start a machine dialogue;

The machine reply module is used to extract the keywords of the voice message, extract the reply voice from the reply vocabulary corresponding to the keywords and send it to the consignee;

The manual conversion module is used to convert the machine dialogue into a manual dialogue when the machine reply module cannot extract the reply voice from the reply lexicon and reply to the consignee.

Further, the order scanning module includes a delivery strategy selection unit, and the selection of the order delivery by the delivery strategy selection unit includes: order delivery according to the distance from the receiving place to the delivery station; order according to the order number scanning order Delivery; the historical call-through rate of the consignee's answering calls is smoothly delivered in sequence; the delivery is in sequence according to the recipient's user tag level.

Further, the recognition of the voice information by the machine recognition module is realized through a voice recognition model, and the establishment of the voice recognition model includes: collecting voice information between the consignee and the dispatcher, training the voice information and storing it .

Further, the machine recognition module is also used for semantic understanding of the voice information, and the semantic understanding is realized through voice rule configuration and voice similarity algorithm.

Further, the extraction of the reply voice by the machine reply module includes converting the text of the voice text into a reply voice, and the text of the voice text is based on the transliteration or closing speech of the consignee and the delivery person. The technique is determined.

The beneficial effects brought about by the technical solutions provided by the embodiments of the present invention are:

The embodiment of the present invention discloses a method and system for making a call based on an order. In the process of making a call through an order, the order is first scanned to determine the phone number to be dialed and the delivery address, and after the consignee is connected through The machine communicates with the consignee, and extracts the key words of the consignee’s voice information. The key words are used to complete the reply to the consignee. Once the machine is unable to communicate with the consignee, it will switch to manual operation through the delivery person. Communicate manually with the consignee. This kind of first communication between the machine and the consignee by phone can significantly improve the efficiency of the distribution during the delivery process of the courier, save the delivery time of the courier, and avoid the consignee from being unable to receive the goods in time during the distribution process. The shortcomings of information, and the direct call through the order improves the success rate of a delivery, reduces the risk of personal information leakage of the consignee, improves the professional service skills of the delivery staff, and reduces the rate of customer complaints.

Description of the drawings

In order to explain the technical solutions in the embodiments of the present invention more clearly, the following will briefly introduce the drawings needed in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

FIG. 1 is a schematic flowchart of a method for making a call based on an order disclosed in an embodiment of the present invention;

2 is a logical schematic diagram of machine communication in the method for making a call based on an order disclosed in an embodiment of the present invention;

Fig. 3 is a schematic diagram of the structural execution of the system for dialing a call based on an order disclosed in an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only Part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

Example one:

As shown in Figure 1, this embodiment provides a method for making a call based on an order, which includes the following steps:

S1: Scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

S2: Automatically connect and the machine recognizes the voice information of the consignee corresponding to the telephone number, and starts a machine dialogue;

S3: Extract the keywords of the voice information, extract the reply voice from the reply vocabulary corresponding to the keywords and send it to the consignee;

S4: When the reply voice cannot be extracted from the reply lexicon and reply to the consignee, the machine dialogue is converted into a manual dialogue.

Specifically, in the process of making a call through an order, the order is first scanned to determine the phone number to be called and the delivery address, after the consignee is connected, the machine communicates with the consignee through the machine, and the consignee’s information is retrieved. The key words of the voice message are used to complete the reply to the consignee. Once the machine cannot communicate with the consignee, it will switch to manual, and the delivery person and the consignee can communicate manually. This kind of first communication between the machine and the consignee by phone can significantly improve the efficiency of the distribution during the delivery process of the courier, save the delivery time of the courier, and avoid the consignee from being unable to receive the goods in time during the distribution process. The shortcomings of information, and the direct call through the order improves the success rate of a delivery, reduces the risk of personal information leakage of the consignee, improves the professional service skills of the delivery staff, and reduces the rate of customer complaints.

Preferably, the distribution strategy includes: sequential distribution according to the distance from the receiving place to the distribution station; or sequential distribution according to the order number scanning sequence; or smooth sequential distribution according to the historical call-through rate of the consignee answering the phone ; Or according to the consignee’s user tag level for delivery. Among them, in the process of automatically dialing a phone call through the machine, the above-mentioned delivery automatic call can be realized through a specific APP downloaded by a mobile phone, a delivery terminal, etc. When using this method to prepare for delivery, the delivery person can do it according to his own delivery needs. The choice of distribution strategy improves the satisfaction of the consignee while improving the efficiency of distribution.

Preferably, the machine recognition of the voice information is realized by a voice recognition model, and the establishment of the voice recognition model includes: collecting voice information between a consignee and a delivery person, training the voice information and storing it. Further, the machine recognition further includes a semantic understanding of the voice information, and the semantic understanding is realized through a voice rule configuration and a voice similarity algorithm. Figure 2 is a logical schematic diagram of the courier's machine communication. The voice recognition model of the consignee is mainly limited to the delivery time and location. The machine's understanding of the consignee is also limited. Once the machine recognizes The voice of the consignee cannot be ignored. For example, there is no response from the machine after a delay of about 5 seconds or the consignee directly hangs up the phone. This will be converted to manual communication. The delivery staff will manually confirm the delivery information with the consignee. , This process saves time mainly in: it can reduce the time of pressing the button for 5-8 seconds before making a call; it can reduce the time of making a call and the user's answering by 5 seconds; after the end of the communication with the user, it can reduce the recording time by 5-8 seconds Time; it can reduce the time for special phone calls. For example, when picking goods, you can make a phone call while picking the goods. In addition, you can improve the service level and unify the language and tone of the call.

Preferably, the reply voice included in the reply vocabulary is a text conversion through voice text, and the text of the voice text is determined according to the transcript or end of the conversation when the consignee and the delivery person communicate. When replying to the consignee by machine, after extracting the consignee’s keywords, it is necessary to search for the reply voice in the reply lexicon corresponding to the keyword. The reply voice is a voice converted from text, where the text includes both The common words of the consignee also include the common words of the delivery staff, such as the delivery location: rookie station, property management office, express cabinet, etc., and also include the delivery time. Among them, the process of machine dialogue is mainly based on the common delivery terms summarized after the manual dialogue between the delivery person and the delivery person after the machine learning.

Embodiment two:

As shown in Figure 3, this embodiment provides a system for dialing calls based on orders, including:

The order scanning module 1 is used to scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

The machine recognition module 2 is used to automatically connect and machine recognize the voice information of the consignee corresponding to the telephone number, and start a machine dialogue;

The machine reply module 3 is used to extract the keywords of the voice information, extract the reply voice from the lexicon corresponding to the keywords, and send it to the consignee;

The manual conversion module 4 is used for converting the machine dialogue into a manual dialogue when the machine reply module 3 cannot extract the reply voice from the reply lexicon and reply to the consignee.

Specifically, during the delivery process, the courier can use the system for delivery. You can choose the traditional manual delivery. The system is used for delivery. First, the order is scanned to determine the phone number to be called and the delivery address. After communication, the machine communicates with the consignee through the machine, and extracts the key words of the consignee’s voice information, and completes the reply to the consignee through the key words. Once the machine cannot communicate with the consignee, switch to manual. Manual communication between the delivery staff and the consignee. This kind of first communication between the machine and the consignee by phone can significantly improve the efficiency of the distribution during the delivery process of the courier, save the delivery time of the courier, and avoid the consignee from being unable to receive the goods in time during the distribution process. The shortcomings of information, and the direct call through the order improves the success rate of a delivery, reduces the risk of personal information leakage of the consignee, improves the professional service skills of the delivery staff, and reduces the rate of customer complaints.

Preferably, the order scanning module 1 includes a distribution strategy selection unit 11, and the selection of the distribution strategy selection unit 11 for order distribution includes: order delivery according to the distance from the receiving place to the distribution station; scanning according to the order number Delivery is in sequence; the historical call-through rate of the consignee is smoothly delivered in sequence; delivery is in sequence according to the user tag level of the consignee. When using the system, the delivery staff can manually initiate automatic dialing. At this time, the operation time is relatively variable, and they can also manually initiate automatic dialing. Or start automatic dialing regularly, and set the automatic dialing time according to the working time. After scanning the order code, the delivery staff can choose the delivery strategy according to their own delivery needs, which can improve the satisfaction of the consignee while providing delivery efficiency.

Preferably, the recognition of the voice information by the machine recognition module 2 is realized through a voice recognition model, and the establishment of the voice recognition model includes: collecting voice information between the consignee and the delivery person, training the voice information and storage. Further, the machine recognition module 2 is also used for semantic understanding of the voice information, which is realized through voice rule configuration and voice similarity algorithm. The voice recognition model of the consignee's voice recognition is mainly limited to the delivery time and delivery location. The machine's understanding of the consignee is also limited. Once the machine cannot recognize the consignee's voice, for example, the machine is still not performed after a delay of about 5 seconds. Or the consignee directly hung up the phone, which will be converted to manual communication. The delivery staff will manually confirm the delivery information with the consignee. This process saves time mainly in: it can reduce the number of calls Before pressing the key for 5-8 seconds; it can reduce the time of making a call and answering the user by 5 seconds; after communicating with the user, it can reduce the recording time by 5-8 seconds; it can reduce the time for special calls, such as when picking goods , Make calls while picking goods, in addition, you can also improve the service level, and unify the language and tone of the call.

Preferably, before the machine reply module 3 extracts the reply voice, it includes converting the text of the voice text into reply voice, and the text of the voice text is based on the transliteration or termination of the communication between the consignee and the delivery person. The words are ok. When replying to the consignee by machine, after extracting the consignee’s keywords, it is necessary to search for the reply voice in the reply lexicon corresponding to the keyword. The reply voice is a voice converted from text, where the text includes both The common words of the consignee also include the common words of the delivery staff, such as the delivery location: rookie station, property management office, express cabinet, etc., and also include the delivery time. Among them, the process of machine dialogue is mainly based on the common delivery terms summarized after the manual dialogue between the delivery person and the delivery person after the machine learning.

All the above-mentioned optional technical solutions can be combined in any way to form an optional embodiment of the present invention, which will not be repeated here.

It should be noted that the system for dialing calls based on orders provided in the above embodiment only uses the division of the above functional modules for example when delivering express delivery or takeaway. In actual applications, the above functional assignments can be divided into different groups according to needs. The function module is completed, that is, the internal structure of the system for making calls based on orders is divided into different function modules to complete all or part of the functions described above. In addition, the system for dialing a call based on an order provided in the above embodiment belongs to the same concept as the embodiment of the method for dialing a call based on an order. For the specific implementation process, please refer to the method embodiment, which will not be repeated here.

Those of ordinary skill in the art can understand that all or part of the steps in the above embodiments can be implemented by hardware, or by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The storage medium mentioned can be a read-only memory, a magnetic disk or an optical disk, etc.

The foregoing descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection of the present invention. Within range.

Claims

A method for making a call based on an order, which is characterized in that it comprises the following steps:

Scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

Automatically connect and the machine recognizes the voice message of the consignee corresponding to the telephone number, and starts the machine dialogue;

Extract the keywords of the voice information, extract the reply voice from the reply vocabulary corresponding to the keywords and send it to the consignee;

When the reply voice cannot be extracted from the reply lexicon and reply to the consignee, the machine dialogue is converted into a manual dialogue.
The method for making a call based on an order according to claim 1, wherein the distribution strategy comprises:

Order delivery according to the distance from the receiving place to the delivery station;

Or deliver them sequentially in the order of scanning the order number;

Or the historical call-through rate of the consignee’s answering calls is smoothly delivered sequentially;

Or according to the consignee’s user tag level for delivery.
The method of making a call based on an order according to claim 1, wherein the machine recognition of the voice information is realized through a voice recognition model, and the establishment of the voice recognition model includes: collecting information between the consignee and the dispatcher The voice information is trained and stored.
The method for making a call based on an order according to claim 1 or 3, wherein the machine recognition further comprises a semantic understanding of the voice information, and the semantic understanding is realized through a voice rule configuration and a voice similarity algorithm.
The method of making a call based on an order according to claim 1, wherein the reply voice included in the reply word database is a text conversion through voice text, and the text of the voice text is based on the communication between the consignee and the delivery person. The timing of the transcript clearing technique or the ending technique is determined.
A system for making a call based on an order, which is characterized in that it includes:

The order scanning module is used to scan the order number to obtain and save the order address and the phone number corresponding to the order, and automatically dial the phone number based on the delivery strategy of the order;

The machine recognition module is used to automatically connect and machine recognize the voice information of the consignee corresponding to the telephone number, and start a machine dialogue;

The machine reply module is used to extract the keywords of the voice message, extract the reply voice from the reply vocabulary corresponding to the keywords and send it to the consignee;

The manual conversion module is used to convert the machine dialogue into a manual dialogue when the machine reply module cannot extract the reply voice from the reply lexicon and reply to the consignee.
The system for making a call based on an order according to claim 6, wherein the order scanning module includes a delivery strategy selection unit, and the selection of the order delivery by the delivery strategy selection unit includes: arriving at the delivery station according to the receiving place Delivered in order according to the order number scanning order; Delivered in order according to the historical call-through rate of the consignee’s answering calls; Delivered in order according to the user’s tag level of the consignee.
The system for making a call based on an order according to claim 6, wherein the recognition of the voice information by the machine recognition module is realized through a voice recognition model, and the establishment of the voice recognition model includes: collecting consignees and The voice information between the delivery personnel is trained and stored.
The system for making a call based on an order according to claim 6 or 8, wherein the machine recognition module is also used for semantic understanding of the voice information, and the semantic understanding is configured by voice rule configuration and voice similarity algorithm. achieve.
The system for making a call based on an order according to claim 6, characterized in that, before the machine reply module extracts the reply voice, it includes converting the text of the voice text into a reply voice, and the text of the voice text is based on the received voice. The transcript or end of the conversation when the shipper and the delivery person communicate are determined.