Method for determining a response device, method for determining and transmitting a probability value policy, computer-readable storage medium, computer program, device, and server

Abstract

A method, device, server and computer-readable storage medium for determining a response device, the method comprising the steps of: after receiving a voice control signal, obtaining a target probability value for the device to be a response device according to a preset probability value policy, wherein the probability value policy is determined based on response status information of the device to the voice control signal (S101); determining a target S / N ratio based on the target probability value and the S / N ratio of the received voice control signal (S102); and determining whether the device is a response device to the voice control signal based on the target S / N ratio (S103).

Description

[Technical Field]

[0001] This disclosure claims priority to a Chinese patent application filed on August 30, 2021, bearing application number "202111007071.0" and entitled "Response device determination method, device and server," the entire contents of which are incorporated herein by reference.

[0002] The present disclosure relates to the field of electrical equipment, and more particularly to a method, device and server for determining a response device. [Background technology]

[0003] In a smart home system, if there are multiple devices with voice response functions in the same space, when a voice wake-up command is received, the S / N ratio of each device is calculated, and the device with the highest S / N ratio is selected and woken up.

[0004] In practical applications, if a user intends to wake up device A, but the device that is actually woken up is device B, there is a problem that the actual response situation does not match the request. Summary of the Invention

[0005] The main objective of the present disclosure is to provide a method, device and server for determining a response device that solves the problem in the prior art that the actual response status of a device in a voice control scenario does not match the request.

[0006] In order to achieve the above objectives, the present disclosure provides: After receiving the voice control signal, obtaining a target probability value of the device becoming a responding device according to a preset probability value policy, the probability value policy being determined based on response status information of the device to the voice control signal; determining a target signal-to-noise ratio based on the target probability value and a signal-to-noise ratio of the received audio control signal; and determining whether the device is a response device for the audio control signal based on the target S / N ratio.

[0007] According to an embodiment of the present disclosure, the probability value policy includes a method for determining a probability value corresponding to each time interval; obtaining a target probability value for the device to be a responder, obtaining a reception time of the audio control signal; and determining the target probability value according to a determination scheme in which the reception time corresponds to a time interval in which the reception time falls.

[0008] According to one embodiment of the present disclosure, the method further comprises: calculating a similarity between the keyword in the voice control signal and a predetermined standard word to obtain reliability information; and a step of not allowing a device to determine whether it will be a response device for the voice control signal if the reliability information is smaller than a predetermined reliability threshold, the reliability threshold being determined by the magnitude of a target probability value.

[0009] According to one embodiment of the present disclosure, the method further comprises: recording the number of times the device becomes a responding device and obtaining response status information; sending response status information; receiving update information for a probability value policy corresponding to the response status information.

[0010] A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements any of the methods described above.

[0011] 1. An apparatus comprising: a memory adapted to store a computer program; a processor adapted to execute said computer program to effectuate any of the methods described above.

[0012] A method for determining a response device, comprising: obtaining response status information of a device to an audio control signal, the response status information including the number of times the device is a response device; determining a probability value policy for the device based on the number of times in the response status information; and transmitting the probability value policy.

[0013] According to an embodiment of the present disclosure, when the response status information includes the number of times of becoming a response device in each time interval, the probability value policy includes a decision method corresponding to each time interval.

[0014] According to one embodiment of the present disclosure, if the response status information further includes validity information of becoming a response device every time, the determination method for each time interval is determined based on the valid number Ti and / or invalid number Fi in each time interval, where i is a positive integer and represents the number of the time interval.

[0015] According to one embodiment of the present disclosure, a method of obtaining a maximum valid count and / or a minimum valid count from the valid counts of all time intervals when all validity information is valid, calculating at least one of the maximum valid count and the minimum valid count and the valid count Ti, and obtaining a first numerical value as a probability value of the i-th time interval; If all the validity information is invalid, a maximum value and / or a minimum value of the invalid count is obtained from the invalid counts of all the time intervals, and at least one of the maximum value and the minimum value of the invalid count and the invalid count Fi are calculated to obtain a second value, and a third value is obtained as a probability value of the i-th time interval based on a difference between a preset standard value and the second value; If some of the validity information is valid and the other part is invalid, a first value corresponding to the valid number of times of the device and a second value corresponding to the invalid number of times are determined, and the probability value of the i-th time interval is determined by one or more methods of obtaining a fourth value as the probability value of the i-th time interval based on the difference between the first value and the second value.

[0016] A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements any of the methods described above.

[0017] A server a memory adapted to store a computer program; a processor adapted to execute said computer program to effectuate any of the methods described above.

[0018] In the technical solution disclosed herein, the response device is determined on the premise of conforming to the user's operating habits, and the determination operation is more suited to the user's requirements, thereby improving the consistency between the determination result and the user's expectations. [Brief explanation of the drawings]

[0019] In order to more clearly describe the embodiments of the present disclosure or the technical solutions of the prior art, the following briefly describes the necessary drawings used in the description of the embodiments or the prior art. The following drawings are only some embodiments of the present disclosure, and it is also obvious to those skilled in the art that they can obtain other related drawings based on the configurations shown in these drawings without any creative efforts. [Figure 1] FIG. 1 is a flow chart of a method for determining a response device provided in a first embodiment of the present disclosure. [Figure 2] FIG. 10 is a flow chart of a method for determining a response device provided in a second embodiment of the present disclosure. [Figure 3] FIG. 10 is a flow chart of a method for determining a response device provided in the third embodiment of the present disclosure. [Figure 4]FIG. 10 is a flow chart of a method for determining a response device provided in a fourth embodiment of the present disclosure. [Figure 5] FIG. 10 is a flow chart of a method for determining a response device provided in the fifth embodiment of the present disclosure.

[0020] The realization of the objects, functional features and advantages of the present disclosure will be further explained in connection with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION OF THE INVENTION

[0021] The technical solutions in the embodiments of the present disclosure will be described below clearly and completely with reference to the drawings in the embodiments of the present disclosure, but it is clear that the described embodiments are only a part of the embodiments of the present disclosure, and are not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without any creative work fall within the scope of protection of the present disclosure.

[0022] It should be noted that all directional indications (e.g., up, down, left, right, front, back, etc.) in the embodiments of the present disclosure are used only to explain the relative positional relationships, movement conditions, etc. between each member in a specific posture (as shown in the drawings), and when that specific posture changes, the indication regarding that directional position also changes accordingly.

[0023] Additionally, references to "first," "second," etc. in this disclosure are used for descriptive purposes only and cannot be understood as indicating or implying the relative importance thereof or as implicitly specifying the number of technical features. Thus, features qualified by "first" and "second" can explicitly or implicitly include at least one of the feature. In the description of this disclosure, "plurality" means at least two, e.g., two, three, unless otherwise clearly and specifically limited.

[0024] In the present disclosure, unless otherwise expressly specified or limited, the terms "connected" and "fixed" should be generally understood, and for example, "fixed" may mean a fixed connection, a detachable connection, or an integral connection, may be a mechanical connection, or an electrical connection, and unless otherwise expressly limited, may be a direct connection or an indirect connection via an intermediate medium, or may be internal communication between two elements or interaction between two elements. Those skilled in the art can understand the specific meanings of the above terms in the present disclosure depending on the specific circumstances.

[0025] It should be noted that the technical solutions in each embodiment of the present disclosure can be combined with each other, but this must be based on what a person skilled in the art can achieve. If the combination of technical solutions is inconsistent with each other or cannot be realized, it should be considered that such combination of technical solutions does not exist and is not included in the scope of protection required by the present disclosure.

[0026] Example 1

[0027] 1 is a flow chart of a method for determining a response device provided in a first embodiment of the present disclosure. As shown in FIG. 1, the method includes the following steps 101 to 103.

[0028] Step 101: When receiving a voice control signal, a target probability value for determining that the device is a response device is obtained according to a preset probability value policy, the probability value policy being determined based on response status information of the device to the voice control signal.

[0029] In one exemplary embodiment, the device has a voice response function and can be a home appliance device such as an air conditioner, a television, a refrigerator, a smart speaker, etc.

[0030] In one exemplary embodiment, the voice control command may be a wake-up command, a parameter setting command, or an on / off command.

[0031] In one example embodiment, the probability value policy may be proactively sent to the device by the cloud platform, or the device may request it from the cloud platform, and the cloud platform may send it to the device in response to the request.

[0032] In one exemplary embodiment, the rule that the probability value in the probability value policy satisfies is: The more times a device is determined to be a response device, the more likely the device is to continue to be a response device in the current environment, indicating a higher probability value corresponding to the device; conversely, the fewer times a device is determined to be a response device, the less likely the device is to continue to be a response device in the current environment, indicating a lower probability value corresponding to the device.

[0033] Since the probability value policy is generated based on the response status information, it can accurately reflect the usage habits of the user who determined the device as a response device during the historical usage process. Compared with the prior art that only applies the probability value under the current environment to determine that a device will be determined as a response device under the current environment, the present disclosure can better support the determination of a response device by combining the user's operation habits reflected in the probability value during historical operation, compared with a scheme based on the probability value that determines that a device will be determined as a response device under the current environment.

[0034] Step 102 is a step of determining a target S / N ratio based on the target probability value and the S / N ratio of the audio control signal obtained in advance.

[0035] Here, the S / N ratio is the ratio between the energy value of the received voice control signal and the energy value of the noise signal, and since any of the conventional methods for calculating the S / N ratio can be applied to this step, they will not be described here.

[0036] In one exemplary embodiment, the target probability value may be multiplied by the S / N ratio, and the result of the multiplication may be taken as the target S / N ratio.

[0037] By correcting the S / N ratio in the current environment using the target probability value, the obtained S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits.

[0038] Step 103 is determining whether the device is a response device for voice control signals based on the target S / N ratio.

[0039] Compared with the conventional technology that directly uses the S / N ratio in the current environment, the solution provided by the embodiments of the present disclosure makes a judgment using a target S / N ratio obtained from the target probability value and the current S / N ratio. The target S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits. Therefore, the response device determined by the target S / N ratio is a response device determined on the premise of conforming to the user's operating habits, so that the decision operation is more in line with the user's requirements and the consistency between the decision result and the user's expectations can be improved.

[0040] According to the method provided in Example 1 of the present disclosure, when a voice control signal is received, a target probability value for determining that the device is an answering device is obtained according to a preset probability value policy, a target S / N ratio is determined according to the target probability value and the S / N ratio of the received voice control signal, and whether the device is an answering device for the voice control signal is determined according to the target S / N ratio. By determining the answering device on the premise that it is in line with the user's operating habits, the decision operation can be more in line with the user's requirements, and the consistency between the decision result and the user's expectations can be improved.

[0041] Example 2

[0042] 2 is a flow chart of a method for determining a response device provided in the second embodiment of the present disclosure. As shown in FIG. 2, the method includes the following steps 201 to 205.

[0043] Step 201: When receiving a voice control signal, a target probability value for determining that the device is a response device is obtained according to a preset probability value policy, the probability value policy being determined based on response status information of the device to the voice control signal.

[0044] In one exemplary embodiment, the device has a voice response function and can be a home appliance device such as an air conditioner, a television, a refrigerator, a smart speaker, etc.

[0045] In one exemplary embodiment, the voice control command may be a wake-up command, a parameter setting command, or an on / off command.

[0046] In one example embodiment, the probability value policy may be proactively sent to the device by the cloud platform, or the device may request it from the cloud platform, and the cloud platform may send it to the device in response to the request.

[0047] In one exemplary embodiment, the rule that the probability value in the probability value policy satisfies is: The more times a device is determined to be a response device, the more likely the device is to continue to be a response device in the current environment, indicating a higher probability value corresponding to the device; conversely, the fewer times a device is determined to be a response device, the less likely the device is to continue to be a response device in the current environment, indicating a lower probability value corresponding to the device.

[0048] Since the probability value policy is generated based on the response status information, it can accurately reflect the usage habits of the user who determined the device as a response device during the historical usage process. Compared with the prior art that only applies the probability value under the current environment to determine that a device will be determined as a response device under the current environment, the present disclosure can better support the determination of a response device by combining the user's operation habits reflected in the probability value during historical operation, compared with a scheme based on the probability value that determines that a device will be determined as a response device under the current environment.

[0049] According to one exemplary embodiment, the probability value policy is a method for determining the probability value of the device corresponding to each time interval.

[0050] When determining the target probability value at which a device is determined to be a response device, the time at which the device receives this voice control signal can be obtained, and the target probability value can be obtained according to the determination method adopted for the target time interval corresponding to that time.

[0051] Here, the length of the time interval can be set as needed. Taking one hour as an example, if the reception time is 11:30, the target time interval will be 11:00 to 12:00, and the target probability value will be obtained according to the determination method corresponding to that target time interval.

[0052] By setting up a determination method in which the device's probability value policy includes different time intervals, the probability information of the same device becoming a response device within different time intervals can be reflected, the user's operating habits for the device at different times can be more accurately reflected, the magnitude of the target probability value can be more accurate, and it can help to more accurately determine the response device.

[0053] Step 202 is a step of calculating the similarity between the keyword and a preset standard word to obtain reliability information, in which the keyword is obtained from the voice control signal.

[0054] Taking the example where the voice control signal is a wake-up command, reliability information can be obtained by calculating the similarity between a keyword in the voice control signal and a standard word corresponding to the wake-up command.

[0055] Step 203 is a step of determining whether the reliability information is greater than a preset reliability threshold.

[0056] Here, the reliability threshold is determined by the magnitude of the target probability value, and the larger the target probability value, the higher the possibility that the device will be determined as a response device and the lower the reliability value; conversely, the smaller the target probability value, the lower the possibility that the device will be determined as a response device and the higher the reliability value.

[0057] For example, if the audio control signal is a wake-up command, when the target probability value is higher, the reliability value is set lower to facilitate smooth wake-up of the device in different environments, thereby increasing the success rate of waking up the device. Conversely, when the target probability value is lower, the reliability value is set higher to reduce the device being woken up due to erroneous environmental sounds, thereby appropriately controlling the success rate of waking up the device.

[0058] As can be seen from the above examples, setting a confidence threshold based on a target probability value can aid in determining that a device is determined to be a responder.

[0059] If the confidence information is greater than the confidence threshold, step 204 is executed; otherwise, the process ends.

[0060] Step 204 is determining a target S / N ratio based on the target probability value and the S / N ratio of the previously received voice control signal.

[0061] Here, the S / N ratio is the ratio between the energy value of the received voice control signal and the energy value of the noise signal, and since any of the conventional methods for calculating the S / N ratio can be applied to this step, they will not be described here.

[0062] In one exemplary embodiment, the target probability value may be multiplied by the S / N ratio, and the result of the multiplication may be taken as the target S / N ratio.

[0063] By correcting the S / N ratio in the current environment using the target probability value, the obtained S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits.

[0064] Step 205 determines whether the device is an answering device for voice control signals based on the target signal-to-noise ratio.

[0065] Compared with the conventional technology that directly uses the S / N ratio in the current environment, the solution provided by the embodiments of the present disclosure makes a judgment using a target S / N ratio obtained from the target probability value and the current S / N ratio. The target S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits. Therefore, the response device determined by the target S / N ratio is a response device determined on the premise of conforming to the user's operating habits, so that the decision operation is more in line with the user's requirements and the consistency between the decision result and the user's expectations can be improved.

[0066] According to the method provided in Example 2 of the present disclosure, when a voice control signal is received, a target probability value for determining that a device is a response device is obtained according to a preset probability value policy, a target S / N ratio is determined according to the target probability value and the S / N ratio of the received voice control signal, and whether the device is a response device for the voice control signal is determined according to the target S / N ratio. By determining the response device based on the user's operating habits, the determination operation can be more tailored to the user's requirements and the degree of consistency between the determination result and the user's expectations can be improved. By setting a determination method in which the device's probability value policy includes different time intervals, the user's operating habits for the device at different times can be more accurately reflected, the magnitude of the target probability value can be more accurate, and more accurate determination of the response device can be supported. By setting a confidence threshold based on the target probability value, the determination that the device is a response device can be supported.

[0067] Example 3

[0068] 3 is a flow chart of a method for determining a response device provided in the third embodiment of the present disclosure. As shown in FIG. 3, the method includes the following steps 301 to 308.

[0069] Step 301: When receiving a voice control signal, a target probability value for determining that the device is a response device is obtained according to a preset probability value policy, the probability value policy being determined based on response status information of the device to the voice control signal.

[0070] In one exemplary embodiment, the device has a voice response function and can be a home appliance device such as an air conditioner, a television, a refrigerator, a smart speaker, etc.

[0071] In one exemplary embodiment, the voice control command may be a wake-up command, a parameter setting command, or an on / off command.

[0072] In one example embodiment, the probability value policy may be proactively sent to the device by the cloud platform, or the device may request it from the cloud platform, and the cloud platform may send it to the device in response to the request.

[0073] In one exemplary embodiment, the rule that the probability value in the probability value policy satisfies is: The more times a device is determined to be a response device, the more likely the device is to continue to be a response device in the current environment, indicating a higher probability value corresponding to the device; conversely, the fewer times a device is determined to be a response device, the less likely the device is to continue to be a response device in the current environment, indicating a lower probability value corresponding to the device.

[0074] Since the probability value policy is generated based on the response status information, it can accurately reflect the usage habits of the user who determined the device as a response device during the historical usage process. Compared with the prior art that only applies the probability value under the current environment to determine that a device will be determined as a response device under the current environment, the present disclosure can better support the determination of a response device by combining the user's operation habits reflected in the probability value during historical operation, compared with a scheme based on the probability value that determines that a device will be determined as a response device under the current environment.

[0075] According to one exemplary embodiment, the probability value policy is a method for determining the probability value of the device corresponding to each time interval.

[0076] When determining the target probability value for determining that the device is a response device, the receiving time of this voice control signal can be obtained, and the target probability value can be obtained according to the determination method corresponding to the target time interval in which the receiving time is located.

[0077] Here, the length of the time interval can be set as needed. Taking one hour as an example, if the reception time is 11:30, the target time interval will be 11:00 to 12:00, and the target probability value will be obtained according to the determination method corresponding to that target time interval.

[0078] By setting a determination method in which the device probability value policy includes different time intervals, the probability information of the same device becoming a response device within different time intervals can be reflected, the user's operating habits for the device at different times can be more accurately reflected, the target probability value can be more accurate, and the response device can be more accurately determined.

[0079] Step 302 is a step of calculating the similarity between the keyword and a preset standard word to obtain reliability information, in which the keyword is obtained from the voice control signal.

[0080] Taking the example where the voice control signal is a wake-up command, reliability information can be obtained by calculating the similarity between a keyword in the voice control signal and a standard word corresponding to the wake-up command.

[0081] Step 303 is a step of determining whether the reliability information is greater than a preset reliability threshold.

[0082] Here, the reliability threshold is determined by the magnitude of the target probability value, and the larger the target probability value, the higher the possibility that the device will be determined as a response device and the lower the reliability value; conversely, the smaller the target probability value, the lower the possibility that the device will be determined as a response device and the higher the reliability value.

[0083] For example, if the audio control signal is a wake-up command, when the target probability value is higher, the reliability value is set lower to facilitate smooth wake-up of the device in different environments, thereby increasing the success rate of waking up the device. Conversely, when the target probability value is lower, the reliability value is set higher to reduce the device being woken up due to erroneous environmental sounds, thereby appropriately controlling the success rate of waking up the device.

[0084] If the confidence information is greater than the confidence threshold, step 303 is executed; otherwise, the process ends.

[0085] Step 304 is determining a target S / N ratio based on the target probability value and the S / N ratio of the previously received audio control signal.

[0086] Here, the S / N ratio is the ratio between the energy value of the received voice control signal and the energy value of the noise signal, and since any of the conventional methods for calculating the S / N ratio can be applied to this step, they will not be described here.

[0087] In one exemplary embodiment, the target probability value may be multiplied by the S / N ratio, and the result of the multiplication may be taken as the target S / N ratio.

[0088] By correcting the S / N ratio in the current environment using the target probability value, the obtained S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits.

[0089] Step 305 determines whether the device is a response device for voice control signals based on the target signal-to-noise ratio.

[0090] Compared with the conventional technology that directly uses the S / N ratio in the current environment, the solution provided by the embodiments of the present disclosure makes a judgment using a target S / N ratio obtained from the target probability value and the current S / N ratio. The target S / N ratio can reflect not only the magnitude of the S / N ratio but also the user's operating habits. Therefore, the response device determined by the target S / N ratio is a response device determined on the premise of conforming to the user's operating habits, so that the decision operation is more in line with the user's requirements and the consistency between the decision result and the user's expectations can be improved.

[0091] Step 306: Recording response status information, wherein the response status information includes information about the number of times the device is determined to be a responding device.

[0092] Step 307 is a step of transmitting response status information.

[0093] The trigger condition for the transmission operation may be that the transmission occurs after the time from the start of recording reaches a preset time threshold, or that the transmission occurs after receiving a request sent from the cloud platform.

[0094] Step 308 is receiving updates to the probability value policy corresponding to the response status information.

[0095] By receiving update information for the probability value policy, the probability value policy can be updated in a timely manner, which can better match the user's operating habits and further improve the consistency between the decision results and the user's expectations.

[0096] According to the method provided in the embodiments of the present disclosure, when a voice control signal is received, a target probability value for determining that a device is a response device is obtained according to a preset probability value policy, a target S / N ratio is determined according to the target probability value and the S / N ratio of the received voice control signal, and whether the device is a response device for the voice control signal is determined according to the target S / N ratio. By determining the response device based on the user's operating habits, the determination operation can be more tailored to the user's requirements and the degree of consistency between the determination result and the user's expectations can be improved. By setting a determination method in which the device's probability value policy includes different time intervals, the user's operating habits for the device at different times can be more accurately reflected, the magnitude of the target probability value can be more accurate, and more accurate determination of the response device can be supported. By setting a confidence threshold based on the target probability value, the determination of the device as a response device can be supported. By updating the probability value policy, the determination can be more tailored to the user's operating habits and the degree of consistency between the determination result and the user's expectations can be further improved.

[0097] Example 4

[0098] 4 is a flow chart of a method for determining a response device according to a fourth embodiment of the present disclosure. As shown in FIG. 4, the method includes the following steps 401 to 403.

[0099] Step 401 is a step of acquiring response status information of a device, and the response status information is a step of recording information on the number of times the device becomes a response device for a voice control signal.

[0100] The trigger condition for the acquisition operation may be one that is actively sent after the time from the start of recording on the device reaches a preset time threshold, or one that is received after an acquisition request is sent from the cloud platform.

[0101] Step 402 is a step of determining a probability value policy of the device based on the number information in the response status information.

[0102] In one exemplary embodiment, the rule that the probability value in the probability value policy satisfies is: The more times a device is determined to be a response device, the more likely the device is to continue to be a response device in the current environment, indicating a higher probability value corresponding to the device; conversely, the fewer times a device is determined to be a response device, the less likely the device is to continue to be a response device in the current environment, indicating a lower probability value corresponding to the device.

[0103] Since the probability value policy is generated based on the response status information, it can accurately reflect the usage habits of the user who determined the device as a response device during the historical usage process. Compared with the prior art that only applies the probability value under the current environment to determine that a device will be determined as a response device under the current environment, the present disclosure can better support the determination of a response device by combining the user's operation habits reflected in the probability value during historical operation, compared with a scheme based on the probability value that determines that a device will be determined as a response device under the current environment.

[0104] Step 403 is a step of transmitting a probability value policy.

[0105] By transmitting a probability value policy that reflects the user's operating habits to the device, the device can determine whether to become a response device to a voice control signal according to the probability value policy.

[0106] The method provided in Example 4 of the present disclosure obtains response status information in the historical data of the device, determines a probability value policy for the device based on the frequency information in the response status information, transmits the probability value policy, and provides data support for the decision operation to meet the user's requirements.

[0107] In the above-described fourth embodiment, the method for determining the probability value policy can be determined using the following means.

[0108] Application Scenario 1 The response status information includes information on the number of times the device was determined to be a responding device during each time interval.

[0109] Using information on the number of times a device is determined to be a response device in each time interval, a method for determining the probability value of a device being determined to be a response device in each time interval can be determined, and a probability value policy for the device can be obtained.

[0110] By setting a determination method in which the device probability value policy includes different time intervals, the probability information of the same device becoming a response device at different time intervals can be reflected, the user's operating habits for the device at different times can be more accurately reflected, the target probability value can be more accurate, and the response device can be more accurately determined.

[0111] Application scenario 2 The response status information includes: Information on the number of times the device was determined to be a responding device in each time interval; Validity information is included that indicates whether the device will be a responding device each time.

[0112] Once a device is determined to be a responding device, if the device is confirmed by the user to be the correct responding device, the validity information is valid.

[0113] If the device is determined to be an answering device due to a false wake-up caused by an environmental sound, and the user confirms that the device is a false answering device, the validity information is invalid.

[0114] Here, if a device is determined to be a response device, if it is identical to the object the user is trying to control, it means that the device is the correct response device, and if it is different from the object the user is trying to control, it means that the device is an incorrect response device.

[0115] Based on the response status information, for each time interval, the valid number Ti and / or invalid number Fi at which the device is determined to be a response device can be determined, where i is the time interval number and is a positive integer. Further, based on the valid number Ti and / or invalid number Fi corresponding to each time interval, a method for determining the probability value corresponding to each time interval is determined, and a probability value calculation policy is obtained.

[0116] Here, the probability value corresponding to the i-th time interval is determined by the following 1) to 3).

[0117] 1) All validity information is valid in the i-th time interval. The effective number of times that the device becomes a responding device in each time interval is counted, and the effective number with the largest value is selected to obtain the maximum effective number Tmax, and / or the effective number with the smallest value is selected to obtain the minimum effective number Tmin.

[0118] At least one of the minimum valid number Tmin and the maximum valid number Tmax, and the valid number Ti are calculated, and a first numerical value is obtained as a probability value corresponding to the i-th time interval.

[0119] In the above method, at least one of the minimum valid number Tmin and the maximum valid number Tmax can be used, so the following three methods are conceivable for determining the first numerical value.

[0120] In method 1, only the maximum valid number of times Tmax is used. A proportional value between the valid number of times Ti and the maximum valid number of times Tmax is calculated as a first numerical value. In method 2, only the minimum valid count Tmin is used. The difference between the effective number of times Ti and the effective number minimum value Tmin is calculated to obtain a first calculated value, and then the ratio between the first calculated value and the effective number minimum value Tmin is calculated. In method 3, the maximum valid number of times Tmax and the minimum valid number of times Tmin are used. The difference between the valid number of times Ti and the minimum valid number of times Tmin is calculated to obtain a first calculated value, and the difference between the maximum valid number of times Tmax and the minimum valid number of times Tmin is calculated to obtain a second calculated value. The first calculated value and the second calculated value can be calculated to obtain a first numerical value. Here, the first calculated value and the second calculated value can be calculated to determine the first numerical value.

[0121] 2) All validity information is invalid in the i-th time interval. The number of invalid times that the device becomes a responding device in each time interval is counted, and the number of invalid times with the largest value is selected to obtain the maximum number of invalid times Fmax, and / or the number of invalid times with the smallest value is selected to obtain the minimum number of invalid times Fmin.

[0122] At least one of the minimum invalid count Fmin and the maximum invalid count Fmax, and the invalid count Fi are calculated, and a second numerical value is obtained as a probability value corresponding to the i-th time interval.

[0123] In the above method, at least one of the minimum invalid count Fmin and the maximum invalid count Fmax can be used, so the following three methods are conceivable for determining the second numerical value.

[0124] In method 1, only the maximum invalid count Fmax is used. A proportional value between the number of invalid operations Fi and the maximum number of invalid operations Fmax is calculated as a second numerical value. In method 2, only the minimum invalid count Fmin is used. The difference between the number of invalid operations Fi and the minimum number of invalid operations Fmin is calculated to obtain a third calculation value, and then the ratio of the first calculation value to the minimum number of invalid operations Fmin is calculated. In method 3, the maximum invalid count Fmax and the minimum invalid count Fmin are used. The difference between the number of invalids Fi and the minimum number of invalids Fmin is calculated to obtain a third calculated value, and the difference between the maximum number of invalids Fmax and the minimum number of invalids Fmin is calculated to obtain a fourth calculated value. The third calculated value and the fourth calculated value can be calculated to obtain the second numerical value. Here, the third calculated value and the fourth calculated value can be calculated to determine the second numerical value.

[0125] 3) In the i-th time interval, some of the validity information is valid and other parts are invalid. Using the above method, a first numerical value corresponding to the validity information being valid and a second numerical value corresponding to the validity information being invalid are obtained.

[0126] A fourth value is calculated as a probability value of the device corresponding to the i-th time interval using the difference between the first value and the second value.

[0127] The probability value determined in the above manner is more accurate and therefore more consistent with the user's operating habits.

[0128] Example 5

[0129] 5 is a flow chart of a method for determining an answering device provided in the fifth embodiment of the present disclosure. As shown in FIG. 5, the method shown in FIG. 5 is described by taking the voice control signal as a wake-up command. The method includes the following steps 501 to 505:

[0130] In step 501, when the time that the device has been connected to the server reaches one week, the server generates a wake-up threshold adjustment table (corresponding to the above probability value policy) from the device's usage record for one week and sends it to the device.

[0131] The threshold adjustment table is obtained in the following manner. Set the time interval to 1 hour, and statistically separate the effective wake-up times (T) and ineffective wake-up times (F) of the device by hour. Obtain the maximum effective wake-up (Tmax), maximum ineffective wake-up (Fmax), minimum effective wake-up (Tmin), and minimum ineffective wake-up (Fmin) over all time intervals.

[0132] Among them, an effective wake-up indicates that an effective intention was recognized within 1 minute before and after the wake-up. An ineffective wake-up indicates that it includes cases where the ineffective wake-up was a false wake-up due to environmental sound or when the current device was wrongly judged to be woken up when the user tried to wake up another device. The determination condition is that there is no effective intention within 1 minute before and after the wake-up.

[0133] Here, the wake-up threshold condition percentage P per hour is P = (T - Tmin) / (Tmax - Tmin) - (F - Fmin) / (Fmax - Fmin) (0% < P < 100%). The content of the adjustment table is the P value per hour.

[0134] Here, the adjustable range of the threshold is the threshold range corresponding to a wake-up rate of 80% to 95%.

[0135] In step 502, after the device receives the wake-up threshold adjustment table, it sets the wake-up threshold in combination with the percentage in the adjustment table according to the preset adjustable range of the threshold, and calculates the threshold every hour.

[0136] The wake-up reliability indicates that after the wake-up model processes the input audio, it outputs the wake-up reliability, which is the similarity between the audio and the wake-up word.

[0137] The wake-up threshold is a fixed value obtained through test verification. Among them, when the wake-up reliability is higher than this value, it is determined as the hit wake-up word. Step 503: When the user wakes up the device, calculate the SNR value, which is the signal-to-noise ratio, and multiply the SNR by the current percentage in the wake-up threshold adjustment table to obtain an SNR value that suits the user's usage habits.

[0138] In step 504, if multiple devices are woken up at the same time, the values ​​calculated in step 503 are compared, and the device with the maximum value is selected and woken up.

[0139] Step 505 is a step in which the server periodically updates the wake-up threshold adjustment table.

[0140] The method provided by Example 5 of the present disclosure collects statistics on a user's hourly usage habits and adjusts the device's wake-up threshold and post-wake-up SNR value, thereby improving the user's success rate in waking up the target device, and lowering the wake-up threshold when the user uses the device less can reduce the probability of false wake-ups caused by environmental sounds.

[0141] An embodiment of the present disclosure provides a storage medium storing a computer program that, when executed, can perform the method of any one of FIGS. 1 to 3 above.

[0142] An embodiment of the present disclosure provides a storage medium storing a computer program that, when executed, can perform the method of FIG. 4 above.

[0143] An embodiment of the present disclosure provides an apparatus including a memory for storing a computer program and a processor capable of performing the method of any one of Figures 1 to 3 above when executing the computer program in the memory.

[0144] The processor in the above device may be located within the device as an independent module or may be integrated within the device's processor and is used to determine whether the device is an answering device when it receives a voice control signal.

[0145] An embodiment of the present disclosure provides a server including a memory for storing a computer program and a processor capable of performing the method of FIG. 4 above when executing the computer program in the memory.

[0146] The processor in the server may be located in the device as an independent module or integrated within the server's processor and is used to generate a probability value policy for the device so that the device determines a responding device according to the probability value policy.

[0147] The above is merely a preferred embodiment of the present disclosure and does not limit the scope of the claims of the present disclosure. Equivalent structural modifications made using the contents of the present disclosure specification and the attached drawings in the inventive concept of the present disclosure, or those directly or indirectly applied in other related technical fields, are also included in the scope of the claims of the present disclosure.

Claims

1. A method for determining a response device when multiple devices with voice response functions exist in the same space, comprising: After receiving the voice control signal, obtaining a target probability value of the device becoming a responding device according to a preset probability value policy, the probability value policy being determined based on response status information of the device to the voice control signal; determining a target signal-to-noise ratio based on the target probability value and a signal-to-noise ratio of the received audio control signal; determining whether the device is a responsive device to the audio control signal based on whether the target S / N ratio of the device is greater than the target S / N ratios of other devices in the same space; The response status information is a history of determining the device as a response device. A method for determining a response device, comprising:

2. The probability value policy includes a method for determining a probability value corresponding to each time interval; The step of obtaining a target probability value for the device to be a responder comprises: obtaining a reception time of the audio control signal; 2. The method of claim 1, further comprising: determining the target probability value according to a determination scheme in which the reception time corresponds to a time interval in which the reception time is located.

3. The method further comprises: calculating a similarity between the keyword in the voice control signal and a predetermined standard word to obtain reliability information; The method for determining a response device according to claim 1, further comprising a step of not allowing a device to determine whether it will be a response device for the voice control signal if the reliability information is smaller than a predetermined reliability threshold, the reliability threshold being determined by the magnitude of a target probability value.

4. The method further comprises: recording the number of times the device becomes a responding device and obtaining response status information; sending response status information; and receiving update information of a probability value policy corresponding to the response status information.

5. A method for determining and transmitting a probability value policy used to determine a response device when multiple devices having a voice response function exist in the same space, comprising: obtaining response status information of the device in response to a voice control signal, the response status information including the number of times the device has been determined to be a response device; determining a probability value policy for the device based on the number of times in the response status information; and transmitting the probability value policy.

6. The probability value policy determination and transmission method described in claim 5, characterized in that when the response status information includes the number of times it has been determined as a response device in each time interval, the probability value policy includes a determination method corresponding to each time interval.

7. When the response status information further includes validity information indicating that the device will be a response device every time, the determination method for each time interval is determined based on the valid number Ti and / or invalid number Fi in each time interval, where i is a positive integer representing the number of the time interval.

8. a method of obtaining a maximum valid count value and / or a minimum valid count value from the valid counts of all time intervals when all validity information is valid, calculating at least one of the maximum valid count value and the minimum valid count value and the valid count Ti, and obtaining a first numerical value as a probability value of the i-th time interval; If all the validity information is invalid, a maximum value of the invalid count and / or a minimum value of the invalid count are obtained from the invalid counts of all the time intervals, and at least one of the maximum value of the invalid count and the minimum value of the invalid count and the invalid count Fi are calculated, and a second value is obtained as a probability value of the i-th time interval; a method of determining a first value corresponding to the number of valid times of the device and a second value corresponding to the number of invalid times of the device when a part of the validity information is valid and another part is invalid, and obtaining a third value as a probability value of the i-th time interval based on a difference between the first value and the second value; 8. The method for determining and transmitting a probability value policy according to claim 7, wherein the probability value of the i-th time interval is determined by one or more of the following methods:

9. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any one of claims 1 to 8.

10. A computer program product for causing a processor to carry out the method according to any one of claims 1 to 8.

11. 1. An apparatus comprising: a memory adapted to store a computer program according to claim 10; a processor adapted to execute said computer program.

12. a server, a memory adapted to store a computer program according to claim 10; a processor adapted to execute said computer program.

Images