Guiding Method, Server, and Storage Medium

Abstract

The present invention discloses a guiding method, a server and a storage medium. The method includes: receiving first reporting information reported by a vehicle that a user in the vehicle initiates navigation to an interest point through an in-vehicle system; querying the collection and note information of all users for the address of the interest point in a note information set according to the first reporting information; determining first voice broadcast content to be fed back according to the query result and the type of the interest point according to a predetermined rule; and sending the first voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to guide the user to add a note name to the interest point after completing the collection of the interest point on the map of the in-vehicle system. When the user in the vehicle initiates navigation to an interest point through the in-vehicle system, the present invention can guide the user to add a note name to the interest point after completing the collection of the interest point on the map of the in-vehicle system through the first voice broadcast content, so that the user can timely know and use the function that the vehicle can navigate to the note name, providing convenience for the user.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicles, and particularly relates to a guiding method, a server, and a storage medium. Background Art

[0002] Currently, when a user often goes to a certain place, the user directly searches for the name of the place by voice every time. Even though the function of the vehicle to navigate to the noted name has been launched, the user cannot timely know that the vehicle has the function of navigating to the noted name, resulting in a poor user experience. For example, every time the user wants to navigate to his uncle's house, he directly searches for the specific address of his uncle's house every time, which will affect the user's navigation experience. Summary of the Invention

[0003] The present invention provides a guiding method, a server, and a storage medium.

[0004] The present invention provides a guiding method. The guiding method includes: receiving first reporting information reported by a vehicle that a user in the vehicle initiates navigation to an interest point through an in-vehicle system; querying all users' collection and note information of the address of the interest point in a note information set according to the first reporting information; determining first voice broadcast content to be fed back according to a query result and the type of the interest point according to a predetermined rule; and sending the first voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to guide the user to add a noted name to the interest point after collecting the interest point through a map of the in-vehicle system.

[0005] In this way, when a user in the vehicle initiates navigation to an interest point through the in-vehicle system, the first voice broadcast content can be used to guide the user to add a noted name to the interest point after collecting the interest point through the map of the in-vehicle system, so that the user can timely know and use the function that the vehicle can navigate to the noted name, providing convenience for the user and improving the user experience.

[0006] After the step of receiving the first reporting information reported by the vehicle that the user in the vehicle initiates navigation to an interest point through the in-vehicle system, the guiding method includes: processing the name of the interest point, the search time and search times of the user for the interest point to obtain the type of the interest point for which a recommended note is to be given.

[0007] In this way, after receiving the first reporting information reported by the vehicle that the user in the vehicle initiates navigation to an interest point through the in-vehicle system, the server side can process the name of the interest point, the search time and search times of the user for the interest point to obtain the type of the interest point for which a recommended note is to be given to the user.

[0008] Determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule, includes: in the case that the user has collected the point of interest and has not added a note name to the point of interest, if the user has performed an operation of initiating navigation to the point of interest within a first preset time, determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to the predetermined rule.

[0009] In this way, in the case that the user has collected the point of interest and has not added a note name to the point of interest, if the user has performed an operation of initiating navigation to the point of interest within a first preset time, the first voice broadcast content to be fed back is determined according to the query result and the type of the point of interest according to the predetermined rule, so as to guide the user to add a note to the point of interest, so that for the next navigation, the user can directly say the note name of the point of interest to initiate navigation, improving the user experience.

[0010] Determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule, includes: according to the note information of the same type as the point of interest in the query result, selecting multiple candidate note names according to the predetermined rule to determine the first voice broadcast content to be fed back.

[0011] In this way, the user can select multiple candidate note names of the same type as the point of interest from the query result, and the feedback is given to the user through voice broadcast.

[0012] Determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule, includes: in the case that the user has not collected the point of interest, determining the first voice broadcast content for prompting the user to collect the point of interest to be fed back.

[0013] In this way, in the case that the user has not collected the point of interest yet, the user can be reminded by voice broadcast that the user can first collect the point of interest, and then reminded that the user can set a note for the point of interest, so that for the next navigation, the user can directly say the note name of the point of interest to initiate navigation, improving the user experience.

[0014] After the step of sending the first voice broadcast content to the vehicle for broadcast to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system, the guiding method includes: receiving second reporting information reported by the vehicle that the user in the vehicle has completed adding a note name to the point of interest; determining second voice broadcast content to be fed back according to the note name in the second reporting information; and sending the second voice broadcast content to the vehicle for broadcast to the user in the vehicle, so as to prompt the user to input a navigation request to navigate to the point of interest after the next voice wakes up the in-vehicle system.

[0015] In this way, when the user completes the operation of adding a note name to the point of interest, the server side can receive the second reporting information reported by the vehicle side in real time, generate second voice broadcast content according to the determined note name of the user, and send it to the vehicle side for voice broadcast feedback to the user in the vehicle, so as to prompt that a navigation request to navigate to the point of interest can be input after the next voice wakes up the in-vehicle system.

[0016] After the step of querying the collection and note information of the address of the point of interest by all users in the note information set according to the first reporting information, the guiding method includes: determining push card content to be fed back according to the query result and the type of the point of interest according to a predetermined rule; and sending the push card content to the vehicle for display to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0017] In this way, the present invention can also provide a way of pushing cards to the user to guide the user to collect and note the point of interest, improving the user experience.

[0018] The step of sending the push card content to the vehicle for display to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system includes: according to the similarity between the point of interest and the type of each point of interest, sequentially sending the push card content corresponding to the type of each point of interest to the vehicle for display to the user in the vehicle within a second preset time, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0019] In this way, the server can send the push card content corresponding to the type of each point of interest to the vehicle for display to the user in the vehicle in order from high to low similarity or from low to high similarity, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0020] The present invention provides a server. The server includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the guiding method described in any one of the above embodiments is implemented.

[0021] In this way, when a user in a vehicle initiates navigation to a point of interest through an in-vehicle system, the server of the present invention applies the above guiding method, and can guide the user to add a note name to the point of interest after completing the collection of the point of interest on the map of the in-vehicle system through the first voice broadcast content, so that the user can timely know and use the function that the vehicle can navigate to the note name, providing convenience for the user and improving the user experience.

[0022] The present invention further provides a non-volatile computer-readable storage medium containing a computer program. When the computer program is executed by one or more processors, the guiding method described in any one of the above embodiments is implemented.

[0023] In this way, when a user in a vehicle initiates navigation to a point of interest through an in-vehicle system, the storage medium of the present invention applies the above guiding method, and can guide the user to add a note name to the point of interest after completing the collection of the point of interest on the map of the in-vehicle system through the first voice broadcast content, so that the user can timely know and use the function that the vehicle can navigate to the note name, providing convenience for the user and improving the user experience.

[0024] The additional aspects and advantages of the present invention will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The above and / or additional aspects and advantages of the present invention will become obvious and easy to understand from the description of the embodiments in conjunction with the following drawings, where:

[0026] Figure 1 is one of the flow diagrams of the guiding method of the present invention;

[0027] Figure 2 is another flow diagram of the guiding method of the present invention;

[0028] Figure 3 is a third scenario diagram of the guiding method of the present invention;

[0029] Figure 4 is one of the flow diagrams of the voice interaction method of the present invention;

[0030] Figure 5 is another scenario diagram of the voice interaction method of the present invention;

[0031] Figure 6It is the third schematic flowchart of the voice interaction method of the present invention;

[0032] Figure 7 It is the fourth schematic flowchart of the voice interaction method of the present invention;

[0033] Figure 8 It is the fifth schematic flowchart of the voice interaction method of the present invention;

[0034] Figure 9 It is the sixth schematic flowchart of the voice interaction method of the present invention;

[0035] Figure 10 It is the seventh schematic flowchart of the voice interaction method of the present invention;

[0036] Figure 11 It is the eighth schematic flowchart of the voice interaction method of the present invention;

[0037] Figure 12 It is the ninth schematic flowchart of the voice interaction method of the present invention. Detailed implementation manners

[0038] The present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0039] When a user uses voice navigation, there are many points of interest that the user visits at a fixed frequency. For example, the gym that the user goes to twice a week, the parents' home that the user goes to once every half month, the dentist that the user goes to once a month, the friend's home that the user often goes to, the third uncle's home that the user occasionally goes to, the son's school that the user goes to every day, etc. These points of interest have extraordinary significance for the user. These meanings are more concise and more personable than the complete statements of the points of interest. Therefore, we hope that through the functions described in this patent, when the user goes to the above locations again, instead of saying the complete address, the user can directly say the note name to complete the navigation action. In this way, both the complexity of voice navigation is reduced, and the voice assistant appears more personable. For example: Suppose a user takes his son to the stadium to play badminton every Sunday: For the navigation of the point of interest, it is rather cumbersome for the user to navigate to this location. For example: Navigate to Gate XXX of XXX Sports Center on XXX Road in XXX District. However, by setting a note name for the point of interest, it is simpler and more personable for the user to navigate to this location. For example: Navigate to the badminton hall.

[0040] In the case where the user has never used the function of "navigating to the noted name" of the vehicle, the user cannot be informed in a timely manner that the vehicle has the function of navigating to the noted name. For example, every time the user wants to navigate to his uncle's house, he directly searches for the specific address of his uncle's house every time, which will affect the user's navigation experience.

[0041] In order to enable the user to complete the navigation task by saying [navigate to "the noted name"], it is necessary to first allow the user to note and collect the point of interest. At this time, according to the frequency of the user's navigation use and combined with the noted names of other users, the noted names for noting a certain point of interest can be provided to the user, so that the user can understand that there is a function of navigating to the noted name.

[0042] Please refer to Figure 1 , the present invention provides a guiding method. The guiding method includes:

[0043] 011: Receive the first reporting information reported by the vehicle that the user in the vehicle initiates navigation to a point of interest through the in-vehicle system;

[0044] 013: Query the collection and note information of all users for the address of the point of interest in the note information set according to the first reporting information;

[0045] 015: Determine the voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule;

[0046] 017: Send the voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to guide the user to add a noted name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0047] The present invention also provides a server. The server includes a processor and a memory, and a computer program is stored on the memory. The processor is used to receive the first reporting information reported by the vehicle that the user in the vehicle initiates navigation to a point of interest through the in-vehicle system; query the collection and note information of all users for the address of the point of interest in the note information set according to the first reporting information; determine the voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule; send the voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to guide the user to add a noted name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0048] The server of the present application is used to implement voice interaction between the vehicle and the user. Among them, the server can send a voice broadcast to the vehicle to conduct voice interaction with the user in the vehicle cockpit when the information reported by the vehicle meets the conditions; or the server can receive a voice request forwarded by the vehicle, process the voice request and issue a control instruction and / or a voice broadcast in combination with natural language understanding to complete the voice interaction, etc.

[0049] Specifically, for the guidance method of this application, first, the server receives the first reporting information reported by the vehicle that the user in the vehicle initiates navigation to a point of interest through the in-vehicle system.

[0050] The user in the vehicle initiates navigation to a point of interest through the in-vehicle system, which means that the user in the vehicle initiates navigation to a certain place by voice searching the map of the in-vehicle system. Among them, the point of interest (POI) refers to a specific address in the map of the in-vehicle system. In a geographic information system, a point of interest can be a house, a store, a mailbox, a bus stop, etc.

[0051] It can be understood that the map of the in-vehicle system can be a map application installed on the vehicle side. Taking the Amap application installed on the vehicle side as an example, there is a favorite folder associated with the user account in the map server corresponding to the Amap application, which is used to store the points of interest added by the user. The vehicle can communicate with the map server through the Amap application to achieve functions such as vehicle positioning, route planning, and navigation to a point of interest.

[0052] When the user in the vehicle initiates navigation to a point of interest through the in-vehicle system, the server of this application can simultaneously receive the first reporting information reported by the vehicle that the user has initiated navigation to a point of interest through the in-vehicle system. The first reporting information includes the user's identity information, the message that the user initiates navigation to a point of interest, the name of the point of interest, the current search time of the user for the point of interest, the historical search time, and the number of searches by the user before the current search time.

[0053] It can be understood that before the navigation function of the user initiating navigation to a point of interest, when the server receives the first reporting information reported by the vehicle, it needs to first determine whether the navigation meets the condition for triggering "voice push" at this time. If it meets the condition for triggering "voice push", then voice push is performed, that is, voice broadcast content is fed back to the user.

[0054] Therefore, the server needs to first query the collection and note information of all users for the address of the point of interest in the note information set according to the first reporting information. Among them, all users refer to all users with account information on the server side.

[0055] For example, the current user can initiate navigation to a point of interest through a voice request: "Navigate to XX Hotel". At this time, the vehicle map searches for "XX Hotel" according to this voice request and queries that other user 1 has collected the address of this "XX Hotel", and the note information is to note this "XX Hotel" as "Company", and other user 2 has collected the address of this "XX Hotel", and the note information is to note this "XX Hotel" as "Frequently visited hotel".

[0056] Then, according to the query results and the types of points of interest, the voice broadcast content to be fed back is determined according to a predetermined rule. Among them, in this application, according to the scenario where the user makes a note for the point of interest, the points of interest can be classified into types of points of interest including entertainment venue type, company type, and home type, realizing the custom classification of points of interest, so as to combine with voice push to guide the user to make a note. The query results are the names collected and noted by the above-mentioned other users 1 and user 2 for the "XX Hotel", which include "company" and "frequently visited hotel" respectively. The predetermined rule refers to the way of comprehensively determining the voice broadcast content to be fed back according to whether the user has collected the point of interest, the point of interest and the query results.

[0057] It can be understood that the purpose of distinguishing the types of noted points of interest in the present invention is to push different voice broadcast contents for different types of points of interest. For example, the voice broadcast content corresponding to the company type of point of interest is "Try navigating to the company", and for the entertainment venue type of point of interest, it is more inclined to guide the user to note it as "frequently visited supermarket", "frequently visited cinema", and the corresponding voice broadcast content is "Try saying to me: Navigate to the frequently visited cinema". For different types of points of interest, the forms in which users are expected to establish memories are different.

[0058] More specifically, for example, if the type of the point of interest "XX Hotel" belongs to the entertainment venue type, combined with the above query results, if the user has completed the collection of the point of interest through the map of the in-vehicle system, the voice broadcast content to be fed back can be determined according to the predetermined rule as "5 vehicle owners have already noted this address, and the noted names are 'frequently visited hotel' and 'company'. What do you want to note it as?"

[0059] Finally, the voice broadcast content is sent to the vehicle to be broadcast to the user in the vehicle, so as to guide the user to add a noted name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

[0060] In this way, when the user in the vehicle initiates navigation to a point of interest through the in-vehicle system, the user can be guided by voice broadcast to first complete the collection of the point of interest through the map of the in-vehicle system and then add a noted name to the point of interest, so that the user can timely know and use the function that the vehicle can navigate to the noted name. This method provides convenience for the user and improves the user experience.

[0061] After step 011, the guiding method includes: processing the name of the point of interest, the search time and search times of the user for the point of interest to obtain the type of the point of interest for which a recommended note is to be made.

[0062] The processor is used to process the name of the point of interest, the search time and search times of the user for the point of interest to obtain the type of the point of interest for which a recommended note is to be made.

[0063] As can be seen from the foregoing, the types of points of interest may include entertainment venue type, company type, and home type. Since the first reported information includes the name of the point of interest, the user's current search time and historical search time for the point of interest, and the number of searches by the user before the current search time, the type of the point of interest recommended for the user to add a note is determined by the name of the point of interest, the user's search time for the point of interest, and the number of searches.

[0064] It can be understood that if the user has searched a relatively large number of times within a predetermined time period before the current search time for the point of interest, for example, the number of searches reaches 2 or more within 1 month before the current time point, it means that this point of interest is an address frequently used by the user. Therefore, the user can be recommended to add a note to this point of interest to improve the user experience.

[0065] For example, when the name of the point of interest is "XX Hotel" and the user has searched this point of interest a total of 4 times within 1 month before the current search time, the type of the point of interest recommended by the server for the user to add a note can be the entertainment venue type.

[0066] In this way, after receiving the first reported information in which the in-vehicle user in the vehicle reports and initiates navigation to the point of interest through the in-vehicle system, the server can process according to the name of the point of interest, the user's search time for the point of interest, and the number of searches to obtain the type of the point of interest recommended for the user to add a note.

[0067] In addition, step 015 includes: in the case where the user has favorited a point of interest and has not added a note name for the point of interest, if the user has performed an operation of initiating navigation to the point of interest within the first preset time, then according to the query result and the type of the point of interest, determine the first voice broadcast content to be fed back according to a predetermined rule.

[0068] The processor is used to, in the case where the user has favorited a point of interest and has not added a note name for the point of interest, if the user has performed an operation of initiating navigation to the point of interest within the first preset time, then according to the query result and the type of the point of interest, determine the first voice broadcast content to be fed back according to a predetermined rule.

[0069] The first preset time refers to within a preset time before the current time point of initiating navigation to the point of interest. For example, the first preset time is 1 month as described above, or it can be other time ranges, which are not limited here.

[0070] The predetermined rule here can be to integrate the query result and the type of the point of interest to obtain the first voice broadcast content.

[0071] When the user has favorited a point of interest and has not added a note name for the point of interest, if the user has performed an operation of initiating navigation to the point of interest within the first preset time. For example, if the first preset time is 1 month, then within 1 month before the current time point of initiating navigation to the point of interest, it is found that the user has performed the operation of initiating navigation to the point of interest, and the number of initiations can be 1 or more than 1 time. And when querying the collection and note information of all users for the address of the point of interest in the note information set, the query result is that 5 car owners have favorited and noted this address as the company, and the note names noted by other users for this address also include 'company', 'friend's company', 'husband's company', etc. If the type of this point of interest belongs to the company type, then according to the query result and the type of the point of interest, determine the first voice broadcast content to be fed back according to a predetermined rule. For example, the content of the first voice broadcast can be: "5 car owners have noted this address as the company, and the note names noted by other users for this address include 'company', 'friend's company', 'husband's company', etc. What do you want to note it as?" Thus, guide the user to make a note for this point of interest.

[0072] In this way, when the user has favorited a point of interest and has not added a note name for the point of interest, if the user has performed an operation of initiating navigation to the point of interest within the first preset time, then according to the query result and the type of the point of interest, determine the first voice broadcast content to be fed back according to a predetermined rule, so as to guide the user to make a note for this point of interest, so that for the next navigation, the user can directly say "navigate to the note name of the point of interest" to initiate navigation, improving the user experience.

[0073] It can be understood that when determining the first voice broadcast content to be fed back to the user, there may be a situation where there are multiple candidate note names selected. At this time, the server side of the present invention can select the note name of the same type as the point of interest and feed it back to the user.

[0074] Specifically, step 015 further includes: according to the note information of the same type as the point of interest in the query result, select multiple candidate note names according to a predetermined rule to determine the first voice broadcast content to be fed back.

[0075] The processor is used to select multiple candidate note names according to the note information of the same type as the point of interest in the query result according to a predetermined rule to determine the first voice broadcast content to be fed back.

[0076] The predetermined rule here can refer to the rule determined according to the degree of conformity between the note name in the query result and the type of the point of interest. The more the note name in the query result conforms to the type of the point of interest, the greater the probability of being fed back to the user as a candidate note name. Correspondingly, the order of voice broadcasting this note name is more forward.

[0077] For example, if the point of interest (POI) that the user initiates navigation to is "XX Office Building", which means the type of this POI is corporate. When querying the collection and annotation information of all users for the address of this POI "XX Office Building" in the annotation information set, the query results are as follows: 4 car owners have collected and annotated this address as a company, and the names annotated by other users for this address also include 'company', 'friend's company', and 'husband's company'. Then, according to the corporate type of the POI, three candidate annotation names, namely 'company', 'friend's company', and 'husband's company', can be selected and recommended to the user. That is, the first voice broadcast content to be fed back can be "4 car owners have annotated this address as a company, and the names annotated by other users for this address are 'company', 'friend's company', 'husband's company', etc. What would you like to annotate it as?" to guide the user to annotate this POI.

[0078] In this way, the user can select multiple candidate annotation names of the same type as the POI from the query results and have them fed back to the user through voice broadcast.

[0079] It can be understood that when determining the first voice broadcast content to be fed back to the user, the user may not have collected this POI, let alone made an annotation. In this case, the user needs to be reminded to first collect this POI and then to annotate this POI.

[0080] Specifically, step 015 further includes: when the user has not collected the POI, determining the first voice broadcast content used to prompt the user to collect the POI.

[0081] The processor is used to determine the first voice broadcast content used to prompt the user to collect the POI when the user has not collected the POI.

[0082] For example, if this POI has not been collected, the first voice broadcast content is "XX car owners have collected this address. You can try saying 'Collect this address'".

[0083] In this way, when the user has not yet collected this POI, the user can be reminded by voice broadcast to first collect this POI, and then be reminded that the user can set an annotation for this POI, so that for the next navigation, the user can directly say "Navigate to the annotation name of the POI" to initiate navigation, improving the user experience.

[0084] In addition, when the user completes the operation of adding an annotation name to the POI, the server can receive this information in real time, generate the second voice broadcast content based on the determined annotation name by the user, and send it to the vehicle end to be fed back to the user in the vehicle through voice broadcast, to prompt that after the vehicle-mounted system is woken up by voice next time, the user can enter a navigation request to navigate to the POI.

[0085] Please refer to Figure 2, after step 017, the guiding method includes:

[0086] 0171: Receive second reporting information reported by the vehicle that the user in the vehicle has completed adding a note name to the point of interest;

[0087] 0172: Determine the second voice broadcast content to be fed back according to the note name in the second reporting information;

[0088] 0173: Send the second voice broadcast content to the vehicle for broadcast to the user in the vehicle, so as to prompt the user to input a navigation request to the point of interest after the next voice wakes up the in-vehicle system.

[0089] The processor is used to receive second reporting information reported by the vehicle that the user in the vehicle has completed adding a note name to the point of interest; determine the second voice broadcast content to be fed back according to the note name in the second reporting information; send the second voice broadcast content to the vehicle for broadcast to the user in the vehicle, so as to prompt the user to input a navigation request to the point of interest after the next voice wakes up the in-vehicle system.

[0090] The second reporting information includes information such as the user's identity information, the message that the user has completed adding a note name to the point of interest, the address of the point of interest, and the note name added to the point of interest.

[0091] For example, after the user notes "My Company" for the point of interest, the server can receive in real time the second reporting information reported by the vehicle that the user in the vehicle has completed adding a note name to the point of interest, and can further determine the second voice broadcast content to be fed back to the user according to the note name in the second reporting information. The second voice broadcast content can be "You can try 'Navigate to My Company' next time".

[0092] Then, the server will send the second voice broadcast content "You can try 'Navigate to My Company' next time" to the vehicle for broadcast to the user in the vehicle, so as to prompt the user to input a navigation request to the point of interest after the next voice wakes up the in-vehicle system. The navigation request to the point of interest is "Navigate to My Company".

[0093] In this way, when the user completes the operation of adding a note name to the point of interest, the server can receive in real time the second reporting information reported by the vehicle end, generate the second voice broadcast content according to the determined note name by the user, and send it to the vehicle end for voice broadcast feedback to the user in the vehicle, so as to prompt that a navigation request to the point of interest can be input after the next voice wakes up the in-vehicle system.

[0094] Understandably, in addition to guiding the user to collect and make notes on the point of interest by means of voice broadcast on the vehicle side, it is also possible to guide the user to collect and make notes on the point of interest by means of card push on the vehicle's central control screen. For example, when the user is not navigating, if the vehicle's voice assistant suddenly makes a voice broadcast at this time, it will seem very strange. Therefore, when the user is not performing a navigation action, a card push method can also be provided to guide the user to collect and make notes on the point of interest.

[0095] Specifically, after step 013, the guiding method includes: determining the content of the push card to be fed back according to the query result and the type of the point of interest according to a predetermined rule; and

[0096] Sending the content of the push card to the vehicle to be displayed to the user in the vehicle, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the vehicle-mounted system.

[0097] The processor is used to determine the content of the push card to be fed back according to the query result and the type of the point of interest according to a predetermined rule; send the content of the push card to the vehicle to be displayed to the user in the vehicle, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the vehicle-mounted system.

[0098] The predetermined rule here may refer to a rule determined according to the degree of conformity between the note name in the query result and the type of the point of interest. The more the note name in the query result conforms to the type of the point of interest, the greater the probability of being fed back to the user as a candidate note name. Correspondingly, the order of displaying the note name in the recommended card on the vehicle's central control screen is more forward. For example, Figure 3 As shown, the note names "Company", "Friend's Company", and "Husband's Company" that conform to the type of the point of interest being a company type are displayed in the card content.

[0099] Determining the content of the push card to be fed back according to the query result and the type of the point of interest, that is, for example, the point of interest for which the user initiates navigation is "XX Office Building", that is, the type of this point of interest is a company type. The query result of querying all users' collection and note information for the address of this point of interest "XX Office Building" in the note information set is: 4 car owners have collected and noted this address as a company, and the note names noted by other users for this address are also 'Company', 'Friend's Company', 'Husband's Company'. Then, according to the type of the point of interest being a company type, the three candidate note names "Company", "Friend's Company", and "Husband's Company" can be selected and recommended to the user. That is, the content of the card to be fed back can be "4 car owners have noted this address as a company, and the note names noted by other users for this address are 'Company', 'Friend's Company', 'Husband's Company', etc. You can select one of the above note names to set a note for the current point of interest", and the card content can be as followsFigure 3 As shown. The user can click on 'Company', 'Friend's Company', or 'Husband's Company' in the card, select a note name from them, or after seeing the card, the user can select a note name through voice input to set a note for the current point of interest.

[0100] Then, after the server determines the card content, the server will push the card content to the vehicle and display it to the user in the vehicle, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the in-vehicle system.

[0101] In this way, the present invention can also provide a way of pushing cards to the user, guiding the user to collect and note the point of interest, and improving the user experience.

[0102] In addition, since the server may not be able to clearly determine which type the selected point of interest of the user conforms to, at this time, it is necessary to compare the similarity between the selected point of interest of the user and the type of each point of interest. Through this similarity, the server can determine to send the push card content corresponding to the type of each point of interest to the vehicle and display it to the user in the vehicle in order from high to low similarity or from low to high similarity, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the in-vehicle system.

[0103] Specifically, step 0142 includes: according to the similarity between the point of interest and the type of each point of interest, within the second preset time, sequentially send the push card content corresponding to the type of each point of interest to the vehicle and display it to the user in the vehicle, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the in-vehicle system.

[0104] The processor is used to, according to the similarity between the point of interest and the type of each point of interest, within the second preset time, sequentially send the push card content corresponding to the type of each point of interest to the vehicle and display it to the user in the vehicle, so as to guide the user to add a note name to the point of interest after collecting the point of interest through the map of the in-vehicle system.

[0105] The similarity between the point of interest and the type of each point of interest can be obtained by matching according to the similarity comparison model.

[0106] The preset order includes the order from high to low similarity or the order from low to high similarity.

[0107] The second preset time can refer to the time after the server determines the push card content corresponding to the type of 1 point of interest. For example, the second preset time can be 1s, 1.2s, 2s, 2.4s, 3s, 4s, 5s, 6s, 7s, and 8s, etc., and there is no limit here.

[0108] For example, if the similarity matching results between the POIs selected by the user and each POI type obtained according to the POI classification model are: company type: 90%, indicating that the similarity between this POI and the POIs of the company type is 90%; home type: 50%, indicating that the similarity between this POI and the POIs of the home type is 50%; entertainment venue type: 70%, indicating that the similarity between this POI and the POIs of the entertainment venue type is 70%.

[0109] Taking the preset order as the order from high to low similarity for example, from the above similarity matching results, it can be seen that the POI selected by the user most conforms to the POIs of the company type. Secondly, the POI selected by the user relatively conforms to the POIs of the entertainment venue type, and the probability that the POI selected by the user conforms to the POIs of the home type is relatively small.

[0110] Therefore, the server can preferentially push cards of the company type to the user. The content of the company type card can be: Other car owners have already noted "XXX Company" as: friend's company, my company. Come and try to choose a note name to make a note?

[0111] After pushing the company type card, within the second preset time, continue to push the cards of the entertainment venue type. The content of the entertainment venue card can be: Other car owners have already noted "XXX Shopping Mall" as: frequently visited cinema, my favorite cinema. Come and try to choose a note name to make a note?

[0112] After pushing the cards of the entertainment venue type, within the second preset time, continue to push the cards of the home type. The content of the home type card can be: Other car owners have already noted "XXX Road, No. XXX" as: friend's home, my home, my third uncle's home. Come and try to choose a note name to make a note?

[0113] In this way, the server can send the content of the push cards corresponding to each POI type to the vehicle in order from high to low similarity or from low to high similarity to display to the user in the vehicle, so as to guide the user to add a note name to the POI after completing the collection of the POI through the map in the in-vehicle system.

[0114] In addition, in the case where the user has already collected a certain address on the map but has not made a note for it, at this time, according to the user's navigation to the unnoted POI in the favorites, remind the user how to make a note for this point, so that the user can understand the function of navigating to the POI.

[0115] Specifically, please refer to Figure 4 , the present invention provides a voice interaction method. The voice interaction method includes:

[0116] 021: Receive the current navigation request forwarded by the vehicle; wherein, the current navigation request is a navigation request initiated by a user in the vehicle cockpit through voice to a point of interest in the favorites folder associated with the user account in the map server;

[0117] 022: Generate a voice broadcast indicating that the navigation route has been selected based on the route confirmed by the user and send it to the vehicle;

[0118] 023: After determining that the point of interest of the current navigation request does not contain a note name, generate a voice broadcast to guide the user to add a note name to the point of interest and send it to the vehicle, so that the user can add a note name to the point of interest;

[0119] 024: When the user in the vehicle cockpit adds a note name to the point of interest, add attribute information including the note name to the corresponding point of interest in the favorites folder associated with the user account in the map server;

[0120] 025: Receive the next navigation request initiated by the user through voice to the note name forwarded by the vehicle;

[0121] 026: Perform intent recognition and slot recognition on the next navigation request. After confirming that the intent recognition result is a navigation intent, compare whether the point of interest in the slot recognition result exactly matches the attribute information including the note name;

[0122] 027: Send the matching result to the map server to complete the navigation in response to the next navigation request.

[0123] The processor is used to receive the current navigation request forwarded by the vehicle; wherein, the current navigation request is a navigation request initiated by a user in the vehicle cockpit through voice to a point of interest in the favorites folder associated with the user account in the map server; generate a voice broadcast indicating that the navigation route has been selected based on the route confirmed by the user and send it to the vehicle; after determining that the point of interest of the current navigation request does not contain a note name, generate a voice broadcast to guide the user to add a note name to the point of interest and send it to the vehicle, so that the user can add a note name to the point of interest; when the user in the vehicle cockpit adds a note name to the point of interest, add attribute information including the note name to the corresponding point of interest in the favorites folder associated with the user account in the map server; receive the next navigation request initiated by the user through voice to the note name forwarded by the vehicle; perform intent recognition and slot recognition on the next navigation request. After confirming that the intent recognition result is a navigation intent, compare whether the point of interest in the slot recognition result exactly matches the attribute information including the note name; send the matching result to the map server to complete the navigation in response to the next navigation request.

[0124] Specifically, first, the navigation request forwarded by the vehicle is received. Among them, this navigation request is a navigation request initiated by the user in the vehicle cabin through voice to navigate to the point of interest in the favorites associated with the user account in the map server. For example, the point of interest in the favorites associated with the user account in the map server is "XXX District XXX Road XXX Sports Center XXX Gate", then the navigation request initiated by the user through voice is "Navigate to XXX District XXX Road XXX Sports Center XXX Gate". At this time, the map on the vehicle side will search for one or more routes based on the navigation request, and a pop-up window of the route list can be popped up.

[0125] Then, a voice announcement indicating that the navigation route has been selected is generated based on the route confirmed by the user and sent to the vehicle. That is, for example, when there are routes 1, 2, and 3 in the route list, the pop-up window can be as follows: Figure 5 When the user clicks to select Route 1 as the navigation route, the server generates a voice announcement indicating that the navigation route has been selected based on Route 1 confirmed by the user, which may be "Route 1 has been selected for you", and sends the voice announcement "Route 1 has been selected for you" to the vehicle, so that the vehicle initiates navigation based on Route 1.

[0126] Next, after the vehicle receives a voice announcement indicating that the navigation route has been selected based on the route confirmed by the user, the voice server may be triggered to add a voice announcement to guide the user to add a note name to the point of interest.

[0127] That is, after determining that the point of interest in this navigation request does not contain a remark name, the server can generate a voice announcement to guide the user to add a remark name to the point of interest and send it to the vehicle so that the user can add a remark name to the point of interest. For example, after determining that the point of interest "XXX District XXX Road XXX Sports Center XXX Gate" in this navigation request "Navigate to XXX District XXX Road XXX Sports Center XXX Gate" does not contain a remark name, the server generates a voice announcement to guide the user to add a remark name to the point of interest, such as:

[0128] "By the way, I can also help you give your destination a name. For example, you can say to me [remark the destination as the old place] and I will remember it."

[0129] Or:

[0130] "Oh, by the way, you can give this destination a name. Try saying [remember the destination as old place] to unlock my learning skill'";

[0131] Or:

[0132] "By the way, you can say to me: [Set the destination note as Badminton Hall]. Next time, you can just say [Navigate to Badminton Hall]. Remember? It's [Set the destination note as Badminton Hall]."

[0133] After that, when the user in the vehicle cockpit adds a note name to a point of interest, add attribute information including the note name to the corresponding point of interest in the favorites folder associated with the user account in the map server, so that when the user initiates the function of navigating to the note name next time, the navigation can be completed according to the point of interest and the attribute information including the note name. It can be understood that each point of interest of the present invention can have multiple attribute information, including specific address, user account, note name and other attribute information. For example, when the user in the vehicle cockpit adds the note name "Badminton Hall" to the point of interest "Navigate to the XXXX Gate of the Sports Center on XXXX Road, XXXX District", the attribute information including the note name "Badminton Hall" can be added to the corresponding point of interest "XXXX Gate of the Sports Center on XXXX Road, XXXX District" in the favorites folder associated with the user account in the map service.

[0134] Then, when the user needs to initiate the function of navigating to the note name for navigation next time, the vehicle side can receive the next navigation request initiated by the user through voice forwarded by the vehicle, perform intent recognition and slot recognition on the next navigation request, and compare whether the point of interest in the slot recognition result exactly matches the attribute information including the note name after confirming that the intent recognition result is a navigation intent. Then the vehicle side sends the matching result to the map server to complete the navigation in response to the next navigation request.

[0135] In this way, the present invention can, on the premise that the user has already favorited a point of interest, trigger the voice server side to append a voice broadcast after the vehicle side receives the voice broadcast indicating that the selected navigation route has been generated according to the route confirmed by the user, so as to guide the user to add a note name to the point of interest, so that the user can use the function of navigating to the note name of the vehicle in the next navigation.

[0136] It should be noted that the frequency of the voice broadcast guiding the user to add a note name to the point of interest can be the default frequency set by the in-vehicle system. For example, it can be once a day, or other frequencies, which are not limited here.

[0137] In addition, the frequency of the voice broadcast can also be restricted. For example, for navigation requests of the same type, the voice broadcast guiding the user to add a note name to the point of interest is triggered at most once every 14 days.

[0138] For the next navigation request sent by the user to navigate to a single destination, for example, the next navigation request is "Navigate to the badminton hall", the point of interest in the slot recognition result and the attribute information including the note name can be completely matched one by one. Correspondingly, step 027 includes: sending the matching result of the complete match between the point of interest in the slot recognition result and the attribute information including the note name to the map server, and the map server and the in-vehicle system complete the map communication to navigate to the point of interest in the favorite folder corresponding to the attribute information including the note name.

[0139] The processor is used to send the matching result of the complete match between the point of interest in the slot recognition result and the attribute information including the note name to the map server, and the map server and the in-vehicle system complete the map communication to navigate to the point of interest in the favorite folder corresponding to the attribute information including the note name.

[0140] When the user's next navigation request to navigate to the note name sent by voice is "Navigate to the badminton hall", after the intention recognition result obtained by performing intention recognition on this navigation request is a navigation intention, and the point of interest in the slot recognition result obtained by performing slot recognition is "badminton hall", and the point of interest in this slot recognition result is completely matched with the attribute information including the note name "badminton hall", then this matching result can be sent from the voice server to the map server.

[0141] The map server and the in-vehicle system complete the map communication to navigate to the point of interest in the favorite folder corresponding to the attribute information including the note name, that is, the map server enables the map of the in-vehicle system to navigate to the point of interest "Gate xx, xx Sports Center, xx Road, xx District" corresponding to the "badminton hall" in the favorite folder.

[0142] In this way, when the user sends a navigation request to the note name by voice next time, the voice server can perform intention recognition and slot recognition on the navigation request. When the intention recognition result is a navigation intention, it sends the matching result of the complete match between the point of interest in the slot recognition result and the attribute information including the note name to the map server, and completes the process of navigating the map of the in-vehicle system to the point of interest in the favorite folder through the map server.

[0143] In addition, step 027 includes: sending the matching result of the incomplete match between the point of interest in the slot recognition result and the attribute information including the note name to the map server, and the map server and the in-vehicle system complete the map communication to navigate to the point of interest selected by the user, and the point of interest selected by the user is determined by navigating and searching based on the point of interest in the slot recognition result.

[0144] The processor is used to send the matching results in which the points of interest in the slot recognition result do not fully match the attribute information including the note name to the map server. The map server communicates with the map of the vehicle-mounted system to complete navigation to the point of interest selected by the user. The point of interest selected by the user is determined by performing a navigation search based on the point of interest in the slot recognition result.

[0145] That is, when the matching results in which the points of interest in the slot recognition result do not fully match the attribute information including the note point are sent to the map server, the map server can perform a navigation search on the map end of the vehicle-mounted system based on the points of interest in the slot recognition result to obtain one or more points of interest in the map favorites. The one or more points of interest can be presented in a list on the map of the vehicle-mounted system for the user to select.

[0146] For example, when the next navigation request initiated by the user is "Navigate to the frequently visited badminton hall", and the point of interest "frequently visited badminton hall" in the slot recognition result does not fully match the attribute information "badminton hall" including the note name, the map server can perform a navigation search in the corresponding map favorites based on the point of interest "frequently visited badminton hall" in the slot recognition result to obtain one or more stadium addresses collected by the user. The one or more stadium addresses are the one or more points of interest in the map favorites.

[0147] After the user selects one of the points of interest in the map favorites, the map server communicates with the map of the vehicle-mounted system to complete navigation to the point of interest selected in the favorites by the user.

[0148] In this way, when the points of interest in the slot recognition result in the next navigation request initiated by the user do not fully match the attribute information including the note name, the map server can perform a navigation search on the map end of the vehicle-mounted system based on the points of interest in the slot recognition result to obtain one or more points of interest in the map favorites, and complete navigation according to the point of interest selected by the user in one of the map favorites.

[0149] In addition, for the multi-purpose next navigation request issued by the user, for example, the next navigation request is "Navigate to the badminton hall first and then go home". That is, when multiple points of interest are included in the user's navigation request, the matching results in which at least one of the multiple points of interest in the slot recognition result fully matches the attribute information including the note name can be sent to the map server, so as to complete navigation to the point of interest in the favorites corresponding to the attribute information of the note name.

[0150] Specifically, step 027 includes: sending the matching results in which at least one of the multiple points of interest in the slot recognition result fully matches the attribute information including the note name to the map server, and the map server communicates with the map of the vehicle-mounted system to complete navigation to the point of interest in the favorites corresponding to the attribute information including the note name.

[0151] The processor is used to send the matching result of fully matching at least one of multiple points of interest in the slot recognition result and the attribute information including the note name to the map server, and the map server communicates with the map of the vehicle system to complete navigation to the point of interest in the favorite corresponding to the attribute information including the note name.

[0152] For example, if the next navigation request is "navigate to the badminton hall first and then go home", the slot recognition result includes 2 points of interest, namely the "badminton hall" and the "home".

[0153] When the points of interest in the favorite of the map of the vehicle system include the attribute information "badminton hall" with a note name but do not include the attribute information "home", the matching result in the voice server can be: one of the 2 points of interest in the slot recognition result, the "badminton hall", fully matches the attribute information "badminton hall" with a note name. Therefore, the map server communicates with the map of the vehicle system to complete navigation to the point of interest in the favorite corresponding to the attribute information "badminton hall" with a note name.

[0154] When the points of interest in the favorite of the map of the vehicle system include the attribute information "badminton hall" with a note name and also include the attribute information "home" with a note name, the matching result in the voice server can be: the 2 points of interest in the slot recognition result, namely the "badminton hall" and the "home" respectively, fully match the attribute information "badminton hall" and "home" with note names respectively. Therefore, the map server communicates with the map of the vehicle system and can combine semantic understanding to first navigate to the point of interest in the favorite corresponding to the attribute information "badminton hall" with a note name, and then navigate to the point of interest in the favorite corresponding to the attribute information "home" with a note name.

[0155] In this way, the present invention can realize a navigation request with multiple points of interest. When at least one of the multiple points of interest in the slot recognition result fully matches the attribute information including the note name, the map server communicates with the map of the vehicle system to complete navigation to the point of interest in the favorite corresponding to the attribute information including the note name, and can navigate to one or more points of interest.

[0156] After determining that the point of interest in the current navigation request does not include a note name, a voice broadcast for guiding the user to add a note name to the point of interest is generated and sent to the vehicle, so that the user can add a note name to the point of interest. The following describes how the user adds a note name to the point of interest.

[0157] Specifically, after step 022, the voice interaction method includes: receiving the voice request from the vehicle for the user to add a note name to the point of interest; and

[0158] Identifying the voice request to obtain the note name of the point of interest added by the user.

[0159] The processor is used to receive the voice request from the vehicle forwarded by the user to add a note name to the point of interest; and identify the voice request to obtain the note name of the point of interest added by the user.

[0160] The voice server first receives the voice request from the vehicle forwarded by the user to add a note name to the point of interest. For example, the voice request for the user to add the note name "Badminton Hall" to the point of interest "Gate of XXX Sports Center, XXX Road, XXX District" can be "Note Gate of XXX Sports Center, XXX Road, XXX District as Badminton Hall". This voice request can also be other sentence patterns expressing the meaning of note, which is not limited here.

[0161] Then, the voice server can perform natural language understanding and recognition on this voice request to obtain the note name of the point of interest added by the user. When the voice request is "Note Gate of XXX Sports Center, XXX Road, XXX District as Badminton Hall", the recognition of this voice request can obtain the note name added by the user as "Badminton Hall".

[0162] In this way, the user can add a note name to the point of interest by voice.

[0163] After the user adds a note name to the point of interest by voice, the map server can add attribute information including the note name to the corresponding point of interest in the favorite folder associated with the user account. Specifically, how to add attribute information including the note name to the corresponding point of interest in the favorite folder associated with the user account in the map server is described below.

[0164] Specifically, step 024 includes: when the user in the vehicle cockpit adds a note name to the point of interest, according to the recognized note name of the point of interest added by the user, add attribute information including the note name to the corresponding point of interest in the favorite folder associated with the user account in the map server.

[0165] The processor is used to, when the user in the vehicle cockpit adds a note name to the point of interest, according to the recognized note name of the point of interest added by the user, add attribute information including the note name to the corresponding point of interest in the favorite folder associated with the user account in the map server.

[0166] For example, when the user in the vehicle cockpit adds the note name "Badminton Hall" to the point of interest "Gate of XXX Sports Center, XXX Road, XXX District", the voice server can, according to the recognized note name "Badminton Hall" of the point of interest added by the user, add attribute information including the note name "Badminton Hall" to the corresponding point of interest "Gate of XXX Sports Center, XXX Road, XXX District" in the favorite folder associated with the user account in the map server, so that when the user issues a navigation request next time, the user can directly issue a voice request to navigate to the note name to initiate navigation.

[0167] After step 024, the voice interaction method includes: issuing a voice broadcast that guides the user to initiate a navigation request to the note name by voice next time according to the added attribute information including the note name, so that the user can initiate a navigation request to the note name by voice next time.

[0168] The processor is used to issue a voice broadcast that guides the user to initiate a navigation request to the note name by voice next time according to the added attribute information including the note name, so that the user can initiate a navigation request to the note name by voice next time.

[0169] That is to say, the server can issue a voice broadcast that guides the user to initiate a navigation request to the note name by voice next time according to the added attribute information including the note name, so that the user can initiate a navigation request to the note name by voice next time.

[0170] For example, when the user adds the attribute information including the note name "Badminton Hall" to the point of interest "Gate XXX, XXX Sports Center, XXX Road, XXX District", the content of the voice broadcast can be "Okay, the Gate XXX, XXX Sports Center, XXX Road, XXX District has been noted as Badminton Hall. Next time, just say [Navigate to Badminton Hall] and it will work."

[0171] In this way, the present invention can issue a voice broadcast that guides the user to initiate a navigation request to the note name by voice next time according to the added attribute information including the note name, so that the user can initiate a navigation request to the note name by voice next time.

[0172] As can be seen from the foregoing, for the current navigation request that does not include a note name, the map on the vehicle side will search for one or more routes according to the current navigation request and can pop up a window of the route list. At this time, the user can also confirm and select a route by voice, so as to determine the route confirmed by the user according to the voice issued by the user.

[0173] Specifically, before step 022, the voice interaction method includes: receiving a voice request from the vehicle for the user to select from the alternative routes provided for the current navigation request; and

[0174] Determining the route confirmed by the user according to the voice request.

[0175] The processor is used to receive a voice request from the vehicle for the user to select from the alternative routes provided for the current navigation request; and determine the route confirmed by the user according to the voice request.

[0176] For example, when there are Route 1, Route 2, and Route 3 in the route list respectively, the pop-up window can be as follows Figure 5As shown in the figure. The voice request for the user to select from the alternative routes provided for this navigation request can be "Select Route 2 for navigation", and then it is determined that the route confirmed by the user according to this voice request is Route 2.

[0177] In this way, the user can select a route by voice without manually clicking on the central control screen of the vehicle system, making it safer for the user to interact with the vehicle end by voice during driving.

[0178] In addition, after the user understands the note function of the navigation destination, the user can initiate a voice request to actively set a note name for the point of interest through voice to achieve collecting and noting the point of interest. When using the navigation to the note name function next time, the user can directly say "Navigate to the note name" instead of saying "Navigate to the corresponding point of interest" as in the traditional voice navigation method.

[0179] Specifically, please refer to Figure 6 , the present invention also provides a voice interaction method. The voice interaction method includes:

[0180] 031: Receive the current voice request forwarded by the vehicle, where the current voice request is a voice request for a user in the vehicle cockpit to set a note name for a point of interest in the map of the vehicle system;

[0181] 032: After completing the collection of the point of interest according to the current voice request, set a note name for the point of interest and synchronize it to the map server;

[0182] 033: Receive the next navigation request forwarded by the vehicle for the user to initiate navigation to the note name by voice;

[0183] 034: Perform intent recognition and slot recognition on the next navigation request. After confirming that the intent recognition result is a navigation intent, compare whether the point of interest in the slot recognition result exactly matches the note name;

[0184] 035: Send the matching result to the map server to complete navigation in response to the next navigation request.

[0185] The processor is used to receive the current voice request forwarded by the vehicle, where the current voice request is a voice request for a user in the vehicle cockpit to set a note name for a point of interest in the map of the vehicle system; after completing the collection of the point of interest according to the current voice request, set a note name for the point of interest and synchronize it to the map server; receive the next navigation request forwarded by the vehicle for the user to initiate navigation to the note name by voice; perform intent recognition and slot recognition on the next navigation request. After confirming that the intent recognition result is a navigation intent, compare whether the point of interest in the slot recognition result exactly matches the note name; send the matching result to the map server to complete navigation in response to the next navigation request.

[0186] First, the voice server receives the current voice request forwarded by the vehicle. The current voice request is a voice request for setting a note name for a point of interest in the map of the vehicle-mounted system by a user in the vehicle cockpit. For example, the current voice request issued by the user is "Note the location of my son's school", where the location of my son's school is the point of interest of the user in the map of the vehicle-mounted system.

[0187] Next, after the voice server completes the collection of the point of interest according to the current voice request, it sets a note name for the point of interest and synchronizes it to the map server. For example, the current voice request is "Note the location of my son's school", where the point of interest is "the location of my son's school". The voice server can complete the collection of the address of the corresponding point of interest "the location of my son's school" in the map of the vehicle-mounted system according to the current voice request, set a note name for the collected point of interest, and synchronize it to the map server. For example, the note name and the point of interest can be synchronously stored in the corresponding map server of the Amap application, so that when the user issues a navigation request to the note name next time, the navigation request can be successfully responded to complete the navigation.

[0188] Then, the voice server can receive the next navigation request forwarded by the vehicle for the user to initiate navigation to the note name by voice, perform intent recognition and slot recognition on the next navigation request, compare whether the point of interest in the slot recognition result exactly matches the note name after confirming that the intent recognition result is a navigation intent, and send the matching result to the map server to complete the navigation in response to the next navigation request. For example, the next navigation request issued by the user can be "Navigate to my son's school".

[0189] In this way, after the user knows about the note function, the present invention can also receive the voice request from the user to actively set a note name for the point of interest, collect and note the point of interest, so that when using the function of navigating to the note name next time, the user can directly say "Navigate to the note name" to navigate to the point of interest.

[0190] Before step 031, the voice interaction method includes: configuring a voice template for the user to set a note name for the point of interest, and the voice template is used to match and recognize the voice request of the user in the vehicle cockpit to set a note name for the point of interest in the map of the vehicle-mounted system.

[0191] The processor is used to configure a voice template for the user to set a note name for the point of interest, and the voice template is used to match and recognize the voice request of the user in the vehicle cockpit to set a note name for the point of interest in the map of the vehicle-mounted system.

[0192] Understandably, before the user actively sends a voice request to add a note name for a point of interest, it is necessary to configure a voice template for the user to set a note name for the point of interest on the server side, so that the voice request of the user in the vehicle cockpit to set a note name for the point of interest in the map of the vehicle-mounted system can be recognized by matching the voice template.

[0193] The voice template includes "[Put][Set][This|This one](destination|endpoint|place to go|navigation endpoint|navigation destination)(Set|Configure|Mark|Note)(as|to|called) #Note name#". For example, for this voice template, the voice requests that the user can send can be "Set the endpoint to my uncle's house" or "Note this address as the old place".

[0194] The voice template also includes "[Vehicle name][You][You help me](Remember|Write down|Take note of|Make a note of|Mark)[This|This one][You][You help me](destination|endpoint|place to go|navigation endpoint|navigation destination)(is called|is named|the note name is|the alias is|the name is|is noted as|is marked as) #Note name#". For example, for this voice template, the voice requests that the user can send can be "Xiaop, take note that this place is my gym" or "You remember that this destination is my second uncle's house".

[0195] The voice template also includes "[Put][Set] #POI (Set|Configure|Mark)(as|to)[My|My] #Note name#". For example, for this voice template, the voice requests that the user can send can be "Set No. 28 Nanjing Road as my son's school".

[0196] The voice template also includes "(Set|Note|Mark) #Note name# [of] (address|location|positioning)". For example, for this voice template, the voice requests that the user can send can be "Set the location of my son's school".

[0197] The voice template also includes "(Set|Note|Mark)[This|This one](place|address|location|positioning|here)(as|to|called) #Note name#". For example, for this voice template, the voice requests that the user can send can be "Set this place as my son's school".

[0198] In this way, by pre-configuring the voice template for the user to set a note name for the point of interest on the server side, when the user sends a corresponding voice request according to the voice template, the server can understand according to the template that the user hopes to add a note for the point of interest and the corresponding note name.

[0199] Understandably, when the server performs recognition through this voice request and sets a note name for the point of interest, it is necessary to first determine whether the map of the in-vehicle system is bound to a user account. If the map of the in-vehicle system is not bound to a user account, the vehicle end can issue a voice reminder to the user that the map of the in-vehicle system needs to be bound to a user account first before setting a note name, and can also open the map account binding page at the same time. For example, the voice reminder to the user can be "You need to bind the Amap account first before setting a note name".

[0200] Specifically, step 032 includes: when the map of the in-vehicle system is bound to a user account, performing slot recognition on this voice request to obtain a recognition result; and

[0201] After completing the collection of the point of interest according to the recognition result, setting a note name for the point of interest and synchronizing it to the map server.

[0202] The processor is used to, when the map of the in-vehicle system is bound to a user account, perform slot recognition on this voice request to obtain a recognition result; after completing the collection of the point of interest according to the recognition result, setting a note name for the point of interest and synchronizing it to the map server.

[0203] Among them, the user account includes the user's identity information. Therefore, it is possible to identify whether the current voice request of this time is a voice request sent by the corresponding user through this user account.

[0204] Understandably, only after determining that the voice request is sent by the user corresponding to this user account can slot recognition be performed on this voice request. Therefore, the point of interest corresponding to this voice request of the user can be found in the user account of the map of the corresponding in-vehicle system, and the collection of the point of interest can be completed according to the recognition result, a note name can be set for the point of interest, and it can be synchronized to the map server.

[0205] In this way, it can be ensured that the point of interest corresponding to this voice request sent by the user can be found in the user account of the map of the corresponding in-vehicle system, so as to realize setting a note name for this point of interest and synchronizing it to the map server.

[0206] More specifically, the present invention can also determine whether the voice request sent by the user has a point-of-interest slot. When the voice request sent by the user has a point-of-interest slot, the point of interest obtained by navigation search can be performed according to the point-of-interest slot, and then a note name can be set for this point of interest.

[0207] In detail, step 0322 includes: if the slot recognition result is that this voice request includes a point-of-interest slot and a note name, then for the point of interest obtained by navigation search according to the point-of-interest slot, sending a voice broadcast to the vehicle to ask the user about the expected point of interest; and

[0208] After completing the collection of the point of interest (POI) based on the POI confirmed by the user's reply, set a note name for the POI and synchronize it to the map server.

[0209] The processor is used to, if the slot recognition result is that the current voice request includes a POI slot and a note name, based on the POI obtained by performing a navigation search according to the POI slot, send a voice broadcast to the vehicle to ask the user about the expected POI; after completing the collection of the POI based on the POI confirmed by the user's reply, set a note name for the POI and synchronize it to the map server.

[0210] For example, if the current voice request issued by the user is "Set No. 28 Nanjing Road as my son's school", then the slot recognition result of the server's slot recognition of this voice request is that this voice request includes a POI slot "No. 28 Nanjing Road" and a note name "my son's school". Then, performing a navigation search according to the POI slot "No. 28 Nanjing Road" may obtain one or more POIs to form a POI list. For example, the obtained POIs include "1. North Gate of No. 28 Nanjing Road", "2. South Gate of No. 28 Nanjing Road", and "3. East Gate of No. 28 Nanjing Road".

[0211] At this time, the server can send a voice broadcast to the vehicle to ask the user about the expected POI. For example, the content of this voice broadcast is "Which one is it, please". If the user replies that it is the second one, then the server, after completing the collection of the POI based on the POI confirmed by the user's reply, sets a note name for the POI and synchronizes it to the map server. At this time, the server can also reply to the user that the note name has been set by sending another voice broadcast. At this time, the content of the voice broadcast is "Okay, No. 28 Nanjing Road South Gate has been set as my son's school. You can just say [Navigate to my son's school] next time".

[0212] In this way, the present invention can also determine whether the current voice request issued by the user has a POI slot. When the current voice request issued by the user has a POI slot, based on the POI obtained by performing a navigation search according to the POI slot, and then complete setting a note name for the POI.

[0213] Step 0322 includes: if the slot recognition result is that the current voice request includes a destination slot and a note name, based on the destination slot and the note name, complete the collection of the POI and set a note name for the POI and synchronize it to the map server.

[0214] The processor is used to, if the slot recognition result is that the current voice request includes a destination slot and a note name, based on the destination slot and the note name, complete the collection of the POI and set a note name for the POI and synchronize it to the map server.

[0215] After initiating voice navigation, the user can set a note name for the point of interest (POI) of the destination. That is to say, the user can conduct a navigation search by voice, and after selecting the destination, request to set a note name for the destination by voice.

[0216] For example, the user can search for "No. 29 Nanjing Road" through voice navigation. After the server identifies the corresponding POI, it conducts a navigation search for the POI and obtains "Building 1, No. 29 Nanjing Road", "Building 2, No. 29 Nanjing Road", and "Building 3, No. 29 Nanjing Road", and asks the user which destination they want to navigate to. After the user replies and confirms that the destination is "Building 1, No. 29 Nanjing Road", it provides route options for the user, and initiates navigation after the user selects a route. At this time, if the user's current voice request is "Set the end point as Uncle San's house", the slot recognition result of the server's slot recognition for this voice request is that this voice request includes a destination slot "end point" and a note name "Uncle San's house". Then, based on the destination slot "end point" and the note name "Uncle San's house", it completes the collection of the POI "Building 1, No. 29 Nanjing Road" corresponding to the "end point" on the map, sets the note name "Uncle San's house" for the POI, and synchronizes the POI and the note name to the map server.

[0217] In addition, after completing the collection of the POI, set a note name for the POI and synchronize it to the map server. At this time, the server can also reply to the user by sending a voice broadcast again to indicate that the note name has been set. For the user's current voice request "Set the end point as Uncle San's house", the content of the voice broadcast at this time is "Okay, the end point has been set as Uncle San's house. You can just say [Navigate to Uncle San's house] next time."

[0218] In this way, the present invention can also determine whether the current voice request sent by the user has a destination slot. When the current voice request sent by the user has a destination slot, it can complete the collection of the POI corresponding to the destination according to the destination slot and the note name, and then complete the setting of the note name for the POI.

[0219] Step 0322 includes: If the slot recognition result is that the current voice request includes an end point slot and a note name, then send a voice broadcast to the vehicle according to the end point slot to ask the user about the expected POI; and

[0220] After completing the collection of the POI confirmed by the user's reply, set a note name for the POI and synchronize it to the map server.

[0221] The processor is used to, if the slot recognition result is that the current voice request includes an end point slot and a note name, then send a voice broadcast to the vehicle according to the POI slot to ask the user about the expected POI; after completing the collection of the POI confirmed by the user's reply, set a note name for the POI and synchronize it to the map server.

[0222] It is understandable that the user can initiate navigation through a voice request and set a note name for the destination of the navigation. When the user has not selected a destination, the server can send a voice broadcast to the vehicle to ask the user about the desired point of interest. After completing the collection of the point of interest confirmed by the user's reply, a note name is set for the point of interest and synchronized to the map server. For example, after the user inputs "Navigate to Nanjing Road" by voice and has not selected the specific address of Nanjing Road as the destination, and the user's voice request for this time is "Set the destination as Uncle San's house", the slot recognition result of the server's slot recognition of this voice request is that this voice request includes a destination slot and the note name "Uncle San's house". Since "Nanjing Road" corresponding to the destination slot is not a point of interest with a specific address, the server conducts a navigation search for "Nanjing Road" to obtain the point of interest, and then sends a voice broadcast to the vehicle to ask the user about the desired point of interest. For example, the content of this voice broadcast is "Okay, please tell the specific address." At this time, the user can reply with the specific address of the point of interest corresponding to the destination, such as "No. 29 Nanjing Road". The server then conducts a navigation search for "No. 29 Nanjing Road" to obtain one or more points of interest, and can pop up a window of the point of interest list on the vehicle's central control screen, and then send a voice broadcast to the vehicle to ask the user about the desired point of interest.

[0223] For example, when there are 3 points of interest obtained from the navigation search, the point of interest list includes "1. Building 1, No. 29 Nanjing Road", "2. Building 2, No. 29 Nanjing Road", and "3. Building 3, No. 29 Nanjing Road". The server sends a voice broadcast to the vehicle to ask the user about the desired point of interest, which is "Which one do you mean?"

[0224] If the user replies that it is the second one, after the server completes the collection of the point of interest confirmed by the user's reply, a note name is set for the point of interest and synchronized to the map server. At this time, the server can also send a voice broadcast to reply to the user that the note name has been set. The content of this voice broadcast is "Okay, Building 2, No. 29 Nanjing Road has been set as Uncle San's house. Next time, you can just say [Navigate to Uncle San's house]."

[0225] In this way, the present invention can also determine whether the voice request sent by the user has a destination slot. When the voice request sent by the user has a destination slot, the point of interest confirmed by asking according to the destination slot is collected, and a note name is set for the point of interest confirmed by the user.

[0226] In addition, when this voice request does not include a point of interest slot, a destination slot, or a destination slot, the server side needs to send a voice broadcast to the vehicle side to prompt the user to input the point of interest by voice, so as to complete the collection of the point of interest and the note for the point of interest according to the point of interest input by the user by voice.

[0227] That is, step 0322 includes: if the slot recognition result indicates that the current voice request does not include a point of interest slot, a destination slot, or an end point slot, then send a voice broadcast to the vehicle to prompt the user to input the point of interest by voice; and

[0228] After completing the collection of the point of interest based on the point of interest input by the user by voice, set a note name for the point of interest and synchronize it to the map server.

[0229] The processor is used to, if the slot recognition result indicates that the current voice request does not include a point of interest slot, a destination slot, or an end point slot, then send a voice broadcast to the vehicle to prompt the user to input the point of interest by voice; after completing the collection of the point of interest based on the point of interest input by the user by voice, set a note name for the point of interest and synchronize it to the map server.

[0230] It can be understood that before the current voice request, when the user has not initiated a navigation request by voice, the current voice request does not include a destination slot or an end point slot. If the current voice request also does not include a point of interest slot, it is impossible to determine which point of interest the user wants to set a note for. At this time, it is necessary to send a voice broadcast to the vehicle to prompt the user to input the point of interest by voice, so as to assist the user in completing the collection and note setting of the point of interest through voice interaction.

[0231] For example, if the current voice request issued by the user is "Set a note as Uncle San's house", then the slot recognition result of the server's slot recognition of this voice request is that this voice request does not include a point of interest slot, a destination slot, or an end point slot. At this time, the server can send a voice broadcast to the vehicle to prompt the user to input the point of interest by voice, and the prompt content of the voice broadcast can be "Which address is to be set as Uncle San's house?". If the user replies by voice "No. 30 Nanjing Road", then the point of interest "No. 30 Nanjing Road" on the map of the in-vehicle system can be collected, and the point of interest "No. 30 Nanjing Road" can be set as a note "Uncle San's house".

[0232] In this way, when the current voice request issued by the user does not include a point of interest slot, a destination slot, or an end point slot, the server can send a voice broadcast to the vehicle to prompt the user to input the point of interest by voice, thereby guiding the user to input the point of interest by voice, and thus completing the collection and note setting of the point of interest.

[0233] More specifically, step 0322 includes: if multiple points of interest are obtained through a navigation search based on the point of interest input by the user by voice, then send a voice broadcast to the vehicle to ask the user about the expected point of interest; and

[0234] After completing the collection of the point of interest based on the point of interest confirmed by the user's reply, set a note name for the point of interest and synchronize it to the map server.

[0235] The processor is used to, if multiple points of interest are obtained through a navigation search based on the points of interest input by the user via voice, send a voice broadcast to the vehicle to ask the user about the expected point of interest; after completing the collection of the point of interest confirmed by the user's reply, set a note name for the point of interest and synchronize it to the map server.

[0236] For example, if the current voice request issued by the user is "Note as my uncle's house", the slot recognition result of the server's slot recognition for this voice request is that this voice request does not include a point of interest slot, a destination slot, or an end point slot. At this time, the server can send a voice broadcast prompting the user to input a point of interest via voice to the vehicle, and the prompt content of the voice broadcast can be "Which address is to be noted as my uncle's house?". If the user replies via voice "No. 30 Nanjing Road", the navigation search can also be performed on "No. 30 Nanjing Road". When there are multiple points of interest in the navigation search results of "No. 30 Nanjing Road", the server asks the user about the point of interest to be noted, and then sets a note name for the point of interest confirmed by the user's reply. For example, when there are 3 points of interest obtained through the navigation search, the point of interest list includes "1. Building A, No. 30 Nanjing Road", "2. Building B, No. 30 Nanjing Road", and "3. Building C, No. 30 Nanjing Road", and the server sends a voice broadcast to the vehicle asking the user about the expected point of interest as "Which one is the expected point of interest?". If the user replies that it is the second one, after the server completes the collection of the point of interest confirmed by the user's reply, sets a note name for the point of interest and synchronizes it to the map server, the server can also send a voice broadcast again to reply to the user that the note name has been set. At this time, the content of the voice broadcast is "Okay, Building B, No. 30 Nanjing Road has been set as my uncle's house. You can just say [Navigate to my uncle's house] next time."

[0237] In this way, when the current voice request does not include a point of interest slot, a destination slot, or an end point slot, and the server sends a voice broadcast prompting the user to input a point of interest via voice, when multiple points of interest are obtained through the navigation search based on the points of interest input by the user via voice, the server can trigger an inquiry about the expected point of interest from the user again, so as to determine a unique point of interest, collect the point of interest, and set a note name for the point of interest.

[0238] The above description provides various scenarios for setting a note name for a point of interest and corresponding processing methods. In addition, the user actively setting a note name for a point of interest by establishing a mapping relationship between the point of interest and the note name to complete the note will be described in detail in the following description.

[0239] Specifically, please refer to Figure 7 , the present invention provides a voice interaction method. The voice interaction method includes:

[0240] 041: Receive the voice request from the vehicle forwarded by the user to set a note for a point of interest in the in-vehicle system navigation map;

[0241] 043: Recognize the voice request to obtain the note name;

[0242] 045: Establish a mapping relationship between the note name and the point of interest;

[0243] 047: According to the mapping relationship, determine the navigation request to navigate to the note name after the user wakes up the in-vehicle system by voice next time as the navigation request for the user to navigate to the corresponding point of interest.

[0244] The processor is used to receive the voice request from the vehicle forwarded by the user to set a note for a point of interest in the in-vehicle system navigation map; recognize the voice request to obtain the note name; establish a mapping relationship between the note name and the point of interest; according to the mapping relationship, determine the navigation request to navigate to the note name after the user wakes up the in-vehicle system by voice next time as the navigation request for the user to navigate to the corresponding point of interest.

[0245] The voice request from the vehicle forwarded by the user to set a note for a point of interest in the in-vehicle system navigation map, that is, the voice request with a note name. For example, the voice request is "Set point of interest 1 as Uncle San's house", where the note name set for point of interest 1 in the in-vehicle system navigation map is "Uncle San's house", and the voice request is "Point of interest 2 is my favorite Japanese restaurant", where the note name set for point of interest 2 in the in-vehicle system navigation map is "my favorite Japanese restaurant".

[0246] The server first receives the voice request from the vehicle forwarded by the user to set a note for a point of interest in the in-vehicle system navigation map, and then recognizes the voice request to obtain the note name. Among them, the method of recognizing the voice request can be recognized by the method of Natural Language Understanding (NLU) to obtain the note name.

[0247] Then, the server can establish and store the mapping relationship between the note name and the point of interest. For example, the note name "Uncle San's house" is mapped to point of interest 1 in the navigation map, and the note name "my favorite restaurant" is mapped to point of interest 2 in the navigation map.

[0248] Finally, the server can, according to the mapping relationship, determine the navigation request to navigate to the note name after the user wakes up the in-vehicle system by voice next time as the navigation request for the user to navigate to the corresponding point of interest, so as to correctly understand the intention of the user's next voice request with a note name and directly navigate to the corresponding point of interest, improving the user experience.

[0249] For example, if the navigation request after the user wakes up the in-vehicle system by voice next time and navigates to the note name is "Navigate to my uncle's house", it can be directly determined that the navigation request for the user to navigate to the point of interest 1 corresponding to the note name "my uncle's house" based on this navigation request, so as to directly initiate navigation to the point of interest 1 and achieve the correct understanding of the intention of the user's voice request with a note.

[0250] In this way, the present invention can identify the note name by recognizing the voice request with a note, and by establishing the mapping relationship between the note name and the point of interest, determine the navigation request for the user to navigate to the note name after the user wakes up the in-vehicle system by voice next time as the navigation request for the user to navigate to the corresponding point of interest according to the mapping relationship, so as to achieve the correct understanding of the intention of the voice request with a note.

[0251] More specifically, please refer to Figure 8 , step 041 includes:

[0252] 0411: Receive the user's current navigation request forwarded by the vehicle;

[0253] 0412: Feedback the navigation search results including multiple points of interest in the in-vehicle system navigation map to the user for confirmation according to the current navigation request;

[0254] 0413: Receive the voice request from the vehicle for the user to set a note for the confirmed point of interest.

[0255] The processor is used to receive the user's current navigation request forwarded by the vehicle; feedback the navigation search results including multiple points of interest in the in-vehicle system navigation map to the user for confirmation according to the current navigation request; receive the voice request from the vehicle for the user to set a note for the confirmed point of interest.

[0256] The user's current navigation request forwarded by the vehicle, for example, the navigation request is "Go to Furongli Community", that is, the vehicle forwards the navigation request "Go to Furongli Community" to the server side.

[0257] Then, the server side feedbacks the navigation search results including multiple points of interest in the in-vehicle system navigation map to the user for confirmation according to the current navigation request. For example, 3 points of interest are obtained by performing a navigation search in the in-vehicle system navigation map according to the navigation request "Go to Furongli Community". For example, the multiple points of interest are point of interest 1, point of interest 2, and point of interest 3 respectively, and then these multiple points of interest are used as the navigation search results to be confirmed by the user.

[0258] If the user selects the confirmed point of interest as point of interest 1, it is determined that the destination for the user to use the navigation is point of interest 1.

[0259] Then, if the server receives a voice request from the vehicle forwarded by the user to set a note for the confirmed point of interest, such as "Set as my uncle's house", the server can set the note for the confirmed point of interest 1 in the navigation destination for the user as "my uncle's house". That is to say, before the server receives the voice request from the user to set a note for the point of interest, when the user has already set point of interest 1 as the navigation destination through a voice request, the voice request to set the note "my uncle's house" for the point of interest this time can directly inherit point of interest 1 that has been confirmed by voice during navigation as the point of interest to be noted, thereby establishing a mapping relationship between point of interest 1 and "my uncle's house". At this time, the content of the voice request for the user to set the note "my uncle's house" for the point of interest this time may not include point of interest 1.

[0260] In this way, the server can receive the voice request from the vehicle forwarded by the user to set a note for the point of interest in the in-vehicle system navigation map, so as to determine the point of interest in the in-vehicle system navigation map corresponding to the voice request related to the note.

[0261] For the case where the user does not set the navigation destination by voice, after step 041, the vehicle voice interaction includes: performing slot recognition on the voice request; and

[0262] In the case where the result of the slot recognition does not include the point of interest slot information, if it is determined using the vehicle's navigation application programming interface that the voice request has a destination field, then set a note for the point of interest corresponding to the destination field.

[0263] The processor is used to perform slot recognition on the voice request; in the case where the result of the slot recognition does not include the point of interest slot information, if it is determined using the vehicle's navigation application programming interface that the voice request has a destination field, then set a note for the point of interest corresponding to the destination field.

[0264] When performing slot recognition, slot recognition can be performed through a pre-trained slot extraction model to obtain slot information.

[0265] It should be noted that in order to meet the requirement of setting a note for a long address in the user's voice request. First, a sentence pattern for the voice request corresponding to the intention of setting a note can be formulated. For example, the sentence patterns for the voice request corresponding to the intention of setting a note include "Note as XXX", "Set as XXX", and "Record XXX as XXX", and a large number of voice requests containing note names are collected to train the slot extraction model, so that the voice requests containing note names can correctly recognize the intention of setting a note name.

[0266] After recognizing that the user's voice request has the intention of setting a note name, a large number of voice requests containing note names can be used. For example, the voice requests include "Set address A as my third uncle's house", "Set address B as my child's school", "Set address C as my favorite store". Then continue to train the slot extraction model so that when the slot extraction model performs slot recognition, it can accurately extract "third uncle's house", "child's school", "favorite store" in the voice request, etc., and recognize the corresponding slot name as "cust_name", and the slot values corresponding to the note names are "third uncle's house", "child's school", "favorite store", that is, the corresponding note name slot information can be accurately recognized. At the same time, when the slot extraction model performs slot recognition, it can accurately extract "address A", "address B", "address C" in the voice request, etc., and recognize the corresponding slot name as "poi_name", and the slot values corresponding to the points of interest are "address A", "address B", "address C", that is, the corresponding point of interest slot information can be accurately recognized.

[0267] The slot recognition result can include the point of interest slot information or may not include the point of interest slot information.

[0268] It can be understood that after the server receives the voice request forwarded by the vehicle, it can first perform slot recognition on the voice request. For example, after the server receives the voice request "Go to my third uncle's house" forwarded by the vehicle, the slot name obtained by slot recognition in the slot recognition result is "cust_name", that is, the note slot, and the slot value is "third uncle's house". At this time, the slot recognition result does not include the point of interest slot information. In this case, the server can call the navigation application programming interface of the vehicle to determine whether there is a destination field in the voice request to determine which point of interest the user wants to set a note for. If it is determined that there is a destination field in the voice request by using the navigation application programming interface of the vehicle, then set a note for the point of interest corresponding to the destination field.

[0269] It can be understood that by calling the navigation application programming interface of the vehicle, it can be determined whether there is a destination field in the current voice request for the user to interact with the vehicle. The destination field is the field indicating going to a certain destination, so as to determine the point of interest corresponding to the destination field of the current user's voice request and set a note for the point of interest.

[0270] For example, the user's voice request is "Set the destination to the company". The result of slot recognition for this voice request shows that the slot name is "cust_name" and the slot value is "company", and the point of interest slot information is not included in the recognition result. At this time, the vehicle's navigation application programming interface can be called to determine whether the current voice request contains a destination field. The result is that the voice request has a destination field, and the destination field is "No. 2 Zhongguancun". Then, a note can be set for the point of interest corresponding to the destination field "No. 2 Zhongguancun" so that when the user enters the voice request "Depart for No. 2 Zhongguancun" again next time, the vehicle can directly navigate to the point of interest corresponding to the destination field "No. 2 Zhongguancun".

[0271] In this way, by calling the application programming interface of the map application on the vehicle's in-vehicle system, the present invention can determine whether the voice request for user-vehicle interaction contains a destination field, thereby determining the point of interest corresponding to the destination field and setting a note for this point of interest, which facilitates directly navigating to the point of interest corresponding to the destination field for the voice request of a user with the same destination field next time.

[0272] It can be understood that when it is determined by using the vehicle's navigation application programming interface that the voice request has a destination field, there may be multiple destination fields. In this case, it is necessary to actively ask and guide the user to confirm a single destination field and set a note for the point of interest corresponding to the confirmed single destination field.

[0273] Specifically, step 0422 includes: if it is determined by using the vehicle's navigation application programming interface that the voice request has multiple destination fields, then asking the user about the desired destination field through voice broadcast; and

[0274] Setting a note for the point of interest corresponding to the destination field confirmed by the user's reply.

[0275] The processor is used to, if it is determined by using the vehicle's navigation application programming interface that the voice request has multiple destination fields, ask the user about the desired destination field through voice broadcast, and is also used to set a note for the point of interest corresponding to the destination field confirmed by the user's reply.

[0276] For example, if the user's voice request is "Set the destination to the company", by calling the vehicle's navigation application programming interface, it is determined that the voice request for the user-vehicle interaction contains destination fields including Address 1: Zhongguancun No. 1, Address 2: Zhongguancun No. 2, and Address 3: Zhongguancun No. 3. At this time, on the vehicle's central control screen, a pop-up window can sequentially display the search lists of the above three destination fields, and at the same time, the vehicle can use voice broadcast to ask the user about the expected destination field. For example, the voice content of the question is "Which one is it?". If the user replies and confirms that the destination field is the second one, it means that the destination field is "Zhongguancun No. 2", and thus the point of interest corresponding to the destination field "Zhongguancun No. 2" is noted as the company.

[0277] In this way, in the case where there are multiple destination fields in the voice request, a unique destination field can be determined.

[0278] It can be understood that when using the vehicle's navigation application programming interface to determine that there is a unique destination field in the voice request, the point of interest corresponding to the unique destination field can be directly set with a note.

[0279] In addition, step 0422 further includes: if it is determined by using the vehicle's navigation application programming interface that there is one destination field in the voice request, a note is set for the point of interest corresponding to the destination field.

[0280] The processor is used to set a note for the point of interest corresponding to the destination field if it is determined by using the vehicle's navigation application programming interface that there is one destination field in the voice request.

[0281] For example, the user's voice request is "Set the destination to the company". By calling the vehicle's navigation application programming interface, it is determined that there is one destination field in the current voice request, and the destination field is "Zhongguancun No. 3". Then, a note can be directly set for the point of interest corresponding to the destination field "Zhongguancun No. 3" as the company, so that when the user inputs the voice request "Go to the company" again next time, the vehicle can directly navigate to the point of interest corresponding to the destination field "Zhongguancun No. 3".

[0282] In addition, in the case where the result of the destination field of the voice request is not found by using the vehicle's navigation application programming interface, the present invention can feedback to the user that no point of interest is found.

[0283] Specifically, the voice interaction method includes: if it is determined by using the vehicle's navigation application programming interface that the result of the destination field of the voice request is not found, feedback that no point of interest is found through voice broadcast.

[0284] The processor is used to feedback that no point of interest is found through voice broadcast if it is determined by using the vehicle's navigation application programming interface that the result of the destination field of the voice request is not found.

[0285] Understandably, when there is no destination planned in the user's voice request and the destination field of the voice request is determined using the vehicle's navigation application programming interface, it is determined that there is no corresponding destination field result.

[0286] At this time, the in-vehicle system needs to perform a voice broadcast prompt to guide the user to say it in another way. For example, the feedback content of the voice broadcast is "The location you mentioned was not found. Please try another way of saying it."

[0287] In this way, when it is determined that there is no result for the destination field of the voice request using the vehicle's navigation application programming interface, by providing a voice broadcast feedback indicating that the point of interest was not found, the user can be promptly informed and can adjust the statement in a timely manner until the destination field is determined and the corresponding point of interest for the destination field is determined.

[0288] In addition, as previously introduced, in the case where the result of slot recognition does not include point of interest slot information, the vehicle's navigation application programming interface can be called to determine whether the voice request contains a destination field, thereby determining the point of interest corresponding to the destination field and setting a note for the point of interest.

[0289] The following describes how to set a note for the point of interest based on the point of interest slot information in the case where the result of slot recognition includes point of interest slot information. That is, for

[0290] Specifically, after step 041, the voice interaction method includes: performing slot recognition on the voice request; and

[0291] In the case where the result of slot recognition includes point of interest slot information, setting a note for the point of interest obtained by navigating and searching the point of interest slot information.

[0292] The processor is used to perform slot recognition on the voice request; in the case where the result of slot recognition includes point of interest slot information, setting a note for the point of interest obtained by navigating and searching the point of interest slot information.

[0293] At this time, the voice request is a voice type that includes both note name slot information and point of interest slot information. For example, the voice request is "Set Zhongguancun as the place where Dad goes to work". Among them, the result obtained by slot recognition includes a slot name of "poi_name", the slot value corresponding to the point of interest is "Zhongguancun", the slot name is "cust_name", and the slot value corresponding to the note name is "the place where Dad goes to work", that is, the point of interest slot information and the note name slot information are recognized. Therefore, the point of interest obtained by navigating and searching the point of interest slot information can be set with a note. Specifically, navigate and search for the point of interest corresponding to the slot value "Zhongguancun" in the slot information, and then set a note for the corresponding point of interest.

[0294] Thus, in the case where the voice request includes the memo name information and the point of interest information, directly perform a navigation search on the point of interest slot information in the slot recognition result to obtain the point of interest, and set a memo for the point of interest.

[0295] Step 0424 includes: if multiple points of interest are obtained through the navigation search of the point of interest slot information, then ask the user about the expected point of interest through voice broadcast; and

[0296] Set a memo for the point of interest confirmed by the user's reply.

[0297] The processor is used to, if multiple points of interest are obtained through the navigation search of the point of interest slot information, then ask the user about the expected point of interest through voice broadcast; set a memo for the point of interest confirmed by the user's reply.

[0298] If multiple points of interest are obtained through the navigation search of the point of interest slot information, at this time, it is also possible to ask the user about the expected point of interest through voice broadcast, and set a memo for the point of interest confirmed by the user's reply.

[0299] For example, the user's voice request is "Set Zhongguancun as the place where Dad goes to work". Since the address range referred to by Zhongguancun in the navigation map is relatively large, multiple points of interest obtained through the navigation search of the point of interest slot information are: Zhongguancun No. 4, Zhongguancun No. 5, and Zhongguancun No. 6. Then, at this time, a pop-up window on the vehicle's central control screen can sequentially display the above three point of interest search lists, and at the same time, the vehicle can ask the user about the expected point of interest through voice broadcast. For example, the voice content of the question is "Which one is it?". If the user replies and confirms that the point of interest is the second one, that is, the point of interest is "Zhongguancun No. 5", then the point of interest "Zhongguancun No. 5" can be noted as the company.

[0300] Thus, in the case where multiple points of interest are obtained through the navigation search of the point of interest slot information, ask the user about the expected point of interest through voice broadcast, and set a memo for the point of interest confirmed by the user's reply.

[0301] In addition, step 0424 includes: if one point of interest is obtained through the navigation search of the point of interest slot information, then set a memo for the point of interest.

[0302] The processor is used to, if one point of interest is obtained through the navigation search of the point of interest slot information, then set a memo for the point of interest.

[0303] It can be understood that if a unique point of interest is obtained through the navigation search of the point of interest slot information, then a memo can be directly set for the point of interest.

[0304] For example, if the user's voice request is "Set No. 7 in Zhongguancun as the place where Dad goes to work", then according to the navigation search of the point of interest slot information, the only point of interest obtained is "No. 7 in Zhongguancun". Therefore, the point of interest "No. 7 in Zhongguancun" can be directly noted as the company.

[0305] In this way, when the navigation search of the point of interest slot information obtains a single point of interest, a note is directly set for this point of interest.

[0306] In addition, step 0424 further includes: if there is no search result in the navigation search of the point of interest slot information, a voice broadcast is used to feedback that the point of interest is not found.

[0307] The processor is used to, if there is no search result in the navigation search of the point of interest slot information, feedback that the point of interest is not found through voice broadcast.

[0308] It can be understood that when the point of interest is clearly indicated in the user's voice request, when using the vehicle's navigation application programming interface to determine the point of interest of the voice request, it is determined that there is no corresponding search result.

[0309] At this time, the in-vehicle system needs to perform a voice broadcast prompt to guide the user to say it in another way. For example, the feedback content of the voice broadcast is "The location you mentioned was not found. Please try another way of saying it".

[0310] In this way, when there is no search result in the navigation search of the point of interest slot information, it is feedback that the point of interest is not found through voice broadcast, which can promptly feedback to the user and promptly adjust the statement until the point of interest is found.

[0311] In the case where the function of navigating to the note name of the point of interest is already known, and the point of interest has been collected and noted, how the server specifically performs intent and slot recognition according to the user's voice request for navigating to the note name of the point of interest, and issues natural language understanding to the vehicle to complete navigation will be described in detail in the following description.

[0312] Please refer to Figure 9 , the present invention provides a voice interaction method. The voice interaction method includes:

[0313] 051: Receive the user's voice request for navigating to the note name of the point of interest forwarded by the vehicle;

[0314] 052: Perform domain classification on the voice request;

[0315] 053: After determining that the classification result of the domain classification is the navigation domain, process the voice request according to the preset sentence pattern set;

[0316] 054: Parse out the note name of the point of interest according to the processing result and obtain the intent of navigating to the note name of the point of interest;

[0317] 055: Match the parsing result with the preset key-value pairs in the navigation field, where the preset key-value pairs include the unique key name composed of the user identifier and the field corresponding to the point of interest (POI) note name;

[0318] 056: According to the intention of navigating to the POI note name, perform slot extraction after the parsing result matches the corresponding field of the preset key-value pair to obtain the first slot extraction result;

[0319] 057: Combine the first slot extraction result and output the first natural language understanding result to the map of the vehicle-mounted system, and the map completes the navigation in response to the voice request according to the first natural language understanding result.

[0320] The present invention also provides a server. The server includes a processor and a memory, and a computer program is stored on the memory. The processor is used to receive the voice request of the user navigating to the POI note name forwarded by the vehicle; perform domain classification on the voice request; after determining that the classification result of the domain classification is the navigation domain, process the voice request according to the preset sentence pattern set; parse out the POI note name according to the processing result and obtain the intention of navigating to the POI note name; match the parsing result with the preset key-value pairs in the navigation field, where the preset key-value pairs include the unique key name composed of the user identifier and the field corresponding to the POI note name; according to the intention of navigating to the POI note name, perform slot extraction after the parsing result matches the corresponding field of the preset key-value pair to obtain the first slot extraction result; combine the first slot extraction result and output the first natural language understanding result to the map of the vehicle-mounted system, and the map completes the navigation in response to the voice request according to the first natural language understanding result.

[0321] First, receive the voice request of the user navigating to the POI note name forwarded by the vehicle. For example, if the user has previously noted the POI "Zhongguancun" as the note name "Company", then the voice request of the user navigating to the POI note name can be "Navigate to the company", where "company" is the POI note name.

[0322] Then, perform domain classification on the voice request. It can be understood that since the voice requests issued by the user can be divided into different domains, such as including the music domain, the weather domain, the vehicle control domain, and the navigation domain, etc.

[0323] Next, after determining that the classification result of the domain classification is the navigation domain, the voice request is processed according to a preset sentence pattern set. It can be understood that since the present invention aims to correctly understand the voice request for the user to navigate to the POI (Point of Interest) note name, therefore, after determining that the classification result of the domain classification of the voice request is the navigation domain, the voice request is processed according to the preset sentence pattern set to obtain whether the intent of the user's navigation is to navigate to the POI note name, that is, whether it is the intent of "navigation_cust" to navigate to the POI note name. The processing method is to match the voice request with the preset sentence pattern set to determine whether the sentence pattern of the voice request conforms to a certain sentence pattern in the preset sentence pattern set. That is to say, if it is determined that the domain of the user's voice request is not the navigation domain but other domains such as the voice domain, the voice request sent by the user is not processed according to the preset sentence pattern combination, and no subsequent related operations are required.

[0324] Among them, the format of the preset sentence pattern set is "sentence pattern: intent", including sentence patterns such as "navigate to XXX", "navigate to my XXX", and "go to XXX". Among them, "XXX" can include the POI note name or a specific place name. The above sentence patterns can initially determine that the user's intent is the intent to navigate to the POI note name. For example, for the sentence pattern "navigate to my company", the corresponding intent is initially determined to be the intent to navigate to the POI note name.

[0325] In addition, the vehicle sometimes cannot correctly recognize the sentence pattern of "navigate to the POI note name" and cannot meet the user's need to navigate to the POI note name. Therefore, the present invention pre-stores a preset sentence pattern set on the server side. When the sentence pattern of the voice request sent by the user matches a certain sentence pattern in the preset sentence pattern set, it is determined that the intent of the current user's voice request is to obtain the intent to navigate to the POI note name. Among them, the parsing method can be sentence pattern matching or regular expression matching, etc.

[0326] Of course, parsing the corresponding POI note name and obtaining the intent to navigate to the POI note name can also be achieved by training a corresponding machine learning model.

[0327] Then, it is necessary to verify the parsing result to verify whether the parsing result is a pre-stored POI note name. By matching the parsing result with the preset key-value pairs in the navigation domain, if the parsing result matches the corresponding field of the preset key-value pair, it indicates that the verification is passed, and the voice request contains the POI note name, that is, the intent of the voice request to navigate to the POI note name is correct.

[0328] Among them, the preset key-value pairs in the navigation field are the key-value pairs pre-stored on the server side. The preset key-value pairs include the unique key name composed of user identifiers and the fields corresponding to the POI (Point of Interest) note names. For example, the preset key-value pairs in the navigation field are "Key: navigation_id, Value: [Uncle San's house, Grandma's house, My favorite place, Child's school]". Here, Key represents the unique key name composed of user identifiers, which is used to identify the user's identity; Value represents the fields corresponding to the POI note names, including all the POI note names that the user has historically noted. The above POI note names, such as "Uncle San's house", "Grandma's house", "My favorite place", and "Child's school", are all the note names that the user has noted for the POIs in the map of the in-vehicle system.

[0329] Afterwards, after verifying that the user's voice request for navigating to the POI note name is correct, the voice request can be subjected to slot extraction according to the intention of navigating to the POI note name to obtain the first slot extraction result. For example, the user's voice request is "Navigate to Uncle San's house". After verifying that the intention of this voice request is to navigate to the POI note name, slot extraction is performed on this voice request, and the first slot extraction result includes the slot name "cust_name" and the corresponding slot value "Uncle San's house".

[0330] Finally, the first natural language understanding result is output to the map of the in-vehicle system in combination with the first slot extraction result, and the map completes the navigation in response to the voice request according to the first natural language understanding result. For example, the server side performs natural language understanding on the voice request "Navigate to Uncle San's house" in combination with the above first slot extraction result "Uncle San's house", and outputs the first natural language understanding result as: "Navigate to Uncle San's house", domain: navigation, intent: navigation_cust, slot: {"cust_name": "Uncle San's house"}. Then, after the map end of the in-vehicle system receives this first natural language understanding result, it requests the POIs pre-stored in the map according to the POI note name in the first natural language understanding result, extracts the correct POIs according to the POI note name, and thus performs navigation.

[0331] In this way, the voice interaction method of the present invention can accurately identify the voice request containing the POI note name for the voice request containing the POI note name, and meet the navigation requirements of the user who wants to navigate to the POI with the note name.

[0332] Understandably, when the user's voice request meets a certain sentence pattern in the preset sentence pattern set but does not include the point of interest (POI) note name, and the user has not modified the POI note name in the preset key-value pair, since the vehicle cannot distinguish whether this voice request is a voice request with a POI note name. At this time, if the vehicle continues to recognize the intention of this voice request as the intention to navigate to the POI note name, since the voice request does not include the POI note name, the parsing result cannot be matched with the preset key-value pair to obtain a corresponding result, nor can the voice request be slot-extracted according to the intention to navigate to the POI note name, thus the voice interaction process cannot be successfully completed.

[0333] In view of this, please refer to Figure 10 , after step 055, the voice interaction method includes:

[0334] 0551: When the field corresponding to the POI note name in the preset key-value pair matched with the parsing result is empty, rewrite the intention to navigate to the POI note name as the intention to navigate to the POI;

[0335] 0552: According to the intention to navigate to the POI, perform slot extraction on the voice request to obtain a second slot extraction result;

[0336] 0553: Combine the second slot extraction result and output a second natural language understanding result to the map, and the map completes the navigation in response to the voice request according to the second natural language understanding result.

[0337] The processor is used to, when the field corresponding to the POI note name in the preset key-value pair matched with the parsing result is empty, rewrite the intention to navigate to the POI note name as the intention to navigate to the POI; according to the intention to navigate to the POI, perform slot extraction on the voice request to obtain a second slot extraction result; combine the second slot extraction result and output a second natural language understanding result to the map, and the map completes the navigation in response to the voice request according to the second natural language understanding result.

[0338] After matching the parsing result with the preset key-value pairs in the navigation domain, when the user has not modified the poi note name in the preset key-value pairs, and the field corresponding to the poi note name in the preset key-value pairs matched with the parsing result is empty, it means that there is no poi note name in the parsing result that matches the preset key-value pairs. At this time, the intention of navigating to the poi note name is rewritten as the intention of navigating to the poi (navigation_poi). For example, if the user's voice request is "Navigate to Zhongguancun", the parsing result is "Zhongguancun", and the field Value corresponding to the poi note name in the preset key-value pairs is empty, then at this time, the intention of navigating to the poi note name can be directly rewritten as the intention of navigating to the poi "Zhongguancun". In this way, it is possible to distinguish whether the voice request sent by the user is a voice request containing a poi note name.

[0339] Then, according to the intention of navigating to the poi, slot extraction is performed on the voice request to obtain the second slot extraction result. That is, after the intention is rewritten as navigating to the poi, slot extraction is performed on the voice request "Navigate to Zhongguancun" in the above example to obtain the second slot extraction result "Zhongguancun".

[0340] After that, the second natural language understanding result is output to the map in combination with the second slot extraction result, and the map completes the navigation in response to the voice request according to the second natural language understanding result. For example, the server combines the slot name "poi_name" and the corresponding slot value "Zhongguancun" included in the above second slot extraction result to perform natural language understanding on the voice request "Navigate to Zhongguancun", and outputs the second natural language understanding result as: "Navigate to Zhongguancun", domain: navigation, intent: navigation_poi, slot: {"poi_name": "Zhongguancun"}, and then after the map end of the in-vehicle system receives this second natural language understanding result, it navigates according to the poi in this first natural language understanding result.

[0341] In this way, when the user's voice request meets a certain sentence pattern in the preset sentence pattern set but does not include a poi note name, and the field corresponding to the poi note name in the preset key-value pairs matched with the parsing result is empty, through the intention rewriting and slot extraction of this voice request, it is also possible to achieve correct intention recognition and slot extraction of this voice request, and then complete the voice interaction.

[0342] Understandably, when the user's voice request meets a certain sentence pattern in the preset sentence pattern set and includes a point of interest (POI) note name, but the user has modified the POI note name in the preset key-value pair, since the POI note name in the preset key-value pair has been modified, therefore, if the POI note name in the parsing result is the POI note name before modification in the preset key-value pair, the POI note name expected by the user cannot be matched. At this time, if the vehicle continues to recognize the intent of this voice request as the intent to navigate to the POI note name, the parsing result cannot be matched with the preset key-value pair to obtain the corresponding result, and the slot extraction cannot be performed on this voice request according to the intent to navigate to the POI note name, thus the voice interaction process cannot be successfully completed.

[0343] In view of this, please refer to Figure 11 , after step 055, the voice interaction method includes:

[0344] 0554: When the parsing result and the preset key-value pair are matched and no corresponding field of the preset key-value pair is matched, rewrite the intent to navigate to the POI note name as the intent to navigate to the POI;

[0345] 0555: According to the intent to navigate to the POI, perform slot extraction on the voice request to obtain a second slot extraction result;

[0346] 0556: Combine the second slot extraction result and output a second natural language understanding result to the map, and the map completes the navigation in response to the voice request according to the second natural language understanding result.

[0347] The processor is used to rewrite the intent to navigate to the POI note name as the intent to navigate to the POI when the parsing result and the preset key-value pair are matched and no corresponding field of the preset key-value pair is matched; perform slot extraction on the voice request according to the intent to navigate to the POI to obtain a second slot extraction result; combine the second slot extraction result and output a second natural language understanding result to the map, and the map completes the navigation in response to the voice request according to the second natural language understanding result.

[0348] After matching the parsing result with the preset key-value pairs in the navigation domain, when the user has modified the POI (Point of Interest) note name in the preset key-value pairs and the result of matching the parsing result with the preset key-value pairs is that the corresponding field of the preset key-value pair is matched, the intention of navigating to the POI note name is rewritten as the intention of navigating to the POI. For example, the voice request is "Navigate to my uncle's house", the parsed POI note name is "my uncle's house", and the corresponding field of the POI note name in the preset key-value pair is "Value: [Company, Grandma's house, My favorite place, Child's school]". Since the POI note name in the parsing result does not match the corresponding field of the POI note name before the modification of the preset key-value pair, at this time, the intention of navigating to the POI note name is rewritten as the intention of navigating to the POI.

[0349] Then, according to the intention of navigating to the POI, slot extraction is performed on the voice request to obtain the second slot extraction result. That is, after the intention is rewritten as navigating to the POI, slot extraction is performed on the voice request "Navigate to my uncle's house" in the above example, and the second slot extraction result includes the slot name "poi_name" and the corresponding slot value "my uncle's house".

[0350] After that, the second natural language understanding result is output to the map in combination with the second slot extraction result, and the map completes the navigation in response to the voice request according to the second natural language understanding result. For example, the server combines the slot name "poi_name" and the corresponding slot value "my uncle's house" included in the above second slot extraction result to perform natural language understanding on the voice request "Navigate to grandma's house", and the output second natural language understanding result is: "Navigate to my uncle's house", domain: navigation, intent: navigation_poi, slot: {"poi_name": "my uncle's house"}. Then, after the map end of the in-vehicle system receives this second natural language understanding result, it performs navigation according to the POI in this first natural language understanding result.

[0351] In this way, when the user's voice request meets a certain sentence pattern in the preset sentence pattern set and includes a POI note name, if the POI note name in the parsing result is the POI note name before the modification in the preset key-value pair, the POI note name expected by the user cannot be matched, that is, the parsing result and the preset key-value pair do not match the corresponding field of the preset key-value pair. Through the intention rewriting and slot extraction of this voice request, correct intention recognition and slot extraction of this voice request can also be achieved, and then the voice interaction can be completed.

[0352] In addition, for each user, different note names actually correspond to different point-of-interest addresses or destination addresses. For example, for user Zhang San, the "best restaurant" is "XXXX Barbecue", while for user Li Si, the "best restaurant" is "XXXX Japanese Cuisine". It is impossible to correctly identify the navigation needs of the user to fulfill the navigation requirement that the user wants to navigate to the place with the note name. Therefore, different user identifiers are included in the preset key-value pairs of the present invention to identify different users, thereby distinguishing voice requests sent by different users and meeting the navigation needs of different users.

[0353] Specifically, please refer to Figure 12 , step 055 includes:

[0354] 05501: Obtain the current user identifier corresponding to the voice request;

[0355] 05502: In the preset key-value pairs in the navigation field, determine the target key-value pair whose unique key name corresponds to the current user identifier;

[0356] 05503: Match the point-of-interest note name in the parsing result with the field corresponding to the point-of-interest note name in the target key-value pair.

[0357] The processor is used to obtain the current user identifier corresponding to the voice request; in the preset key-value pairs in the navigation field, determine the target key-value pair whose unique key name corresponds to the current user identifier; match the point-of-interest note name in the parsing result with the field corresponding to the point-of-interest note name in the target key-value pair.

[0358] When matching the parsing result with the preset key-value pairs in the navigation field, the current user identifier corresponding to the voice request can be obtained first. For example, the current user identifier can be identified according to the user's voice tone, that is, the navigation_id1 of the current user is obtained.

[0359] Then, in the preset key-value pairs in the navigation field, determine the target key-value pair whose unique key name corresponds to the current user identifier. For example, the preset key-value pairs in the navigation field pre-stored on the server side include: "Key: navigation_id1, Value: [company, home, my favorite restaurant, child's school]" and "Key: navigation_id2, Value: [uncle's home, grandma's home, my favorite place, child's school]", then the target key-value pair corresponding to the current user identifier navigation_id1 can be determined as: "Key: navigation_id1, Value: [company, home, my favorite restaurant, child's school]".

[0360] Therefore, at this time, the POI note name in the parsing result can be matched with the field corresponding to the POI note name in the target key-value pair. For example, if the current user's voice request is "Navigate to my favorite restaurant", then the POI note name "my favorite restaurant" in the parsing result is matched with the field corresponding to the POI note name in the target key-value pair "Value: [company, home, my favorite restaurant, children's school]".

[0361] In this way, the preset key-value pair of the present invention includes different user identifiers for identifying different users. By obtaining the current user identifier corresponding to the voice request, in the preset key-value pair in the navigation field, the target key-value pair corresponding to the unique key name and the current user identifier is determined, and the POI note name in the parsing result is matched with the field corresponding to the POI note name in the target key-value pair, so that the voice requests issued by different users can be distinguished, and the different navigation needs of different users can be met.

[0362] In addition, since users have different note name requirements at different times, users will need to update the preset key-value pairs pre-stored on the server side to meet the navigation needs of users at different times. The following introduces the solution for updating the preset key-value pairs pre-stored on the server side of the present invention.

[0363] Specifically, when adding a note name to the preset key-value pair on the server side to update the preset key-value pair, the voice interaction method includes:

[0364] 0581: Receive the note name information of the user adding a note name forwarded by the vehicle;

[0365] 0582: Update the preset key-value pair according to the note name information.

[0366] The processor is used to receive the note name information of the user adding a note name forwarded by the vehicle; update the preset key-value pair according to the note name information.

[0367] When adding a note name to the preset key-value pair on the server side, the user can add a note name on the map end of the vehicle. At this time, the server side can call the application programming interface to synchronously receive the note name information of the user adding a note name forwarded by the vehicle, and update the preset key-value pair according to the note name information.

[0368] For example, the user can issue a voice request "Set POI 1 as my mother's home" on the map end of the vehicle to add a note name "my mother's home". At this time, the server side calls the application programming interface to synchronously receive the added note name "my mother's home", so as to add the note name "my mother's home" to the field corresponding to the POI note name in the preset key-value pair.

[0369] Thus, when it is necessary to add a note name to the preset key-value pairs on the server side, the server side of the present invention can synchronously receive the note name information of the user adding the note name forwarded by the vehicle by calling the application program interface, and update the preset key-value pairs according to the note name information.

[0370] More specifically, step 0582 includes: determining the current user identifier corresponding to the note name information; and

[0371] Adding the note name of the note name information to the field corresponding to the point of interest note name in the preset key-value pair whose unique key name matches the current user identifier.

[0372] The processor is used to determine the current user identifier corresponding to the note name information; and add the note name of the note name information to the field corresponding to the point of interest note name in the preset key-value pair whose unique key name matches the current user identifier.

[0373] When adding a note name to the preset key-value pair according to the note name information, the current user identifier corresponding to the note name information can be determined first, and the preset key-value pair whose unique key name matches the current user identifier can be determined according to the current user identifier, so as to add the note name of the note name information to the field corresponding to the point of interest note name in the matching preset key-value pair, thereby distinguishing the voice requests for adding note names sent by different users and meeting the adding requirements of different users.

[0374] For example, if the current user identifier corresponding to the note name information is determined to be "navigation_id2", the preset key-value pair whose unique key name matches the current user identifier is "Key: navigation_id2, Value: [My uncle's house, Grandma's house, My favorite place, The child's school]". Therefore, the note name of the note name information can be accurately added to the field corresponding to the point of interest note name in the preset key-value pair. For example, if the note name of the note name information is "The hospital closest to my home", the preset key-value pair after adding the note name becomes "Key: navigation_id2, Value: [My uncle's house, Grandma's house, My favorite place, The child's school, The hospital closest to my home]", thus completing the process of adding the note name to the voice request for adding the note name sent by this user in the preset key-value pair.

[0375] Thus, when updating the preset key-value pair according to the note name information, the current user identifier corresponding to the note name information can be determined first, and the note name of the note name information can be added to the field corresponding to the point of interest note name in the preset key-value pair whose unique key name matches the current user identifier, so as to meet the requirement of adding the note name to the voice requests for adding note names sent by different users in the preset key-value pair.

[0376] In addition, when it is necessary to modify the note name in the preset key-value pair on the server side to update the preset key-value pair, the voice interaction method includes:

[0377] Obtaining the note name information after the user modifies the note name in the vehicle by means of polling; and

[0378] Updating the preset key-value pair according to the note name information.

[0379] The processor is used to obtain the note name information after the user modifies the note name in the vehicle by means of polling; and update the preset key-value pair according to the note name information.

[0380] Polling is a way for a processor (CPU) to decide how to provide services to peripheral devices, also known as "Programmed I / O". The concept of the polling method is that the CPU sends out inquiries at regular intervals, sequentially asking each peripheral device whether it needs its services. If so, it provides the service. After the service is completed, it asks the next peripheral, and then repeats this process continuously.

[0381] That is to say, when it is necessary to modify the note name in the preset key-value pair on the server side, the server side can obtain the note name information after the user modifies the note name in the vehicle by means of polling. The server side can send an inquiry to the vehicle side at regular intervals to ask whether it is necessary to modify the note name. When the server side receives the service that the user on the vehicle side needs to modify the note name, it obtains the note name information after the user modifies the note name in the vehicle, and modifies the preset key-value pair according to the note name information to update the preset key-value pair.

[0382] For example, the server side sends an inquiry to the vehicle side every 1 minute to ask whether there is a modification to the note name. When the user modifies the note name, the server side can obtain the note name information after the user modifies the note name on the vehicle side. For example, the note name information after the modification is to change "my uncle's house" to "my aunt's house". Thus, the "my uncle's house" in the field corresponding to the point of interest note name in the preset key-value pair can be changed to "my aunt's house" to complete the modification of the preset key-value pair.

[0383] In this way, when it is necessary to modify the note name in the preset key-value pair on the server side, the server side of the present invention can obtain the note name information after the user modifies the note name in the vehicle by means of polling, and update the preset key-value pair according to the note name information.

[0384] Step 0592 includes: determining the current user identifier corresponding to the note name information; and

[0385] Modifying the field corresponding to the point of interest note name in the preset key-value pair whose unique key name for note name modification matches the current user identifier according to the note name information.

[0386] The processor is used to determine the current user identifier corresponding to the note name information; and modify the field corresponding to the point of interest note name in the preset key-value pair whose unique key name matches the current user identifier according to the note name of the note name information.

[0387] When modifying the preset key-value pair according to the note name information, the current user identifier corresponding to the note name information can be determined first, and the preset key-value pair whose unique key name matches the current user identifier can be determined according to the current user identifier, so as to modify the field corresponding to the point of interest note name in the matching preset key-value pair according to the note name of the note name information, and further distinguish the voice requests for modifying the note name sent by different users and meet the modification requirements of different users.

[0388] For example, if the current user identifier corresponding to the note name information is determined to be "navigation_id2", the preset key-value pair whose unique key name matches the current user identifier is "Key: navigation_id2, Value: [Uncle San's house, Grandma's house, My favorite place, Child's school]". Therefore, the field corresponding to the point of interest note name in the matching preset key-value pair can be modified according to the note name information. For example, if the user modifies the note name and changes "Uncle San's house" to "Aunt's house", the preset key-value pair after adding this note name becomes "Key: navigation_id2, Value: [Aunt's house, Grandma's house, My favorite place, Child's school]", thus completing the process of modifying the note name in the preset key-value pair for the voice request of modifying the note name sent by this user.

[0389] In this way, when updating the preset key-value pair according to the note name information, the current user identifier corresponding to the note name information can be determined first, and the field corresponding to the point of interest note name in the preset key-value pair whose unique key name matches the current user identifier can be modified according to the note name of the note name information, so as to meet the voice requests for modifying the note name sent by different users and the requirements for modifying the note name in the preset key-value pair.

[0390] The present invention also provides a non-volatile computer-readable storage medium containing a computer program. When the computer program is executed by one or more processors, the voice interaction method described in any of the above embodiments is implemented.

[0391] It can be understood that a computer program includes computer program code. The computer program code can be in the form of source code, object code, executable files, or some intermediate forms, etc. The computer-readable storage medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc.

Claims

1. A guiding method for adding a note name to a point of interest in a vehicle system, characterized in that, it includes: Receiving first reporting information reported by the vehicle that a user in the vehicle initiates navigation to a point of interest through the vehicle system; Querying the collection and note information of all users for the address of the point of interest in the note information set according to the first reporting information; Determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule; Sending the first voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the vehicle system; The determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule includes: In the case where the user has collected the point of interest and has not added a note name to the point of interest, if the user has performed an operation of initiating navigation to the point of interest within a first preset time, then according to the query result and the type of the point of interest, determining the first voice broadcast content to be fed back according to the predetermined rule.

2. The guiding method according to claim 1, characterized in that, After the step of receiving the first reporting information reported by the vehicle that a user in the vehicle initiates navigation to a point of interest through the vehicle system, the guiding method includes: Processing the name of the point of interest, the search time and search times of the user for the point of interest to obtain the type of the point of interest for which a recommended note is to be given.

3. The guiding method according to claim 1, characterized in that, The determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule includes: Selecting multiple candidate note names according to the predetermined rule according to the note information in the query result that is of the same type as the point of interest to determine the first voice broadcast content to be fed back.

4. The guiding method according to claim 1, characterized in that, The determining the first voice broadcast content to be fed back according to the query result and the type of the point of interest according to a predetermined rule includes: In the case where the user has not collected the point of interest, determining the first voice broadcast content for prompting the user to collect the point of interest.

5. The guiding method according to claim 1, characterized in that, After the step of sending the first voice broadcast content to the vehicle to broadcast to the user in the vehicle to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the vehicle system, the guiding method includes: Receiving second reporting information reported by the vehicle that the user in the vehicle has completed adding a note name to the point of interest; Determining the second voice broadcast content to be fed back according to the note name in the second reporting information; Sending the second voice broadcast content to the vehicle to broadcast to the user in the vehicle, so as to prompt the user to input a navigation request for navigating to the point of interest after the next voice wakes up the vehicle system.

6. The guiding method according to claim 1, characterized in that, After the step of querying the collection and note information of all users for the address of the point of interest in the note information set according to the first reported information, the guiding method includes: Determine the push card content to be fed back according to the query result and the type of the point of interest according to a predetermined rule; Send the push card content to the vehicle for display to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

7. The guiding method according to claim 6, wherein, The sending the push card content to the vehicle for display to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system, includes: According to the similarity between the point of interest and the type of each point of interest, within a second preset time, sequentially send the push card content corresponding to the type of each point of interest to the vehicle for display to the user in the vehicle, so as to guide the user to add a note name to the point of interest after completing the collection of the point of interest through the map of the in-vehicle system.

8. A server, wherein, The server includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the guiding method according to any one of claims 1-7 is implemented.

9. A non-volatile computer-readable storage medium containing a computer program, wherein, When the computer program is executed by one or more processors, the guiding method according to any one of claims 1-7 is implemented.

Images