CN-116827705-B - Speech scene interaction dynamic display method and household appliance interaction system based on near field communication

CN116827705BCN 116827705 BCN116827705 BCN 116827705BCN-116827705-B

Abstract

The invention particularly relates to a near field communication-based dynamic display method of voice scene interaction and a household appliance interaction system, wherein the near field communication-based dynamic display method of voice scene interaction comprises the following steps of obtaining the number N of existing voice scene interactions in a unit time period t; when the number N=1, the voice scene is the display scene of the near field communication tag, when the number N >1, the maximum dynamic display value Y in the unit time period t is obtained, the voice scene corresponding to the maximum dynamic display value Y is set as the display scene of the near field communication tag, when the number N=0, the number M of the voice scene interactions in the unit time period t is obtained, wherein the maximum dynamic display value Y is determined according to the interaction time sequence, the interaction times and the interaction duration of the voice scenes in the corresponding time period. The invention displays the voice scene by utilizing the screen of the intelligent equipment, and has the advantages of better man-machine property, rationality and realizability.

Inventors

YAN YANG
KONG LINGLEI
ZHANG JINGRUI
LI GUIXI

Assignees

青岛海尔电冰箱有限公司
海尔智家股份有限公司

Dates

Publication Date: 20260512
Application Date: 20220321

Claims (7)

1. A dynamic display method of voice scene interaction based on near field communication is characterized by comprising the following steps: acquiring the number N of existing voice scene interactions in a unit time period t; When the number N=1, the voice scene is the display scene of the near field communication tag; When the number N is more than 1, acquiring a maximum dynamic display value Y in a unit time period t, and setting a voice scene corresponding to the maximum dynamic display value Y as a display scene of the near field communication tag; when the number n=0, the number M of the stored voice scene interactions in the unit time period t is acquired; When the number m=1, the stored voice scene in the unit time period t is the display scene of the near field communication tag; When the number M is more than 1, acquiring a last interactive continuous voice scene in a previous unit time period t-1, and setting the last interactive continuous voice scene as a display scene of a near field communication tag; When the number m=0, acquiring a latest unit time period t-x before the unit time period t, in which the voice scene interaction exists; when the latest unit time period t-x is acquired, acquiring the maximum dynamic display value Y in the latest unit time period t-x, and setting a voice scene corresponding to the maximum dynamic display value Y as a display scene of the near field communication tag; The maximum dynamic display value Y is determined according to the interaction time sequence, the interaction times and the interaction duration of the voice scene in the corresponding time period, and specifically is: Acquiring weight vectors W T of single scenes of non-repeated interaction in the previous unit time period t-1, wherein ,W T ={α,β,λ,η},α= o i /Σo i ,β= f i /Σf i ,λ= d normi /Σd normi ,η= p i /Σp i ,i∈(1,n-m); Upon obtaining the weight vector W T , a maximum dynamic display value Y is calculated, where y=max ({ o i ,f i ,d normi ,p i } W T , i e (1, n-m); when the number m=0, a latest unit time period t-x and a next latest unit time period t-y of the voice scene existing before the unit time period t are acquired; When the latest unit time period t-x and the next latest unit time period t-Y are obtained, respectively obtaining the maximum dynamic display value Y t-x 、Y t-y corresponding to the two time periods; when the maximum dynamic display value Y t-x 、Y t-y is obtained, calculating the maximum dynamic display value Y, wherein Y= (Y t-x ,Y t-y ) (theta, phi), wherein (Y t-x ,Y t-y ) and (theta, phi) are both set as vectors, theta is the weighted dynamic display value of the nearest unit time period t-x, phi is the weighted dynamic display value of the next nearest unit time period t-Y, and theta > phi; when the maximum dynamic display value Y is obtained, the voice scene corresponding to the maximum dynamic display value Y is the display scene of the near field communication tag.
2. The method for dynamically displaying voice scene interaction based on near field communication according to claim 1, wherein in the unit time period t, the larger the maximum dynamic display value Y corresponding to the voice scene with the newer set interaction time is, the larger the maximum dynamic display value Y corresponding to the more set interaction times is, and the larger the maximum dynamic display value Y corresponding to the longer set interaction time is.
3. The dynamic presentation method of near field communication based voice scene interactions of claim 2, wherein: acquiring the total interaction times f of each single scene of non-repeated interaction in a unit time period t; When the total interaction times F of each single scene are obtained, a single scene interaction times set F, F= { F 1 i ,f 2 j ,…,f n-m-1 k ,f n-m l } generated by combining the total interaction times F of each single scene is obtained, wherein i, j, k and l are the interaction times of the single scene which are not repeatedly interacted in a unit time period t, and i, j, k and l belong to N.
4. A method for dynamic presentation of near field communication based voice scene interactions as claimed in claim 3 wherein: Acquiring interaction time d of each single scene of non-repeated interaction in a unit time period t; when the interaction time length d of each single scene of non-repeated interaction in the unit time period t is obtained, the average interaction time length u of the interaction time length d of each single scene and the variance sigma of the interaction time length d are obtained; When the average interaction time length u and the variance sigma of the interaction time length are obtained, carrying out normalization processing on the interaction time length d of each single scene to obtain normalized interaction time length d norm of each single scene, wherein d norm = (d-u)/sigma; When the normalized interaction duration D norm of each single scene is obtained, the normalized interaction durations D norm are combined to generate a normalized interaction duration set D norm as D norm ={d norm1 ,d norm2 ,…,d norm,n-m-1 ,d norm,n-m .
5. The method for dynamically exhibiting near field communication-based voice scene interactions according to claim 4, wherein: acquiring single scene preamble decision times p of each single scene of non-repeated interaction in a previous unit time period t-1; When the single scene preamble decision times P are obtained, the single scene preamble decision times P are combined to generate a single scene interaction times set P which is P= { P 1 ,p 2 ,…,p n-m-1 ,p n-m }.
6. The method for dynamically exhibiting voice scene interaction based on near field communication as recited in claim 1, wherein θ is 0.6 and ψ is 0.4.
7. The household appliance interaction system is characterized by comprising an intelligent voice household appliance, a cloud server, a decision center, a near field communication tag and intelligent equipment with video display; Wherein the smart device is for identifying a near field communication tag, the decision center being configured to execute to implement the near field communication based dynamic presentation method of voice scene interaction as claimed in any of claims 1-6.

Description

Speech scene interaction dynamic display method and household appliance interaction system based on near field communication Technical Field The invention relates to the field of intelligent household appliances, in particular to a dynamic display method of voice scene interaction based on near field communication and a household appliance interaction system. Background With the continuous improvement of the living standard of people, refrigerators with intelligent voice modules are increasingly appeared in families of people, and the convenience of using the refrigerators by people is improved. People can trigger corresponding scene instructions only by waking up the refrigerator, and the refrigerator can be controlled without limb actions, so that hands of people are greatly liberated. However, although voice refrigerators facilitate people's lives, it is impractical to remember voice replies by only hearing and brains of people when inquiring and operating in some scenes (e.g., menu, food management, weather, etc.), and particularly, there is a problem in a non-screen voice refrigerator. For example, people need to make a dish, and if the dish is made only according to the corpus broadcasted on the refrigerator, the dish can be just barely finished by repeatedly playing the dish for many times. In addition, although the appearance of NFC function can show the scene required by the user through the intelligent device, how to reasonably show the scene required by the user when multiple scenes are continued or exist simultaneously is also a problem to be solved. Most of the current voice scene display adopts a method of fixing a scene, namely fixing the last user voice interaction scene, and writing the links required by the scene into a designated NFC tag. Although the display method can always save the last interactive scene of the user, when a plurality of scenes are continuously or parallelly performed, for example, when a menu and a music scene are played in the same step, the situation that the music is still being played after the menu is broadcast is generated due to the fact that the broadcasting time of the menu is short and the music time is long, if only the scene links of the menu are recorded at the moment and the scene links of the music are not recorded, when the user wants to view the detail page of the music scene, the situation that the detail page of the music cannot be displayed is generated, and the experience satisfaction degree of the user is greatly reduced. Therefore, it is necessary to research a dynamic display method of voice scene interaction based on near field communication and a household appliance interaction system to solve the above problems. Disclosure of Invention The invention aims to provide a dynamic display method of voice scene interaction based on near field communication. In order to achieve the above object, an embodiment of the present invention provides a method for dynamically exhibiting voice scene interaction based on near field communication, including the steps of: acquiring the number N of existing voice scene interactions in a unit time period t; When the number N=1, the voice scene is the display scene of the near field communication tag; When the number N is more than 1, acquiring a maximum dynamic display value Y in a unit time period t, and setting a voice scene corresponding to the maximum dynamic display value Y as a display scene of the near field communication tag; when the number n=0, the number M of the stored voice scene interactions in the unit time period t is acquired; When the number m=1, the stored voice scene in the unit time period t is the display scene of the near field communication tag; When the number M is more than 1, acquiring a last interactive continuous voice scene in a previous unit time period t-1, and setting the last interactive continuous voice scene as a display scene of a near field communication tag; When the number m=0, acquiring a latest unit time period t-x before the unit time period t, in which the voice scene interaction exists; when the latest unit time period t-x is acquired, acquiring the maximum dynamic display value Y in the latest unit time period t-x, and setting a voice scene corresponding to the maximum dynamic display value Y as a display scene of the near field communication tag; The maximum dynamic display value Y is determined according to the interaction time sequence, the interaction times and the interaction duration of the voice scene in the corresponding time period. As a further improvement of an embodiment of the present invention, in the unit time period t, the larger the maximum dynamic display value Y corresponding to the voice scene with the newer set interaction time, the larger the maximum dynamic display value Y corresponding to the larger set interaction times, and the larger the maximum dynamic display value Y corresponding to the longer set interaction time. As a further improvement