WO2020181748A1

WO2020181748A1 - Speech data interaction method and interaction system for use when adding food material into refrigerator

Info

Publication number: WO2020181748A1
Application number: PCT/CN2019/105072
Authority: WO
Inventors: 台锡敬; 党广明; 吴勇; 王庆福
Original assignee: 青岛海尔电冰箱有限公司; 海尔智家股份有限公司
Priority date: 2019-03-13
Filing date: 2019-09-10
Publication date: 2020-09-17
Also published as: CN111696541B; CN111696541A

Abstract

Provided are a speech data interaction method and interaction system for use when adding food materials into a refrigerator. The method comprises: when a command for adding food materials into a refrigerator is received, receiving continuous speech information (S21); recognizing food material information in the continuous speech information (S22); adjusting a speech information output mode according to a recognized food material type (S23); and updating the recognized food material type to a refrigerator information library (S3). Accordingly, by means of speech data exchange, a refrigerator can automatically update information of a background database.

Description

Voice data interaction method and interaction system for adding ingredients to refrigerator

This application claims the priority of a Chinese patent application whose application date is March 13, 2019, the application number is 201910187877.9, and the invention title is "Voice data interaction method and interaction system for adding ingredients to the refrigerator", the entire content of which is incorporated by reference Incorporated in this application.

Technical field

The invention relates to the field of household appliances, and in particular to a voice data interaction method and an interaction system for adding ingredients to a refrigerator.

Background technique

The Internet of Things is another global informatization wave after computers, the Internet and mobile communications. It is a new stage in the development of global informatization. With the rise of the concept of the Internet of Things, products in the home appliance industry have accelerated their intelligence and launched a series of smart refrigerators.

The existing smart refrigerator sets up a database in the background to store the information of the ingredients in the refrigerator. When adding ingredients to the refrigerator, the user usually manually changes the database information so that the ingredients in the storage space of the refrigerator are consistent with those in the background database. The data is consistent, but the traditional manual input method is more inconvenient. In the process of updating the database, if the user holds other items in his hand, the database cannot be changed. In addition, manually changing the database information is more cumbersome, and the user experience effect is better. difference.

Summary of the invention

The purpose of the present invention is to provide a voice data interaction method and interaction system for adding ingredients to the refrigerator.

In order to achieve one of the above-mentioned objects of the invention, an embodiment of the present invention provides a voice data interaction method for adding ingredients to a refrigerator. The method includes: receiving continuous voice information when a command for adding ingredients to the refrigerator is received;

Recognize the food information in the continuous voice information;

Adjust the voice information output mode according to the recognized food type,

If the type of food in the food information is less than the first threshold preset by the system, the voice output is performed in the first voice information output mode; in the first voice information output mode, the voice output information is the specific type name of the food;

If the type of food in the food information is not less than the first threshold preset by the system, the second voice information output mode is used for voice output; in the second voice information output mode, the voice output information is the number of food types;

If the food is not recognized, the voice output is performed in the third voice information output mode; in the third voice information output mode, the voice output information is warning information;

Update the recognized types of ingredients to the refrigerator information database.

As a further improvement of an embodiment of the present invention, text output is synchronized in the voice information mode;

In the first voice information output mode and in the third voice information output mode, the text output information is the same as the voice output information in the voice information output mode;

In the second voice information output mode, the text output information is the number of types of ingredients and the names of specific types of ingredients.

As a further improvement of an embodiment of the present invention, before "receiving continuous voice information", the method further includes:

M1. After the voice information output mode is turned on for the first time, adjust the current state to the listening state;

M2. In the listening state, receive continuous voice information, and if the voice information meets one of the listening end conditions, confirm the end of listening, and adjust the current state to the recognition state; the listening rules include: Rule 1: continuous listening When the occurrence time of voice information is not less than the first time threshold, rule two: the pause interval for listening to continuous voice information is not less than the second time threshold, rule three: external interrupt command trigger;

M3. In the recognition state, receive continuous voice information.

As a further improvement of an embodiment of the present invention, the step M3 further includes:

The obtained continuous voice information is a keyword query database. If the continuous voice information contains the types of ingredients that match the database, the types of ingredients are acquired, and the number of types of ingredients is counted to adjust the voice information according to the recognized types of ingredients Output mode.

As a further improvement of an embodiment of the present invention, the database includes: an ingredient name database, an ingredient alias database, and a custom ingredient database, wherein the ingredient information in the ingredient alias database and the custom ingredient database all match the ingredient name database;

The step M3 also includes:

The obtained continuous voice information is used as a keyword to query the ingredient name database. If the voice information completely matches the ingredient name database, the query is ended;

If at least part of the voice information does not completely match, then use the unmatched voice information as a keyword to query the food alias database; if the voice information in the query food alias database exactly matches the food alias database, the query ends;

If at least part of the voice information does not completely match the ingredient alias database, use the unmatched voice information as a keyword to query the custom ingredient database. If the voice information of the query custom ingredient database exactly matches the custom ingredient database, the query ends;

If the unmatched voice information cannot match the custom ingredient library, the unmatched voice information will be ignored, and the query will end after the custom ingredient library is searched;

Unify all the matching food types in the voice message as the food name in the food name library for output.

In order to achieve another objective of the above-mentioned invention, an embodiment of the present invention provides a data interaction system for a refrigerator. The system includes: a passive voice processing module for receiving continuous voice message;

Recognize the food information in the continuous voice information;

Adjust the voice information output mode according to the recognized food type,

The update module is used to update the identified types of ingredients to the refrigerator information database.

As a further improvement of an embodiment of the present invention, the trigger module is also used to: synchronously output text in the voice information mode;

As a further improvement of an embodiment of the present invention, the passive voice processing module is also used to: before receiving continuous voice information:

After the voice information output mode is turned on for the first time, adjust the current state to the listening state;

In the listening state, continuous voice information is received. If the voice information meets one of the listening end conditions, the end of listening is confirmed and the current state is adjusted to the recognition state; the listening rules include: Rule 1: Continuous voice information is monitored When the occurrence time of is not less than the first time threshold, rule 2: the pause interval for listening to continuous voice information is not less than the second time threshold, rule 3: external interrupt command trigger;

In the recognition state, continuous voice information is received.

As a further improvement of an embodiment of the present invention, the passive voice processing module is also used for:

The continuous voice information obtained in the recognition state is used as a keyword to query the database. If the continuous voice information contains the type of food material that matches the database, the type of food material is obtained, and the number of food material types is counted to according to the recognized food material Type to adjust the voice information output mode.

The passive voice processing module is specifically used for:

The beneficial effects of the present invention are: the voice data interaction method and interactive system for adding ingredients to the refrigerator of the present invention, after adding ingredients in the refrigerator, the background database information is automatically updated through voice data exchange, the operation is more convenient, and the user experience effect is improved .

Description of the drawings

FIG. 1 is a schematic flowchart of a data interaction method for refrigerators according to an embodiment of the present invention;

FIG. 2 is a schematic flowchart of the voice data interaction method for adding ingredients to the refrigerator based on FIG. 1;

FIG. 3 is a schematic flowchart of a voice data interaction method for removing ingredients from inside and outside the refrigerator provided in FIG. 1;

4 is a schematic flowchart of a voice data interaction method for adding ingredients to a refrigerator according to a preferred embodiment of the present invention;

FIG. 5 is a schematic flow diagram based on the preferred implementation method provided in FIG. 4 when the command is received to remove ingredients from the refrigerator;

Fig. 6 is a schematic diagram of a specific implementation flow of one of the steps in Fig. 4;

Fig. 7 is a schematic diagram of modules of a voice data interaction system for adding ingredients to a refrigerator according to an embodiment of the present invention.

detailed description

The present invention will be described in detail below in conjunction with the specific embodiments shown in the drawings. However, these embodiments do not limit the present invention, and the structural, method, or functional changes made by those skilled in the art based on these embodiments are all included in the protection scope of the present invention.

As shown in FIG. 1, the present invention provides a data interaction method for refrigerators, and the interaction method includes:

S1. When receiving a signal that any door of the refrigerator is closed, a voice interaction message is sent out immediately, asking whether there is an ingredient added or deleted, and if so, the voice information output mode corresponding to adding or deleting an ingredient is activated.

In the specific implementation process of the present invention, the start signal of the voice information output mode is a door closing signal, and this process is a passive triggering process. When receiving any door closing signal of the refrigerator, the user may operate the food in the refrigerator. Either add ingredients to the refrigerator or remove ingredients from the refrigerator, or just open the refrigerator door, as described in the following.

Further, as shown in FIG. 2 and FIG. 3, when a command to add ingredients to the refrigerator is received or a command to remove ingredients from the refrigerator is received, the method further includes: S2, and the step S2 specifically includes : S21, S22, S23.

S21. Receive continuous voice information in the voice information output mode;

S22. Recognizing the food material information in the continuous voice information;

S23. Adjust the voice information output mode according to the recognized food type. If the food type in the food information is less than the first threshold preset by the system, perform the voice output in the first voice information output mode; in the first voice information output mode, the voice The output information is the specific type name of the food; if the food type in the food information is not less than the first threshold preset by the system, the second voice information output mode is used for voice output; in the second voice information output mode, the voice output information It is the quantity of the types of ingredients; if the ingredients are not recognized, the voice output is performed in the third voice information output mode; in the third voice information output mode, the voice output information is warning information.

The system presets the first threshold to be a quantity threshold, and its size can be specified according to needs, for example: 2, 3, 4, etc.

In order to improve the user experience and meet the needs of multiple groups of people, in a preferred embodiment of the present invention, the method further includes: synchronously outputting text in the voice information mode; in the first voice information output mode and the third voice information output In the mode, the text output information is the same as the voice output information in the voice information output mode; in the second voice information output mode, the text output information is the number of food types and the specific types of food names.

In practical applications, during the voice information output mode corresponding to the current door closing signal, the current door closing signal or other door closing signals may be received again. In order to avoid missing operations, the preferred embodiment of the present invention In, when the current shutdown signal is received again, the signal is ignored. When receiving the closing signal of other doors, do the following: If the signal that the other door of the refrigerator is closed occurs during the voice information output mode after the current door closing signal occurs, the interrupt voice information output mode corresponds to the current door instruction Respond and ask whether to continue the current operation or respond to the new door closing signal. If the current operation is continued, the new door closing signal will be recorded in time sequence, and during the voice information output mode corresponding to the current door, follow other doors The time sequence of the closing signal starts the voice information output mode corresponding to other closed door signals; if the response is corresponding to a new door closing signal, during the voice information output mode corresponding to the current door, the voice information output mode corresponding to other closed doors is activated immediately The voice information output mode of the signal, and after completing the command response of other doors, continue the command response of the current door.

In the specific implementation process of the present invention, as shown in FIG. 4, before receiving continuous voice information, the method further includes: M1, after the voice information output mode is turned on for the first time, adjusting the current state to the listening state;

M3. In the recognition state, receive continuous voice information to continue step S2.

Preferably, for step M1, after the voice information output mode is turned on for the first time and before the current state is adjusted to the listening state, the method further includes: outputting active interactive voice information and synchronously outputting text. The active interactive voice information It includes at least: the name of the refrigerator door that generates the door closing signal; the text output information is at least the same as the active interactive voice information.

In a specific example of the present invention, the voice interaction information is for example: "Master, what did you do when you just opened the left refrigerator compartment".

In a preferred embodiment of the present invention, when a signal that any door of the refrigerator is closed is received, the text output can be synchronized when the voice interaction information is sent out. The content of the text output is the same as the voice interaction information, and can also be based on the voice interaction information. Guide information is added to the upper part, and the guide information is a trigger instruction corresponding to adding or deleting an ingredient.

For example: the content of the text output is "Master, what did you do when you just opened the left refrigerator compartment" + the guidance message is "You can say to add ingredients" or "You can say to remove ingredients", at this time, wait to receive " After the voice input of "Add ingredients" or "Remove ingredients", enter the listening state.

In the specific implementation of the present invention, when the command to add ingredients is received, the voice interaction information is continued to be sent out, and text output is synchronized. The voice interaction information that continues to be output includes the specific names of the types of ingredients, for example: "You can say add AA, After BB", it enters the listening state. The "AA" and "BB" indicate specific types of ingredients, such as apples and potatoes.

It should be noted that if the received command is to add ingredients to the refrigerator, the specific types of ingredients in the active interactive voice are preset by the system, and the specific names can be arbitrarily designated by the user or by the system. No specific details will be given here. Repeat.

In the specific embodiment of the present invention, after receiving the command to remove the ingredients, the voice interaction information is continued to be sent out, and text output is performed synchronously, and the text output information is the same as the active interaction voice information.

In an achievable manner of the present invention, the active interactive voice information when removing food materials is determined according to the types of food materials that are already in the refrigerator and are out of date.

Preferably, in conjunction with FIG. 5, if the received command is to remove ingredients from the refrigerator, the method M1 further includes:

If there are no ingredients in the refrigerator, the active interactive voice message that continues to be output is the first deletion reminder information, and the first deletion reminder information identifies that the refrigerator is empty and no operation is required;

If there are ingredients in the refrigerator, query the database, search for the types of existing and expired ingredients in the refrigerator, and record the number and names of the existing and expired types of ingredients;

And adjust the specific output content of the active interactive voice information according to the record;

If there is no expired food type, the output active interactive voice message is the second deletion reminder information, and the second deletion reminder information identifies that there is no expired food in the refrigerator;

If the number of existing and expired food types is less than the second threshold preset by the system, the active interactive voice message is the specific name of each food type;

If the number of existing and expired food types is not less than the second threshold preset by the system, the food types with the longest expiration time and equal to the second threshold are selected, and the active interactive voice message is the specific name of the selected food type .

The second threshold is preset by the system as a quantity threshold, and its size can be specified as required, for example: 2, 3, 4, etc. In the specific example of the present invention, if there is no food in the refrigerator, the voice reminds "no food in the refrigerator"; When there are ingredients in the refrigerator, but no expired ingredients, the voice reminder "there is no expired ingredients in the refrigerator".

In another example of the present invention, the preset second threshold value is 2, and there is 1 expired ingredient, which is "potato". If this is the case, the active interactive voice message may be "You can say remove potatoes";

In another example of the present invention, the preset second threshold is 2, and there are 3 expired ingredients, which are "potato", "apple", and "tomato" in the order of their expiration time. In this way, the voice information is actively exchanged Can be "You can say remove potatoes, apples".

For step M2, the first time threshold and the second time threshold involved in the listening rules are both a fixed time value. In practical applications, the size can be arbitrarily specified. The first time threshold is greater than the second time threshold, for example: 40s , 45s, 50s; the second time threshold is usually small, for example: 5s, 10s, 15s, and the external interrupt command triggers, for example: manual interruption, activation of other functions of the refrigerator, etc., which will not be further described here.

Preferably, for step M3, the step M3 further includes: the obtained continuous voice information is a keyword query database, and if the continuous voice information contains the type of food material matching the database, then the food type is acquired, and Count the number of food types to adjust the voice information output mode according to the recognized food types.

In a specific embodiment of the present invention, the database includes: an ingredient name database, an ingredient alias database, and a custom ingredient database, wherein the ingredient information in the ingredient alias database and the custom ingredient database matches the ingredient name database; correspondingly , As shown in FIG. 6, the step M3 further includes: querying the ingredient name database with the obtained continuous voice information as a keyword; if the voice information completely matches the ingredient name database, the query ends; if at least part of the voice information is not If there is a complete match, use the unmatched voice information as a keyword to query the alias database. If the voice information in the query alias database matches the alias database completely, the query ends; if at least part of the voice information does not completely match the alias database, then Use the unmatched voice information as a keyword to query the custom ingredient library. If the voice information in the query custom ingredient library matches the custom ingredient library, the query ends; if the unmatched voice information does not match the custom ingredient library, Ignore the unmatched voice messages, and end the query after querying the custom food library; and after querying the custom food library, end the query; unify all the matching food types in the voice message as those in the food name library The name of the ingredient is output.

Further, if the received command is to remove the ingredients from the inside of the refrigerator, the step M3 further includes: querying the refrigerator memory bank for the types of ingredients that match the names of the ingredients, and the memory bank stores the names of the ingredients actually stored in the refrigerator , Determine whether the food type matches the refrigerator memory bank, if yes, output the matched food material type according to the rule shown in step S2, and further update the refrigerator storage library; if the refrigerator memory bank cannot be matched, ignore the unmatched The types of ingredients.

For ease of understanding, for the process of step S2, a specific example for adding and deleting ingredients is described in detail.

In Embodiment 1, the received command is a command to add ingredients to the refrigerator.

When the listening is over, the continuous voice information is received. In this example, the received continuous voice information is "add potatoes, apples, corn, pineapple". Further, the food information in the continuous voice information is recognized, respectively "Potato", "Apple", "Corn" and "Pineapple". After querying the database and comparing them, it is confirmed that the added ingredients are "Potato", "Apple", "Corn" and "Pineapple", and "Pineapple" is "Pineapple". "Corresponds to the identification of the ingredient name library.

Further, in an embodiment, the first threshold is set to 5, and corresponding to the first threshold, the first voice information output mode is adopted for output, and the voice output information and text output information are both "potato" and "apple". "Corn" and "pineapple".

In another embodiment, the first threshold is set to 2, corresponding to the first threshold, the second voice information output mode is used for output, the voice output information "add 4 kinds of ingredients", and the text output information is "add 3 kinds" Ingredients" + "Potato" "Apple" "Corn" "Pineapple"

In still another embodiment, after the listening is finished, no food is recognized. Thus, the third voice information output mode is used for output, and the voice output information and text output information are both "not recognized added food."

In Embodiment 2, the received command is a command to remove ingredients from the refrigerator.

After listening, the continuous voice message is received. In this example, the received continuous voice message is "remove potatoes, apples, corn, pineapple". Further, the food information in the continuous voice message is recognized, which They are "potatoes", "apples", "corns" and "pineapples". After querying the database and comparing them, it is confirmed that the removed ingredients are "potatoes", "apples", "corns" and "pineapples", where "pineapples" are "Pineapple" corresponds to the identity of the ingredient name library.

Further, in an embodiment, the first threshold is set to 4, corresponding to the first threshold, the first voice information output mode is adopted for output, and the voice output information and text output information are both "potato" and "apple". "Corn" and "pineapple".

In another embodiment, the first threshold is set to 2, corresponding to the first threshold, the second voice information output mode is used for output, the voice output information "remove 4 kinds of ingredients", and the text output information is "remove 3 kinds of ingredients" + "potatoes", "apples", "corns" and "pineapples".

In yet another embodiment, after the listening is finished, no food is recognized, so the third voice information output mode is used for output, and the voice output information and text output information are both "unrecognized removed food."

Further, after confirming the addition or deletion of ingredients, the method further includes: S3. Update the refrigerator information database according to the added or deleted ingredients to unify refrigerator information.

As shown in FIG. 7, in an embodiment of the present invention, a data interaction system for a refrigerator includes: a trigger module 100, a passive voice processing module 200, an update module 300, and a storage module 400. The storage module includes a database 401 and a refrigerator. Repository 403.

The trigger module 100 is used to immediately send out a voice interaction message when receiving a signal that any door of the refrigerator is closed, asking whether there is an added or deleted ingredient, and if so, activate a voice information output mode corresponding to the added or deleted ingredient.

Further, when a command to add ingredients to the refrigerator is received, or a command to remove ingredients from the refrigerator is received, the passive voice processing module 200 is used to receive continuous voice information in the voice information output mode; The food information in the voice information; adjust the voice information output mode according to the recognized food type, if the food type in the food information is less than the first threshold preset by the system, the voice output is performed in the first voice information output mode; first voice In the information output mode, the voice output information is the specific type name of the food; if the food type in the food information is not less than the first threshold preset by the system, the second voice information output mode is used for voice output; the second voice information output mode Below, the voice output information is the number of food types; if the food is not recognized, the voice output is performed in the third voice information output mode; in the third voice information output mode, the voice output information is warning information.

In order to improve the user experience and meet the needs of various groups of people, in the preferred embodiment of the present invention, the passive voice processing module 200 is also used for synchronous text output in the voice information mode; in the first voice information output mode and the third voice In the information output mode, the text output information is the same as the voice output information in the voice information output mode; in the second voice information output mode, the text output information is the number of food types and the specific types of food names.

In practical applications, during the voice information output mode corresponding to the current door closing signal, the current door closing signal or other door closing signals may be received again. In order to avoid missing operations, the preferred embodiment of the present invention When the trigger module 100 confirms that it receives the current shutdown signal again, it ignores the signal. When receiving the closing signal of other doors, the following operation is performed: if the trigger module 100 receives the signal of closing the other doors of the refrigerator and occurs during the voice information output mode after the current door closing signal occurs, the interrupt voice information output mode corresponds to the current Respond to the instruction of the door, and ask whether to continue the current operation or respond to the new door closing signal. If the current operation is continued, the new door closing signal is recorded in time sequence, and during the voice information output mode corresponding to the current door, Start the voice information output mode corresponding to other closed door signals according to the time sequence of other door closing signals; if the response is corresponding to the new door close signal, during the voice information output mode corresponding to the current door, the immediate activation corresponds to The voice information output mode of other door closing signals, and after completing the command response of other doors, continue the command response of the current door.

In the specific implementation process of the present invention, before receiving continuous voice information, the passive voice processing module 200 is also used to adjust the current state to the listening state after the voice information output mode is turned on for the first time; in the listening state, receiving continuous voice information, If the voice information satisfies one of the listening end conditions, confirm the end of listening and adjust the current state to the recognition state; the listening rules include: Rule 1: When the occurrence time of listening to continuous voice information is not less than the first time threshold, Rule 2: The pause interval for listening to continuous voice information is not less than the second time threshold; Rule 3: Triggered by an external interrupt command; in the recognition state, continuous voice information is received.

Preferably, the passive voice processing module 200 after the voice information output mode is turned on for the first time and before adjusting the current state to the listening state, is also used to: output active interactive voice information and synchronously output text. The active interactive voice information is at least It includes: the name of the refrigerator door that generates the door closing signal; the text output information is at least the same as the active interactive voice information.

In a preferred embodiment of the present invention, when the passive voice processing module 200 receives the signal that any door of the refrigerator is closed, it can synchronize the text output when sending out the voice interactive information. The content of the text output is the same as the voice interactive information, or it can be in the voice Guide information is added on the basis of the interactive information, and the guide information is a trigger instruction corresponding to adding or deleting an ingredient.

In a specific embodiment of the present invention, after the passive voice processing module 200 confirms that it receives the command to remove the ingredients, it continues to send out the voice interaction information and synchronously outputs the text. The text output information is the same as the active interaction voice information.

In an implementable manner of the present invention, the active interactive voice information when the passive voice processing module 200 removes food materials is determined according to the types of food materials that are already in the refrigerator and are out of date.

Preferably, if the received command is to remove ingredients from the refrigerator, the passive voice processing module 200 is also used for if there are no ingredients in the refrigerator, the active interactive voice message that continues to be output is the first deletion reminder message. One deletes the reminder message to indicate that the refrigerator is empty and does not require operation;

Preferably, the passive voice processing module 200 is also used to obtain continuous voice information as a keyword query database. If the continuous voice information contains the type of food material matching the database, the type of the food material is obtained and the statistics of the type of food material are obtained. Quantity to adjust the voice information output mode according to the recognized food type.

In a specific embodiment of the present invention, the database 401 includes: an ingredient name database, an ingredient alias database, and a custom ingredient database, wherein the ingredient information in the ingredient alias database and the custom ingredient database all match the ingredient name database; correspondingly; Yes, the passive voice processing module 200 is also used to obtain continuous voice information to query the food name database as keywords. If the voice information completely matches the food name database, the query ends; if at least part of the voice information does not completely match, then Use the unmatched voice information as the key to query the ingredient alias database. If the voice information in the query ingredient alias database exactly matches the ingredient alias database, the query ends; if at least part of the voice information does not completely match the ingredient alias database, the unmatched The voice information is a keyword to query the custom ingredient library. If the voice information of the query custom ingredient library exactly matches the custom ingredient library, the query is ended; if the unmatched voice information cannot match the custom ingredient library, ignore the unmatched After querying the custom ingredient library, end the query; and after querying the custom ingredient database, end the query; unify all the matching ingredients in the voice message as the name of the ingredient in the ingredient name database for output .

Further, if the received command is to remove ingredients from the inside of the refrigerator, the passive voice processing module 200 is also used to query the refrigerator memory bank for the types of ingredients that match the names of the ingredients, and the memory bank stores the ingredients actually stored in the refrigerator. Name, determine whether the food type matches the refrigerator memory, if yes, output the matched food type, and further update the refrigerator storage 403; if it cannot match the refrigerator memory 403, ignore the unmatched food type .

Further, after confirming the addition or deletion of ingredients, the update module 300 is used to update the refrigerator information database according to the added or deleted ingredients to unify the refrigerator information.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the modules described above can refer to the corresponding process in the foregoing method implementation, which will not be repeated here.

To sum up, the data interaction method and interaction system for refrigerators of the present invention, after adding or removing ingredients from the refrigerator, through voice data exchange, the background database information is automatically updated, the operation is more convenient, and the user is improved Experience the effect.

In the several implementation manners provided by the present invention, it should be understood that the implementation manners of the structures, systems, and methods described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods, for example, multiple modules or components may be combined or integrated into another device, or some features may be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or modules, and may be in electrical, mechanical or other forms.

The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed to multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the objectives of the solutions of this embodiment.

In addition, the functional modules in the various embodiments of the present invention may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware, or in the form of hardware plus software functional modules.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments are modified, or some of the technical features are equivalently replaced; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

A voice data interaction method for adding ingredients to a refrigerator, characterized in that the method includes:

When receiving a command to add ingredients to the refrigerator, receive continuous voice messages;

Recognize the food information in the continuous voice information;

Adjust the voice information output mode according to the recognized food type,

If the type of food in the food information is less than the first threshold preset by the system, the voice output is performed in the first voice information output mode; in the first voice information output mode, the voice output information is the specific type name of the food;

If the type of food in the food information is not less than the first threshold preset by the system, the second voice information output mode is used for voice output; in the second voice information output mode, the voice output information is the number of food types;

If the food is not recognized, the voice output is performed in the third voice information output mode; in the third voice information output mode, the voice output information is warning information;

Update the recognized types of ingredients to the refrigerator information database.
The voice data interaction method for adding ingredients to the refrigerator according to claim 1, wherein the method further comprises: synchronously outputting text in the voice information mode;

In the first voice information output mode and in the third voice information output mode, the text output information is the same as the voice output information in the voice information output mode;

In the second voice information output mode, the text output information is the number of types of ingredients and the names of specific types of ingredients.
The voice data interaction method for adding ingredients to the refrigerator according to claim 1, wherein before "receiving continuous voice information", the method further comprises:

M1. After the voice information output mode is turned on for the first time, adjust the current state to the listening state;

M2. In the listening state, receive continuous voice information, and if the voice information meets one of the listening end conditions, confirm the end of listening, and adjust the current state to the recognition state; the listening rules include: Rule 1: continuous listening When the occurrence time of voice information is not less than the first time threshold, rule two: the pause interval for listening to continuous voice information is not less than the second time threshold, rule three: external interrupt command trigger;

M3. In the recognition state, receive continuous voice information.
The voice data interaction method for adding ingredients to the refrigerator according to claim 3, wherein:

The step M3 also includes:

The obtained continuous voice information is a keyword query database. If the continuous voice information contains the types of ingredients that match the database, the types of ingredients are acquired, and the number of types of ingredients is counted to adjust the voice information according to the recognized types of ingredients Output mode.
The voice data interaction method for adding ingredients to the refrigerator according to claim 4, wherein:

The database includes: an ingredient name database, an ingredient alias database, and a custom ingredient database, wherein the ingredient information in the ingredient alias database and the custom ingredient database all match the ingredient name database;

The step M3 also includes:

The obtained continuous voice information is used as a keyword to query the ingredient name database. If the voice information completely matches the ingredient name database, the query is ended;

If at least part of the voice information does not completely match, then use the unmatched voice information as a keyword to query the food alias database; if the voice information in the query food alias database exactly matches the food alias database, the query ends;

If at least part of the voice information does not completely match the ingredient alias database, use the unmatched voice information as a keyword to query the custom ingredient database. If the voice information of the query custom ingredient database exactly matches the custom ingredient database, the query ends;

If the unmatched voice information cannot match the custom ingredient library, the unmatched voice information will be ignored, and the query will end after the custom ingredient library is searched;

Unify all the matching food types in the voice message as the food name in the food name library for output.
A voice data interaction system for adding ingredients to a refrigerator, characterized in that the system includes:

Passive voice processing module, used to receive continuous voice information when receiving a command to add ingredients to the refrigerator;

Recognize the food information in the continuous voice information;

Adjust the voice information output mode according to the recognized food type,

If the type of food in the food information is less than the first threshold preset by the system, the voice output is performed in the first voice information output mode; in the first voice information output mode, the voice output information is the specific type name of the food;

If the type of food in the food information is not less than the first threshold preset by the system, the second voice information output mode is used for voice output; in the second voice information output mode, the voice output information is the number of food types;

If the food is not recognized, the voice output is performed in the third voice information output mode; in the third voice information output mode, the voice output information is warning information;

The update module is used to update the identified types of ingredients to the refrigerator information database.
The voice data interaction system for adding ingredients to the refrigerator according to claim 6, wherein the passive voice processing module is further used to: synchronously output text in the voice information mode;

In the first voice information output mode and in the third voice information output mode, the text output information is the same as the voice output information in the voice information output mode;

In the second voice information output mode, the text output information is the number of types of ingredients and the names of specific types of ingredients.
The voice data interaction system for adding ingredients to the refrigerator according to claim 6, wherein the passive voice processing module is further used for: before receiving continuous voice information:

After the voice information output mode is turned on for the first time, adjust the current state to the listening state;

In the listening state, continuous voice information is received. If the voice information meets one of the listening end conditions, the end of listening is confirmed and the current state is adjusted to the recognition state; the listening rules include: Rule 1: Continuous voice information is monitored When the occurrence time of is not less than the first time threshold, rule 2: the pause interval for listening to continuous voice information is not less than the second time threshold, rule 3: external interrupt command trigger;

In the recognition state, continuous voice information is received.
The voice data interaction system for adding ingredients to the refrigerator according to claim 8, wherein the passive voice processing module is further used for:

The continuous voice information obtained in the recognition state is used as a keyword to query the database. If the continuous voice information contains the type of food material that matches the database, the type of food material is obtained, and the number of food material types is counted to according to the recognized food material Type to adjust the voice information output mode.
The voice data interaction system for adding ingredients to the refrigerator according to claim 9, wherein:

The database includes: an ingredient name database, an ingredient alias database, and a custom ingredient database, wherein the ingredient information in the ingredient alias database and the custom ingredient database all match the ingredient name database;

The passive voice processing module is specifically used for:

The obtained continuous voice information is used as a keyword to query the ingredient name database. If the voice information completely matches the ingredient name database, the query is ended;

If at least part of the voice information does not completely match, then use the unmatched voice information as a keyword to query the food alias database; if the voice information in the query food alias database exactly matches the food alias database, the query ends;

If at least part of the voice information does not completely match the ingredient alias database, use the unmatched voice information as a keyword to query the custom ingredient database. If the voice information of the query custom ingredient database exactly matches the custom ingredient database, the query ends;

If the unmatched voice information cannot match the custom ingredient library, the unmatched voice information will be ignored, and the query will end after the custom ingredient library is searched;

Unify all the matching food types in the voice message as the food name in the food name library for output.