A volume adjusting method, device, equipment and medium

Abstract

The application discloses a volume adjusting method and device, equipment and medium. In the embodiment of the application, the target text sent by the target user is input into the pre-trained intention recognition model to determine the target intention carrying the target action to be executed and the target identification information of the intelligent device executing the target action. According to the target action and the corresponding relationship between the pre-stored action and the volume reduction intention score, the first target score corresponding to the target action is determined. If the first target score is greater than the preset score threshold, the volume adjusting instruction carrying the information of the preset target volume to be adjusted is sent to the intelligent device of the target identification information, that is, when the target user has a strong intention degree to adaptively adjust the volume of the intelligent device according to the first target score corresponding to the target action, the volume adjusting instruction is sent to the intelligent device executing the target action, the adaptive adjustment of the volume of the intelligent device is realized, and the experience of the target user is improved.

Description

Technical Field

[0001] This application relates to the field of smart device technology, and in particular to a volume adjustment method, apparatus, device and medium. Background Technology

[0002] When users play audio on smart devices with voice playback capabilities outside of their rest time, they typically turn the volume up quite high. These devices could be smart speakers, smart refrigerators, smart TVs, or smart robot vacuums. However, if they use the device again during their rest time, the volume may still be at the previously set high level, potentially disturbing the user or their family. Therefore, finding a method that can adaptively adjust the volume of smart devices is crucial.

[0003] In related technologies, the volume of smart devices is generally adjusted based on a pre-saved relationship between ambient noise and broadcast volume. This solves the problem of users not being able to hear clearly when the ambient noise is high, and the sound being too loud when the ambient noise is low. Alternatively, users can send a working mode switching command. Upon receiving the command, the smart device switches to sleep mode or do-not-disturb mode, and then adjusts the volume based on light intensity and ambient noise. However, regardless of the method used, the determination of volume adjustment is primarily based on ambient noise. Since ambient noise is uncontrollable, it can be high even during rest periods. Therefore, relying solely on ambient noise cannot accurately determine when to adjust the smart device's volume. Incorrect volume adjustment can severely impact the user experience. For example, if the ambient noise is high during a user's rest period, increasing the volume at that time will significantly affect the user experience. Summary of the Invention

[0004] This application provides a volume adjustment method, apparatus, device, and medium to solve the problem in the prior art that it is impossible to accurately determine whether to adaptively adjust the volume of a smart device, thus affecting the user experience.

[0005] This application provides a volume adjustment method, the method comprising:

[0006] The target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be performed and the target identification information of the smart device that performs the target action;

[0007] Based on the target action and the pre-saved correspondence between the action and the volume reduction intention score, determine the first target score corresponding to the target action;

[0008] If the first target score is greater than a preset score threshold, a volume adjustment command carrying information about the preset target volume to be adjusted is sent to the smart device containing the target identification information.

[0009] Furthermore, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information, the method further includes:

[0010] Based on the correspondence between the current time and the pre-saved time and the user's rest probability score, determine the second target score corresponding to the current time;

[0011] The product of the second target score and the first target score is updated to the first target score.

[0012] Furthermore, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information, the method further includes:

[0013] Determine the difference between the current time and the end time of the last action performed by the smart device, which is pre-saved with the target identification information;

[0014] Based on the difference and the preset normalization function, the normalized difference is obtained;

[0015] Based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined; wherein, the third target score is inversely proportional to the normalized difference;

[0016] The quotient of the first target score and the third target score is updated to the first target score.

[0017] Furthermore, the volume adjustment command also carries the target action to be executed, so that the smart device with the target identification information executes the target action.

[0018] Furthermore, after determining the target user's target intent, and before determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between actions and volume-down intent scores, the method further includes:

[0019] Determine whether the smart device corresponding to the target identifier information carried in the target intent is a smart device with voice broadcast function. If so, proceed with the subsequent operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0020] This application embodiment also provides a text processing apparatus, the apparatus comprising:

[0021] The determination module is used to input the target text sent by the target user into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be executed and the target identification information of the smart device executing the target action; and determines the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0022] The sending module is used to send a volume adjustment command carrying information about the preset target volume to be adjusted to to the smart device containing the target identification information if the first target score is greater than a preset score threshold.

[0023] Furthermore, the determining module is also used to determine the second target score corresponding to the current time based on the current time and the pre-saved correspondence between the time and the user's rest probability score; and update the first target score by multiplying the second target score by the first target score.

[0024] Furthermore, the determining module is also used to determine the difference between the current time and the end time of the last action performed by the smart device using the pre-saved target identification information; obtain a normalized difference based on the difference and a preset normalization function; determine a third target score corresponding to the probability score of the smart device based on the normalized difference; wherein the third target score is inversely proportional to the normalized difference; and update the quotient of the first target score and the third target score to the first target score.

[0025] Furthermore, the determining module is also used to determine whether the smart device corresponding to the target identification information carried in the target intent is a smart device with voice broadcasting function. If so, the subsequent operation of determining the first target score corresponding to the target action is performed based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0026] This application also provides an electronic device, which includes a processor for executing a computer program stored in a memory to implement the steps of any of the volume adjustment methods described above.

[0027] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the volume adjustment methods described above.

[0028] In this embodiment, the target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent. The target intent carries the target action to be performed and the target identifier information of the smart device performing the target action. Based on the target action and the pre-saved correspondence between the action and the volume reduction intent score, a first target score corresponding to the target action is determined. If the first target score is greater than a preset score threshold, a volume adjustment command carrying information about the preset target volume to be adjusted is sent to the smart device with the target identifier information. This embodiment can determine the target user's willingness to adaptively adjust the volume of the smart device based on the first target score corresponding to the target action carried in the target user's target intent. Then, when the willingness is determined to be strong, a volume adjustment command is sent to the smart device performing the target action, thereby achieving adaptive volume adjustment of the smart device and improving the target user's experience. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A schematic diagram of a volume adjustment process provided for some embodiments of this application;

[0031] Figure 2 A schematic diagram illustrating the relationship between a target action and the degree of intent, provided for some embodiments of this application;

[0032] Figure 3 This application provides a correspondence between actions and volume-down intention scores in some embodiments;

[0033] Figure 4 This is a schematic diagram illustrating a process for determining whether to adjust the volume based on whether the target action is an action of adjusting the volume, according to some embodiments of this application.

[0034] Figure 5 A schematic diagram illustrating the correspondence between time and user rest probability scores provided for some embodiments of this application;

[0035] Figure 6 A schematic diagram illustrating a volume adaptive adjustment process provided in some embodiments of this application;

[0036] Figure 7 A schematic diagram illustrating the relationship between various target scores and the degree of willingness of a target user to adaptively adjust the volume, provided for some embodiments of this application;

[0037] Figure 8 A schematic diagram illustrating the process of determining whether the execution result before the output of the smart device needs to be simplified, provided for some embodiments of this application;

[0038] Figure 9 A comparative diagram showing the data table before and after the simplified execution result provided in some embodiments of this application;

[0039] Figure 10 A schematic diagram of a volume control device provided in some embodiments of this application;

[0040] Figure 11 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] In order to determine whether adaptive adjustment of the volume of a smart device is needed to improve the experience of the target user, embodiments of this application provide a volume adjustment method, apparatus, device, and medium.

[0043] In this embodiment, the target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent. The target intent carries the target action to be performed and the target identification information of the smart device performing the target action. Based on the target action and the pre-saved correspondence between the action and the volume reduction intent score, a first target score corresponding to the target action is determined. If the first target score is greater than a preset score threshold, a volume adjustment command carrying the preset target volume to be adjusted is sent to the smart device with the target identification information.

[0044] Figure 1 A volume adjustment process is illustrated in some embodiments of this application, the process including the following steps:

[0045] S101: Input the target text sent by the target user into the pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be performed and the target identification information of the smart device performing the target action.

[0046] The volume adjustment method provided in this application is applied to an electronic device, which may be a smart terminal, a PC, or a server, etc.

[0047] In this embodiment of the application, in order to determine which smart device the target user wants to perform which action, the electronic device can interact with the target user and obtain the target text input by the target user during the interaction. The target text is text that carries the target user's intention, such as "Turn on the air conditioner for me" or "Let the robot vacuum sweep the floor".

[0048] Specifically, electronic devices can obtain target text input by the target user through text interaction; they can also obtain voice input by the target user through voice interaction. Specifically, the electronic device can capture the user's voice input through its microphone, and then convert the voice into text, which is the target text. Specifically, speech-to-text conversion can be based on Hidden Markov Models (HMMs), Tacotron models, or FastSpeech models.

[0049] To determine the target user's intent, in this embodiment, an intent recognition model is pre-trained. After the target text is input into the intent recognition model, the model outputs the target user's intent, which carries the target action to be performed and the target identifier information of the smart device performing the action. The target action to be performed could be lowering the volume, turning on the air conditioner, raising the air conditioner temperature, playing JJ Lin's music, telling a story, etc., and the smart device performing the action could be a smart speaker, smart curtains, smart washing machine, smart TV, smart robot, robot vacuum cleaner, etc. Which smart device corresponds to which identifier information, and which smart device can perform which actions, are pre-set and known. For example, a smart speaker can perform actions such as adjusting volume, playing music, and telling stories, and its corresponding identifier information could be 1; a smart TV can perform actions such as adjusting volume and playing audio, and its corresponding identifier information could be 2, and so on.

[0050] The pre-trained intent recognition model can be an ELMO model, a generative pre-training (GPT) model, a pre-trained language representation model (Bidirectional Encoder Representation from Transformers, BERT), etc.

[0051] For example, if the target text sent by the target user is "Please turn on the air conditioner for me", then the target intent corresponding to the target text is "turn on the air conditioner", where the target action to be performed in the target intent is "turn on the air conditioner", and the target identification information of the smart device that performs the target action is the identification information corresponding to "air conditioner"; if the target text sent by the target user is "lower the volume of the music", then the target intent corresponding to the target text is "turn down the volume of the music", where the target action to be performed in the target intent is "turn down the volume", and the target identification information of the smart device that performs the target action is the identification information corresponding to "smart speaker".

[0052] S102: Determine the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intention score.

[0053] In the embodiments of this application, the target action is different, and the degree to which the target user is willing to adaptively adjust the volume of the smart device will also be different. For example, if the target action is "turn up the volume", the target user's degree of willingness to adaptively adjust the volume of the smart device is particularly low, and they do not want the volume of the smart device to be adaptively turned down. If the target action is a smart device control action, such as "turn on the air conditioner", then the smart device will only output a prompt message indicating that the action has been completed after the target action is executed. In this case, if the target user is in a rest period, the degree of willingness to adaptively adjust the volume is particularly high, that is, they want the volume of the smart device to be adaptively turned down.

[0054] Figure 2 This application provides a schematic diagram illustrating the relationship between a target action and the degree of intent in some embodiments. The following is specifically addressed... Figure 2 Please provide an explanation.

[0055] The main types of actions include: volume control, audio control, weather control, chat, furniture control, time query, and alarm clock.

[0056] The actions under the audio control category include: adjusting volume; actions under the audio category include: listening to music, telling stories, performing crosstalk, storytelling, etc.; actions under the weather category include: checking the weather; actions under the chat category include: chatting; actions under the furniture control category include: controlling the air conditioner, refrigerator, washing machine, oven, etc.; actions under the time query category include: querying the time and modifying the time; actions under the alarm clock category include: setting an alarm and modifying the alarm time.

[0057] It should be noted that there are several execution action classes, and the specific actions contained in each execution action class are pre-set and known.

[0058] Actions with a high degree of willingness to adaptively adjust volume are identified by the identifier "volume_strong," which includes actions such as adjusting volume, listening to music, telling stories, performing crosstalk, and storytelling. Actions with a neutral degree of willingness to adaptively adjust volume are identified by the identifier "volume_medium," which includes actions such as checking the weather and chatting. Actions with a low degree of willingness to adaptively adjust volume are identified by the identifier "volume_weak," which includes actions such as controlling the air conditioner, refrigerator, washing machine, oven, checking the time, modifying the time, setting an alarm, and modifying the alarm time. Here, "volume_strong," "volume_medium," and "volume_weak" are used to characterize the degree of willingness to adaptively adjust volume.

[0059] To accurately determine whether adaptive volume adjustment is needed to improve user experience, this embodiment of the application pre-stores the correspondence between actions and volume-down intention scores. The higher the volume-down intention score, the stronger the target user's willingness to adaptively adjust the volume. To determine the target user's willingness to adaptively adjust the volume, this embodiment of the application, after determining the target action to be executed carried in the target user's target intention, determines a first target score corresponding to the target action based on the target action and the pre-stored correspondence between actions and volume-down intention scores. The higher the first target score, the stronger the target user's willingness to adaptively adjust the volume, and the greater the likelihood that the smart device executing the target action will adaptively adjust the volume.

[0060] Figure 3 The correspondence between actions and volume-down intention scores provided in some embodiments of this application is now addressed. Figure 3 Please provide an explanation.

[0061] Actions with a high degree of willingness to adaptively adjust the volume correspond to a volume reduction intention score of 0.4, actions with a neutral degree of willingness to adaptively adjust the volume correspond to a volume reduction intention score of 0.2, and actions with a low degree of willingness to adaptively adjust the volume correspond to a volume reduction intention score of 0.

[0062] In other words, if the target action is any of the following actions: adjusting volume, listening to music, telling a story, performing crosstalk, or storytelling, the first target score for that target action is 0.4. If the target action is any of the following actions: checking the weather or chatting, the first target score for that target action is 0.2. If the target action is any of the following actions: controlling the air conditioner, refrigerator, washing machine, oven, checking the time, modifying the time, setting an alarm, or modifying the alarm time, the first target score for that target action is 0.

[0063] To improve the efficiency of determining whether to perform adaptive volume adjustment, in this embodiment of the application, the electronic device can pre-store actions with a low degree of willingness to adjust the volume. Before determining the first target score corresponding to the target action, it is first determined whether the target action is an action with a low degree of willingness to adjust the volume. If so, it is not necessary to determine the first target score corresponding to the target action, and it is directly determined that the volume of the smart device performing the target action does not need to be adjusted.

[0064] Specifically, after determining the target action to be executed carried in the target intent, before determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between actions and volume reduction intent scores, it can be first determined whether the target action is a pre-saved action with a low degree of willingness to adjust the volume. If not, the operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between actions and volume reduction intent scores is performed. If so, an execution instruction carrying the target action is directly sent to the smart device that executes the target action so that the smart device executes the target action.

[0065] Figure 4 This application provides a schematic diagram illustrating a process for determining whether to adjust the volume based on whether the target action is an action of adjusting the volume. The following is a description of this process. Figure 4 Please provide an explanation.

[0066] After determining the target action to be executed carried in the target intent, it is determined whether the target action is a pre-saved action with a low degree of willingness to adjust the volume. If so, it is determined that the smart device does not need to perform adaptive volume adjustment, and the execution instruction carrying the target action is sent directly to the smart device that executes the target action. If not, the first target score corresponding to the target action is determined according to the target action and the correspondence between the pre-saved action and the volume reduction intent score.

[0067] S103: If the first target score is greater than a preset score threshold, send a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information.

[0068] To determine whether to adaptively adjust the volume of the smart device performing the target action, that is, to determine whether to adaptively adjust the volume of the smart device with the target identification information, in this embodiment of the application, it can first be determined whether the first target score is greater than a preset score threshold. If so, it is determined that the volume of the smart device with the target identification information will be adaptively adjusted. Specifically, a volume adjustment command carrying information about the preset target volume to be adjusted can be sent to the smart device with the target identification information. If not, it is determined that the volume of the smart device with the target identification information will not be adaptively adjusted.

[0069] The preset target volume is 0.12, 0.13, etc., and can be set according to needs.

[0070] It should be noted that the volume of the smart device with the target identification information may be greater than or less than the preset target volume. If the volume of the smart device with the target identification information is less than the preset target volume, the smart device may not adjust its own volume after receiving a volume adjustment command from the electronic device carrying information to be adjusted to the preset target volume. If the volume of the smart device with the target identification information is greater than the preset target volume, the smart device may adjust its own volume to the preset target volume after receiving a volume adjustment command from the electronic device carrying information to be adjusted to the preset target volume.

[0071] This application embodiment can determine the target user's willingness to adaptively adjust the volume of the smart device based on the first target score corresponding to the target action carried in the target user's target intent. Then, when the willingness is determined to be strong, a volume adjustment command is sent to the smart device that performs the target action, thereby realizing adaptive adjustment of the smart device's volume and improving the target user's experience.

[0072] To determine whether to adaptively adjust the volume of the smart device performing the target action, based on the above embodiments, in this embodiment, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device with the target identification information, the method further includes:

[0073] Based on the correspondence between the current time and the pre-saved time and the user's rest probability score, determine the second target score corresponding to the current time;

[0074] The product of the second target score and the first target score is updated to the first target score.

[0075] Since people's rest time is generally fixed, such as midnight or lunch break, and for the same target action, the willingness of the target user to adaptively adjust the volume of the smart device will differ between rest and non-rest times, and the willingness of the target user to adaptively adjust the volume is very high during rest times and very low during non-rest times, in order to more accurately determine whether to adaptively adjust the volume of the smart device performing the target action to improve the user experience, in this embodiment of the application, the electronic device also pre-stores the correspondence between time and user rest probability score, wherein the user rest probability score is higher during rest times and lower during non-rest times.

[0076] Figure 5 This application provides a schematic diagram illustrating the correspondence between time and user rest probability scores in some embodiments. The following is specifically for... Figure 5 Please provide an explanation.

[0077] For each time period within the 0:00-7:00 time range, the corresponding user rest probability score is 1; for each time period within the 7:00-9:00 time range, the corresponding user rest probability score is 0.6; for each time period within the 9:00-12:00 time range, the corresponding user rest probability score is 0.2; for each time period within the 12:00-14:00 time range, the corresponding user rest probability score is 0.6; for each time period within the 14:00-21:00 time range, the corresponding user rest probability score is 0.2; and for each time period within the 21:00-24:00 time range, the corresponding user rest probability score is 0.6.

[0078] After determining the first target score corresponding to the target action, a second target score corresponding to the current time can be determined based on the current time and the pre-saved correspondence between the time and the user's rest probability score. The first target score is then updated based on this second target score. The higher the second target score, the greater the likelihood that the volume of the smart device performing the target action will be adaptively adjusted.

[0079] In order to update the first target score based on the second target score, in one possible implementation, the sum of the second target score and the first target score can be used to update the first target score.

[0080] In another possible implementation, the product of the second target score and the first target score can also be updated to the first target score.

[0081] To determine whether to adaptively adjust the volume of the smart device performing the target action, based on the above embodiments, in this embodiment, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device with the target identification information, the method further includes:

[0082] Determine the difference between the current time and the end time of the last action performed by the smart device, which is pre-saved with the target identification information;

[0083] Based on the difference and the preset normalization function, the normalized difference is obtained;

[0084] Based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined; wherein, the third target score is inversely proportional to the normalized difference;

[0085] The quotient of the first target score and the third target score is updated to the first target score.

[0086] For each smart device, the target user may have used the smart device recently or may not have used it at all. If the target user has used the smart device recently or is currently using it, it indicates that the target user has a strong willingness to use the smart device and can obtain a good user experience when using it. The shorter the time interval between the current time and the last time the smart device was used, the higher the target user's willingness to adaptively adjust the volume of the smart device. Therefore, in order to more accurately determine whether to adaptively adjust the volume of the smart device performing the target action to improve the user experience, in this embodiment, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying the preset target volume information to be adjusted to the smart device with the target identification information, the difference between the current time and the end time of the last action performed by the smart device with the target identification information (pre-saved) is first determined.

[0087] Since the current time and the end time of the last action performed by the smart device with the pre-saved target identification information may be very large, for convenience, the difference can be normalized first. Specifically, the normalized difference is obtained based on the difference and a preset normalization function. The preset normalization function can be an S-shaped growth curve (sigmoid function). The process of normalizing the difference based on the sigmoid function is existing technology and will not be described in detail here.

[0088] Then, based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined, wherein the third target score is inversely proportional to the normalized difference.

[0089] To determine the third objective score, in one possible implementation, the third objective score can be determined according to the following formula:

[0090] Z(T_now,T_previous)=1 / log(sigmoid|T_now–T_previous|+α)

[0091] Where Z(T_now,T_previous) is the third target score, T_now is the current time, T_previous is the end time of the last action performed by the smart device, and α is a preset parameter.

[0092] In another possible implementation, the third target score can be determined according to the following formula:

[0093] Z(T_now,T_previous)=1 / [sigmoid|T_now–T_previous|+α]

[0094] Where Z(T_now,T_previous) is the third target score, T_now is the current time, T_previous is the end time of the last action performed by the smart device, and α is a preset parameter.

[0095] It should be noted that if the smart device does not stop using the device after performing the last action, then T_now–T_previous is 0.

[0096] Finally, the first target score is updated based on the third target score. The higher the third target score, the greater the likelihood that the volume of the smart device performing the target action will be adaptively adjusted. The first target score can be determined based on the target action and a pre-saved correspondence between the action and the volume-down intention score, or it can be obtained by updating the first target score based on a second target score, where the second target score is determined based on the current time and a pre-saved correspondence between the time and the user's rest probability score.

[0097] In order to update the first target score, in one possible implementation, the difference between the first target score and the third target score can be updated to the first target score.

[0098] In another possible implementation, the quotient of the first target score and the third target score can be updated to the first target score.

[0099] Figure 6 This is a schematic diagram illustrating a volume adaptive adjustment process provided in some embodiments of this application. Now, regarding... Figure 6 Please provide an explanation.

[0100] The electronic device first obtains the voice input by the target user, then performs speech recognition, converts the speech into target text, and inputs the target text into a pre-trained intent recognition model to obtain the target intent corresponding to the target text. The target intent carries the target action to be performed and the target identification information of the smart device that performs the target action.

[0101] Based on the target action and the pre-saved correspondence between the action and the volume reduction intention score, a first target score is determined for the target action. Based on the current time and the pre-saved correspondence between the time and the user's rest probability score, a second target score is determined for the current time. Based on the difference between the current time and the end time of the last action performed by the smart device in the pre-saved target identification information, a third target score is determined for the probability score of using the smart device. The product of the first target score and the second target score is determined, and the quotient of the product and the third target score is determined. The first target score is then updated based on the quotient.

[0102] If the first target score is greater than a preset score threshold, then it is determined that the smart device with the target identification information needs to be adaptively adjusted. A volume adjustment command carrying the preset target volume information to be adjusted is sent to the smart device with the target identification information so that the smart device with the target identification information adjusts the volume to the preset target volume.

[0103] Figure 7 This application provides a schematic diagram illustrating the relationship between various target scores and the degree of a target user's willingness to adaptively adjust the volume, based on some embodiments of the present application. Figure 7 Please provide an explanation.

[0104] The first and second target scores are directly proportional to the target user's willingness to adaptively adjust the volume, while the third target score is inversely proportional to the target user's willingness to adaptively adjust the volume.

[0105] In order to enable the smart device that performs the target action to perform the target action, based on the above embodiments, in this embodiment of the application, the volume adjustment command also carries the target action to be performed, so that the smart device with the target identification information performs the target action.

[0106] In this application embodiment, in order to enable the smart device that performs the target action to perform the target action, in one possible implementation, the electronic device can send an execution instruction carrying the target action to the smart device that performs the target action, so that the smart device that performs the target identification information can execute the target action after receiving the execution instruction.

[0107] To improve efficiency, in another possible implementation, the electronic device may also carry the target action to be performed in the volume adjustment command, and send the volume adjustment command carrying the target action to the smart device with target identification information, so that the smart device with target identification information can execute the target action after receiving the volume adjustment command.

[0108] In this embodiment of the application, after the smart device completes the target action, it will output the execution result obtained after completing the target action. If the target action is a query action, the execution result can be a query result. For example, if the target action is to query the weather, the execution result can be "Sunny". If the target action is a control action, the execution result can be a prompt result after completing the target action. For example, if the target action is "Turn on the air conditioner", the execution result can be "The air conditioner has been turned on for you".

[0109] In this embodiment of the application, if the output execution result contains too much text after the target action is completed, it will often affect the user experience. Therefore, in order to avoid the output execution result being too long and affecting the user experience, in one possible implementation, for some cases where the desire for adaptive volume adjustment is high, the possible output execution result in this case can be simplified in advance. In order to facilitate the subsequent output of the execution result, each smart device can store two data tables: one data table before the execution result is simplified and the other data table after the execution result is simplified. After receiving the volume adjustment command, the smart device can output the execution result of the target action from the data table after the simplified execution result.

[0110] In another possible implementation, the electronic device can also send an Artificial Intelligence of Things (AIOT) command to the smart device. This AIOT command carries a key value; if a simplified execution result is required, the electronic device sets this key value to `auo_volume`. Generally, after determining that the smart device's volume needs adaptive adjustment, the electronic device sends an AIOT command with the key value `auo_volume` to the smart device. Upon receiving the AIOT command, the smart device determines whether the key value in the AIOT command is `auo_volume`. If it is, it outputs the execution result of the target action from the simplified execution result data table; otherwise, it outputs the execution result of the target action from the unsimplified execution result data table.

[0111] Figure 8 This application provides schematic diagrams illustrating the process of determining whether the execution result before the smart device outputs needs simplification, based on some embodiments of the present application. Figure 9 This is a comparative diagram showing the data table before and after the simplified execution result provided in some embodiments of this application. The following is a summary of the comparison. Figure 8 and Figure 9 Please provide an explanation.

[0112] In this embodiment, the execution results of actions with high willingness and actions with neutral willingness can be left unsimplified, while the execution results of actions with low willingness can be simplified. The action is categorized into volume control, audio, weather, and chat categories. If it is, the subsequent output is based on the execution result of the action in the data table before simplification; otherwise, the subsequent output is based on the execution result of the action in the data table after simplification.

[0113] Specifically, the execution results for actions such as air conditioner control, refrigerator control, washing machine control, oven control, time query, time modification, setting alarms, and modifying alarm times need to be simplified. For example, the execution result for the target text "Adjust the air conditioner temperature to 25 degrees" before simplification is "Okay, the air conditioner temperature has been adjusted to 25 degrees," while the simplified execution result is "Okay." Similarly, the execution result for the target text "Query the current time" before simplification is "It's 10 PM now, take a shower and relax," while the simplified execution result is "22 o'clock." See below for details. Figure 9 .

[0114] To accurately determine whether adaptive volume adjustment of smart devices is needed to improve user experience, based on the above embodiments, in this embodiment, after determining the target user's target intent and before determining the first target score corresponding to the target action according to the target action and the pre-saved correspondence between the action and the volume reduction intent score, the method further includes:

[0115] Determine whether the smart device corresponding to the target identifier information carried in the target intent is a smart device with voice broadcast function. If so, proceed with the subsequent operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0116] In this embodiment, since some smart devices cannot provide voice broadcast capabilities, such as lights and curtains, to improve the efficiency of determining whether adaptive volume adjustment is needed, this embodiment can, after determining the target user's target intent, and before determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score, first determine whether the smart device corresponding to the target identifier information carried in the target intent is a smart device with voice broadcast functionality. If so, then proceed with the subsequent operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score. If not, then it is not necessary to determine whether the smart device corresponding to the target identifier information needs adaptive volume adjustment; instead, an execution instruction carrying the target action can be directly sent to the smart device corresponding to the target identifier information, so that the smart device corresponding to the target identifier information executes the target action after receiving the execution instruction.

[0117] Specifically, which smart devices can provide voice broadcast functionality is pre-set and known.

[0118] Figure 10 This application provides a schematic diagram of a volume control device structure, which includes:

[0119] The determination module 1001 is used to input the target text sent by the target user into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be executed and the target identification information of the smart device executing the target action; and determines the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0120] The sending module 1002 is used to send a volume adjustment command carrying information about the preset target volume to be adjusted to to the smart device containing the target identification information if the first target score is greater than a preset score threshold.

[0121] In one possible implementation, the determining module 1001 is further configured to determine a second target score corresponding to the current time based on the current time and the pre-saved correspondence between the time and the user's rest probability score; and update the first target score by multiplying the second target score by the first target score.

[0122] In one possible implementation, the determining module 1001 is further configured to: determine the difference between the current time and the end time of the last action performed by the smart device using the pre-saved target identification information; obtain a normalized difference based on the difference and a preset normalization function; determine a third target score corresponding to the probability score of the smart device based on the normalized difference; wherein the third target score is inversely proportional to the normalized difference; and update the quotient of the first target score and the third target score to the first target score.

[0123] In one possible implementation, the determining module 1001 is further configured to determine whether the smart device corresponding to the target identification information carried in the target intent is a smart device with voice broadcasting function. If so, the subsequent operation of determining the first target score corresponding to the target action is performed based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0124] Based on the above embodiments, this application also provides an electronic device. Figure 11 This application provides a schematic diagram of an electronic device structure, such as... Figure 11 As shown, it includes: processor 1101, communication interface 1102, memory 1103 and communication bus 1104, wherein processor 1101, communication interface 1102 and memory 1103 communicate with each other through communication bus 1104.

[0125] The memory 1103 stores a computer program, which, when executed by the processor 1101, causes the processor 1101 to perform the following steps:

[0126] The target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be performed and the target identification information of the smart device that performs the target action;

[0127] Based on the target action and the pre-saved correspondence between the action and the volume reduction intention score, determine the first target score corresponding to the target action;

[0128] If the first target score is greater than a preset score threshold, a volume adjustment command carrying information about the preset target volume to be adjusted is sent to the smart device containing the target identification information.

[0129] In one possible implementation, the processor 1101 is further configured to determine a second target score corresponding to the current time based on the current time and the pre-saved correspondence between the time and the user's rest probability score; and update the first target score by multiplying the second target score by the first target score.

[0130] In one possible implementation, the processor 1101 is further configured to: determine the difference between the current time and the end time of the last action performed by the smart device using the pre-saved target identification information; obtain a normalized difference based on the difference and a preset normalization function; determine a third target score corresponding to the probability score of the smart device based on the normalized difference; wherein the third target score is inversely proportional to the normalized difference; and update the quotient of the first target score and the third target score to the first target score.

[0131] In one possible implementation, the processor 1101 is further configured to determine whether the smart device corresponding to the target identification information carried in the target intent is a smart device with voice broadcasting function. If so, the processor performs a subsequent operation to determine the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0132] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0133] The communication interface 1102 is used for communication between the aforementioned electronic device and other devices. The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0134] The processors mentioned above can be general-purpose processors, including central processing units, network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits, field-programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0135] Based on the above embodiments, this invention also provides a computer-readable storage medium storing a computer program executable by a processor. When the program runs on the processor, it causes the processor to perform the following steps:

[0136] The target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be performed and the target identification information of the smart device that performs the target action;

[0137] Based on the target action and the pre-saved correspondence between the action and the volume reduction intention score, determine the first target score corresponding to the target action;

[0138] If the first target score is greater than a preset score threshold, a volume adjustment command carrying information about the preset target volume to be adjusted is sent to the smart device containing the target identification information.

[0139] Furthermore, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information, the method further includes:

[0140] Based on the correspondence between the current time and the pre-saved time and the user's rest probability score, determine the second target score corresponding to the current time;

[0141] The product of the second target score and the first target score is updated to the first target score.

[0142] Furthermore, after determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information, the method further includes:

[0143] Determine the difference between the current time and the end time of the last action performed by the smart device, which is pre-saved with the target identification information;

[0144] Based on the difference and the preset normalization function, the normalized difference is obtained;

[0145] Based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined; wherein, the third target score is inversely proportional to the normalized difference;

[0146] The quotient of the first target score and the third target score is updated to the first target score.

[0147] Furthermore, the volume adjustment command also carries the target action to be executed, so that the smart device with the target identification information executes the target action.

[0148] Furthermore, after determining the target user's target intent, and before determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between actions and volume-down intent scores, the method further includes:

[0149] Determine whether the smart device corresponding to the target identifier information carried in the target intent is a smart device with voice broadcast function. If so, proceed with the subsequent operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

[0150] This application embodiment can determine the target user's willingness to adaptively adjust the volume of the smart device based on the first target score corresponding to the target action carried in the target user's target intent. Then, when the willingness is determined to be strong, a volume adjustment command is sent to the smart device that performs the target action, thereby realizing adaptive adjustment of the smart device's volume and improving the target user's experience.

[0151] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0152] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0153] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0155] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A volume adjustment method, characterized in that, The method includes: The target text sent by the target user is input into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be performed and the target identification information of the smart device that performs the target action; Based on the target action and the pre-saved correspondence between the action and the volume reduction intention score, determine the first target score corresponding to the target action; Based on the correspondence between the current time and the pre-saved time and the user's rest probability score, determine the second target score corresponding to the current time; The product of the second target score and the first target score is updated to the first target score; If the first target score is greater than a preset score threshold, a volume adjustment command carrying information about the preset target volume to be adjusted is sent to the smart device containing the target identification information.

2. The method according to claim 1, characterized in that, After determining the first target score corresponding to the target action, if the first target score is greater than a preset score threshold, before sending a volume adjustment command carrying information about the preset target volume to be adjusted to the smart device containing the target identification information, the method further includes: Determine the difference between the current time and the end time of the last action performed by the smart device, which is pre-saved with the target identification information; Based on the difference and the preset normalization function, the normalized difference is obtained; Based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined; wherein, the third target score is inversely proportional to the normalized difference; The quotient of the first target score and the third target score is updated to the first target score.

3. The method according to claim 1, characterized in that, The volume adjustment command also carries the target action to be executed, so that the smart device with the target identification information executes the target action.

4. The method according to claim 1, characterized in that, After determining the target user's target intent, and before determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between actions and volume-down intent scores, the method further includes: Determine whether the smart device corresponding to the target identifier information carried in the target intent is a smart device with voice broadcast function. If so, proceed with the subsequent operation of determining the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score.

5. A volume control device, characterized in that, The device includes: The determination module is used to input the target text sent by the target user into a pre-trained intent recognition model to determine the target user's target intent, wherein the target intent carries the target action to be executed and the target identification information of the smart device executing the target action; and determines the first target score corresponding to the target action based on the target action and the pre-saved correspondence between the action and the volume reduction intent score. The determining module is further configured to determine a second target score corresponding to the current time based on the current time and the pre-saved correspondence between the time and the user's rest probability score; and update the first target score by multiplying the second target score by the first target score. The sending module is used to send a volume adjustment command carrying information about the preset target volume to be adjusted to to the smart device containing the target identification information if the first target score is greater than a preset score threshold.

6. The apparatus according to claim 5, characterized in that, The determining module is further configured to determine the difference between the current time and the end time of the last action performed by the smart device, which is pre-saved with the target identification information; and to obtain a normalized difference based on the difference and a preset normalization function. Based on the normalized difference, a third target score corresponding to the probability score of using the smart device is determined; wherein the third target score is inversely proportional to the normalized difference; the quotient of the first target score and the third target score is updated as the first target score.

7. An electronic device, characterized in that, The electronic device includes a processor that executes a computer program stored in a memory to implement the steps of the volume adjustment method as described in any one of claims 1-4.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the volume adjustment method as described in any one of claims 1-4.

Images