CN111949323B - Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium - Google Patents
Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium Download PDFInfo
- Publication number
- CN111949323B CN111949323B CN202010898830.6A CN202010898830A CN111949323B CN 111949323 B CN111949323 B CN 111949323B CN 202010898830 A CN202010898830 A CN 202010898830A CN 111949323 B CN111949323 B CN 111949323B
- Authority
- CN
- China
- Prior art keywords
- wake
- rate
- target
- threshold
- current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000005457 optimization Methods 0.000 title claims abstract description 21
- 230000002618 waking effect Effects 0.000 title claims description 18
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/4401—Bootstrapping
- G06F9/4418—Suspend and resume; Hibernate and awake
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Electric Clocks (AREA)
Abstract
The application discloses an intelligent equipment awakening optimization method, an intelligent equipment awakening optimization device, an intelligent equipment and a storage medium, wherein the intelligent equipment awakening optimization method is applied to intelligent equipment and comprises the following steps: acquiring the current wake-up rate of a target user in a preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user in the total wake-up times under the condition of being based on a standard wake-up threshold value; when the current wake-up rate of the target user is determined to exceed the preset wake-up rate range, the standard wake-up threshold is adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold, so that the current wake-up rate of the target user based on the target wake-up threshold is in the preset wake-up rate range. The method can effectively adjust the standard wake-up threshold of the target user according to the current wake-up rate of the target user, enable the wake-up rates of different users to be consistent, and improve user experience.
Description
Technical Field
The present application relates to the field of speech recognition technologies, and in particular, to an optimization method and apparatus for waking up an intelligent device, and a storage medium.
Background
The voice wake-up achieves the purpose of waking up the intelligent equipment by understanding voice data of a user, so that a first gate leading to artificial intelligence by human beings is opened, and the voice wake-up device is widely applied to various intelligent terminals, such as intelligent wearing equipment, mobile phones, tablet computers, intelligent household appliances and the like.
When traditional voice wakes, a fixed wake-up threshold is set for the intelligent device, when the similarity between the audio frequency and the preset wake-up word when the user wakes up the intelligent device is larger than the wake-up threshold, the intelligent device can be waken, and the similarity between the audio frequency and the preset wake-up word for users with different ages or different accents can be different, so that the wake-up success rates of different users can be different, and the experience of the users is further affected.
Disclosure of Invention
In view of the above problems, the application provides an optimization method and device for waking up an intelligent device, the intelligent device and a storage medium, which can adjust the standard waking up threshold of a target user according to the current waking up rate of the target user, enable the waking up rates of different users to be consistent, and promote the user experience.
In a first aspect, an embodiment of the present application provides an optimization method for waking up an intelligent device, which is applied to the intelligent device, where the optimization method for waking up the intelligent device includes: acquiring the current wake-up rate of a target user in a preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user in the total wake-up times under the condition of being based on a standard wake-up threshold value; when the current wake-up rate of the target user is determined to exceed the preset wake-up rate range, the standard wake-up threshold is adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold, so that the current wake-up rate of the target user based on the target wake-up threshold is in the preset wake-up rate range.
In a second aspect, an embodiment of the present application provides an apparatus for optimizing wake-up of an intelligent device, where the apparatus for optimizing wake-up of an intelligent device includes: the current wake-up rate acquisition module is used for acquiring the current wake-up rate of the target user in a preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user based on a standard wake-up threshold value to the total wake-up times; and the adjustment updating module is used for adjusting the standard wake-up threshold value when the current wake-up rate of the target user exceeds the preset wake-up rate range, and updating the adjusted standard wake-up threshold value into the target wake-up threshold value, so that the current wake-up rate of the target user based on the target wake-up threshold value is in the preset wake-up rate range.
In a third aspect, an embodiment of the present application provides an intelligent device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the method for optimizing wake-up of the intelligent device provided in the first aspect when the processor executes the computer program.
In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, where a program code is stored, where the program code may be invoked by a processor to perform an optimization method for waking up a smart device as provided in the first aspect above.
According to the scheme provided by the application, the intelligent device acquires the current wake-up rate of the target user in the preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user based on the standard wake-up threshold value in the total wake-up times, when the current wake-up rate of the target user exceeds the preset wake-up rate range, the standard wake-up threshold value is adjusted, and the adjusted standard wake-up threshold value is updated to be the target wake-up threshold value, so that the current wake-up rate of the target user based on the target wake-up threshold value is in the preset wake-up rate range, and therefore, the standard wake-up threshold value of the target user is adjusted according to the current wake-up rate of the target user, the wake-up rates of different users tend to be consistent, and user experience is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following description will briefly explain the drawings needed in the description of the embodiments, and it is obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an optimization method for waking up an intelligent device according to an embodiment of the present application;
fig. 2 is a schematic flow chart of an optimization method for waking up an intelligent device according to another embodiment of the present application;
fig. 3 is a schematic structural diagram of an optimization device for waking up an intelligent device according to an embodiment of the present application;
fig. 4 shows a block diagram of an intelligent device according to an embodiment of the present application;
Fig. 5 shows a memory unit for storing or carrying program code for implementing an optimization method for smart device wake-up according to an embodiment of the present application.
Detailed Description
In order to enable those skilled in the art to better understand the present application, the following description will make clear and complete descriptions of the technical solutions according to the embodiments of the present application with reference to the accompanying drawings.
Speech recognition technology has made remarkable progress in recent years, and the technology has entered various fields such as industry, home appliances, smart home and the like. Voice wake-up is a form of voice recognition technology that can wake up a device to run by voice without directly contacting the hardware device. The voice wake-up is to make the intelligent device (such as an intelligent sound box, an intelligent mobile phone, an intelligent toy, an intelligent household appliance and the like) detect the voice (set voice command, i.e. wake-up word) of the user in the sleep state or the screen locking state, so that the intelligent device in the sleep state directly enters the waiting command state, and the first step of voice interaction is started. In general, most intelligent devices are awakened or operated by physical keys. However, this is not good for the user experience. Voice is used as the most natural communication mode of people, and the non-contact mode of waking up the intelligent equipment through voice is definitely more friendly.
The scheme in the traditional voice awakening technology generally adopts a mode of presetting an awakening threshold value, when the similarity between the audio frequency and a preset awakening word when a user awakens the intelligent device is larger than the awakening threshold value, the intelligent device can be awakened, and the similarity between the audio frequency and the preset awakening word for users with different ages or different accents can be different, so that the awakening success rates of different users can be different, and the user experience is further influenced.
Aiming at the problems, the inventor researches and puts forward the optimization method, the device, the intelligent device and the storage medium for the wake-up of the intelligent device for a long time, the intelligent device is used for acquiring the current wake-up rate of the target user in a preset time period, when the current wake-up rate of the target user is determined to exceed the preset wake-up rate range, the standard wake-up threshold is automatically adjusted, the adjusted standard threshold is updated to the target wake-up threshold, so that the wake-up rates of different users tend to be consistent, and the user experience is improved.
Referring to fig. 1, which is a flowchart illustrating a method for optimizing wake-up of an intelligent device according to an embodiment of the present application, the flow illustrated in fig. 1 will be described in detail, and the method for optimizing wake-up of an intelligent device may include the following steps:
step S110: and acquiring the current wake-up rate of the target user in a preset time period.
In the embodiment of the application, the intelligent device can acquire the current wake-up rate of the target user in the preset time period based on the current wake-up threshold value so as to determine whether the current wake-up rate in the preset time period exceeds the preset wake-up rate range or not, and further determine whether the wake-up threshold value needs to be adjusted or not. The current wake-up rate is the proportion of the number of wake-up success times in the total wake-up times under the condition that the target user is based on the current wake-up threshold value; the current wake-up threshold value can be a standard wake-up threshold value preset by a system, or can be a target wake-up threshold value adjusted according to the specific wake-up success rate of the target user; the preset time period may be the previous day, the previous week, the previous month, the previous year, or the like, which is not limited herein.
In some embodiments, in order to obtain the current wake-up rate of the target user in the preset time period, the intelligent device may obtain a similarity corresponding to each wake-up audio in a plurality of wake-up audios of the target user in the preset time period, where the similarity is a similarity value between the wake-up audio and a standard wake-up audio, compare the plurality of similarities with a standard wake-up threshold, determine the number of wake-up audios with the similarity greater than or equal to the standard wake-up threshold, and then calculate a proportion of the number of wake-up audios with the similarity greater than or equal to the standard wake-up threshold to the total number of the plurality of wake-up audios, so as to obtain the current wake-up rate. For example, the standard wake-up threshold may be 0.4, the smart device may acquire that, in 100 wake-up audios of the last week of the target user, the wake-up audio with a similarity of 0.5 is 5, the wake-up audio with a similarity of 0.45 is 10, the wake-up audio with a similarity of 0.4 is 70, the wake-up audio with a similarity of 0.35 is 10, the wake-up audio with a similarity of 0.3 is 5, and the smart device may determine that the number of wake-up audios greater than or equal to the standard wake-up threshold of 0.4 is 85, and then calculate to obtain the current wake-up rate as 85%, where the standard wake-up threshold may also be 0.7, may also be 0.3, may also be 0.2, and so on, and is not limited herein.
As an implementation mode, when the intelligent device detects the wake-up audio of the target user, the intelligent device can acquire voiceprint information in the wake-up audio, and acquire current wake-up rates of different users according to the voiceprint information, so that the current wake-up rate corresponding to the target user can be determined.
In some embodiments, the smart device may acquire the current wake-up rate within a preset period of time before the current time every a preset period, which means that the smart device may periodically acquire the current wake-up rate within the preset period of time, so that the smart device periodically adjusts the wake-up threshold of the smart device according to the acquired current wake-up rate. It will be appreciated that the preset period may be one day, one week, one month, etc., and is not limited herein. Taking a preset time period as an example of one week, the number of successful awakenings and the total number of awakenings in the previous week can be acquired in the early morning every day, and the current awakening rate is obtained according to the proportion of the acquired number of successful awakenings to the total number of awakenings, and if the time period before the current time period cannot meet the preset time period (namely, less than one week), the current awakening rate can be obtained according to the actually acquired number of successful awakenings and the total number of awakenings.
It can be appreciated that when the smart device is first used, the smart device can only wake up the threshold value with a preset standard as a criterion for determining the similarity of wake-up audio of the user.
Step S120: when the current wake-up rate of the target user is determined to be beyond the preset wake-up rate range, the standard wake-up threshold is adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold.
In the embodiment of the application, after the intelligent device acquires the current wake-up rate of the target user in the preset time period, the intelligent device can determine whether the current wake-up rate of the target user exceeds the preset wake-up rate range. When the current wake-up rate of the target user is determined to be beyond the preset wake-up rate range, the standard wake-up threshold can be adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold, so that the current wake-up rate of the target user based on the target wake-up threshold is within the preset wake-up rate range. The preset wake-up rate range may be a range value or a set value, which is not limited herein.
In some embodiments, the intelligent device may acquire a plurality of similarities of multiple wake-up audios of the target user in a preset time period, where the plurality of similarities are in one-to-one correspondence with the multiple wake-up audios, the similarities are similarity values of the wake-up audios and the standard audio, and then determine a target wake-up threshold based on the similarities corresponding to the multiple wake-up audios, where a proportion of a number of the similarities reaching the target wake-up threshold in the plurality of similarities to a total number of the plurality of similarities is within a preset proportion range, then update the standard wake-up threshold to the target wake-up threshold, where the target wake-up threshold is a criterion for judging whether the wake-up audio of the target user by the intelligent device meets a wake-up condition or not, and is used for executing wake-up on the intelligent device when the similarity between the audio data of the target user received by the intelligent device and the standard audio is greater than or equal to the target wake-up threshold. The preset ratio range may be a ratio range or a ratio value, which is not limited herein.
In one embodiment, in order to determine a target wake-up threshold of the intelligent device, the intelligent device may determine a plurality of kinds of similarities which occur in the plurality of kinds of similarities, respectively use each kind of similarity in the plurality of kinds of similarities as a first specified wake-up threshold, calculate a first proportion of a number of similarities reaching the first specified wake-up threshold in the plurality of kinds of similarities to a total number of the plurality of kinds of similarities, obtain a plurality of first proportions, and then obtain, from the plurality of kinds of similarities, a similarity corresponding to a first target proportion within a preset proportion range, as the target wake-up threshold.
In this embodiment, when the degree of similarity corresponding to the first target proportion within the preset proportion range obtained from the multiple degrees of similarity is multiple, the intelligent device may select the maximum degree of similarity from the degrees of similarity corresponding to the multiple first target proportions, as the target wake-up threshold.
For example, the preset scale range may be a scale value greater than or equal to 95%, when the smart device detects that among 100 wake-up audios similar to the target user 1 in the last week, the wake-up audio with a similarity of 0.6 is 2 times, the wake-up audio with a similarity of 0.55 is 4 times, the wake-up audio with a similarity of 0.5 is 3 times, the wake-up audio with a similarity of 0.45 is 6 times, the wake-up audio with a similarity of 0.4 is 35 times, the wake-up audio with a similarity of 0.35 is 45 times, and the wake-up audio with a similarity of 0.3 is 5 times, the smart device may respectively have a similarity of 0.6, 0.55, 0.5, 0.45, 0.4, 0.35 and 0.3 as a first specified wake-up threshold, respectively calculated as: the first proportion of the similarity reaching 0.6 is 2%, the first proportion of the similarity reaching 0.55 is 6%, the first proportion of the similarity reaching 0.5 is 9%, the first proportion of the similarity reaching 0.45 is 15%, the first proportion of the similarity reaching 0.4 is 50%, the first proportion of the similarity reaching 0.35 is 95%, the first proportion of the similarity reaching 0.3 is 100%, then the similarity corresponding to the first target proportion of more than or equal to 95% is selected from the above-mentioned various similarities to obtain 0.3 and 0.35 respectively, then selecting the maximum similarity 0.35 from the similarity 0.3 and the similarity 0.35 as a target wake-up threshold; when the smart device detects that among 100 similar wake-up tones of the target user 2 in the previous week, the wake-up tone with the similarity of 0.6 is 35 times, the wake-up tone with the similarity of 0.55 is 25 times, the wake-up tone with the similarity of 0.5 is 20 times, the wake-up tone with the similarity of 0.45 is 15 times, the wake-up tone with the similarity of 0.4 is 1 time, the wake-up tone with the similarity of 0.35 is 4 times, and the wake-up tone with the similarity of 0.3 is 0 times, the smart device may respectively take the similarities of 0.6, 0.55, 0.5, 0.45, 0.4, 0.35 and 0.3 as the first specified wake-up threshold, and respectively calculating to obtain: the first proportion of the similarity reaching 0.6 is 35%, the first proportion of the similarity reaching 0.55 is 60%, the first proportion of the similarity reaching 0.5 is 80%, the first proportion of the similarity reaching 0.45 is 95%, the first proportion of the similarity reaching 0.4 is 96%, the first proportion of the similarity reaching 0.35 is 100%, the first proportion of the similarity reaching 0.3 is 100%, then the similarity corresponding to a first target proportion of more than or equal to 95% is selected from the above-mentioned various similarities to obtain 0.3, 0.35, 0.4 and 0.45 respectively, and then the first proportion of the similarity reaching 0.3, Selecting the maximum similarity 0.45 from the similarity 0.35, the similarity 0.4 and the similarity 0.45 as a target wake-up threshold; for another example, the preset proportion range may be 94% -96%, the intelligent device selects, from multiple similarities of the target user 1, a similarity 0.35 corresponding to a first proportion of 95% within the range of 94% -96% as the target wake-up threshold, and the intelligent device selects, from multiple similarities of the target user 2, a similarity 0.45 corresponding to a first proportion of 95% within the range of 94% -96% and a similarity 0.4 corresponding to a first proportion of 96%, and then selects, from the similarities 0.45 and the similarities 0.4, the largest similarity 0.45 as the target wake-up threshold.
As another embodiment, in order to determine the target wake-up threshold of the smart device, the smart device may obtain a plurality of preset second specified wake-up thresholds, calculate a second proportion of the number of the similarities reaching each second specified threshold to the total number of the similarities in the plurality of similarities, obtain a plurality of second proportions, and then obtain, as the target wake-up threshold, a second specified wake-up threshold corresponding to a second target proportion within a preset proportion range from the plurality of second specified wake-up thresholds.
In this embodiment, when the second designated wake-up threshold corresponding to the second target proportion within the preset proportion range obtained from the plurality of second designated wake-up thresholds is a plurality of, the smart device may select the largest second designated wake-up threshold from the second designated wake-up thresholds corresponding to the plurality of second target proportions as the target wake-up threshold.
For example, the predetermined scale range may be a scale value. In one embodiment, the ratio value is 95%, and the plurality of second specified wake-up thresholds preset by the intelligent device may be respectively: 0.55, 0.5, 0.45, 0.4, 0.35 and 0.3, when the smart device detects 100 similar wake-up audios of the target user 3 in the last week, the wake-up audio with the similarity of 0.55 is 6 times, the wake-up audio with the similarity of 0.5 is 3 times, the wake-up audio with the similarity of 0.45 is 6 times, the wake-up audio with the similarity of 0.4 is 35 times, the wake-up audio with the similarity of 0.35 is 46 times, and the wake-up audio with the similarity of 0.3 is 4 times, the intelligent device can respectively calculate: a second proportion of 0.55 to 6%, a second proportion of 0.5 to 9%, a second proportion of 0.45 to 15%, a second proportion of 0.4 to 50%, a second proportion of 0.35 to 96%, a second proportion of 0.3 to 100%, selecting a second designated wake-up threshold value corresponding to a second target proportion of 95% or more from the plurality of second designated wake-up thresholds, obtaining 0.3 and 0.35 of similarity respectively, selecting the largest similarity 0.35 from 0.3 and 0.35 of similarity, as a target wake-up threshold; when the intelligent device detects that the 100 similar wake-up audios of the target user 4 in the last week are 60 wake-up audios with the similarity of 0.55, 20 wake-up audios with the similarity of 0.5, 15 wake-up audios with the similarity of 0.45, 1 wake-up audio with the similarity of 0.4, 4 wake-up audios with the similarity of 0.35 and 0 wake-up audio with the similarity of 0.3, the intelligent device may respectively calculate to obtain the following second specified wake-up thresholds with the similarity of 0.55, 0.5, 0.45, 0.4, 0.35 and 0.3: the second ratio of similarity to 0.55 is 60%, a second proportion of 0.5 to 80%, a second proportion of 0.45 to 95%, a second proportion of 0.4 to 96%, a second proportion of 0.35 to 100%, a second proportion of 0.3 to 100%, selecting a second designated wake-up threshold value corresponding to a second target proportion of 95% or more from the plurality of second designated wake-up thresholds to obtain 0.3, 0.35, 0.4, 0.45, respectively, selecting a largest second designated wake-up threshold value 0.45 from the second designated wake-up threshold value 0.3, the second designated wake-up threshold value 0.35, the second designated wake-up threshold value 0.4 and the second designated wake-up threshold value 0.45, as a target wake-up threshold; for another example, the preset ratio range may be 94% -96%, the intelligent device selects a second designated wake-up threshold value 0.35 corresponding to a second ratio of 96% in the range of 94% -96% from the plurality of second designated wake-up thresholds of the target user 3 as the target wake-up threshold value, and the intelligent device selects a second designated wake-up threshold value 0.45 corresponding to a second ratio of 95% in the range of 94% -96% and a second designated wake-up threshold value 0.4 corresponding to a second ratio of 96% from the plurality of second designated wake-up thresholds of the target user 4, and then selects a second set-up threshold value from the second designated wake-up threshold values 0.4 and the second designated wake-up threshold value 0.45, and selecting the largest second designated wake-up threshold value of 0.45 as a target wake-up threshold value. The number and specific values of the second plurality of specified wake-up thresholds are not limited herein.
According to the scheme provided by the application, the intelligent device acquires the current wake-up rate of the target user in the preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user based on the standard wake-up threshold value in the total wake-up times, when the current wake-up rate of the target user exceeds the preset wake-up rate range, the standard wake-up threshold value is adjusted, and the adjusted standard wake-up threshold value is updated to be the target wake-up threshold value, so that the current wake-up rate of the target user based on the target wake-up threshold value is in the preset wake-up rate range, and therefore, the standard wake-up threshold value of the target user is adjusted according to the current wake-up rate of the target user, the wake-up rates of different users tend to be consistent, and user experience is improved.
Referring to fig. 2, a flowchart of an intelligent device wake-up optimization method according to another embodiment of the present application is shown, where the intelligent device wake-up optimization method may be applied to an intelligent device, and the flowchart shown in fig. 2 will be described in detail, where the intelligent device wake-up optimization method may include the following steps:
step S210: and acquiring the current wake-up rate of the target user in a preset time period.
In the embodiment of the present application, step S210 may refer to the content of the foregoing embodiment, and is not described herein.
Step S220: and determining whether the current wake-up rate of the target user exceeds a preset wake-up rate range.
In the embodiment of the application, after the intelligent device acquires the current wake-up rate of the target user in the preset time period, whether the current wake-up rate of the target user exceeds the preset wake-up rate range can be determined, so that the intelligent device can adjust the wake-up threshold of the intelligent device when the current wake-up rate of the target user exceeds the preset wake-up rate range.
In some embodiments, the intelligent device may compare the current wake-up rate corresponding to the target user with a maximum preset wake-up rate and a minimum preset wake-up rate of the preset wake-up rate range, so as to determine whether the current wake-up rate exceeds the preset wake-up rate range according to the current wake-up rate, the maximum preset wake-up rate and the minimum preset wake-up rate.
If the current wake-up rate is smaller than the minimum preset wake-up rate, determining that the current wake-up rate exceeds a preset wake-up rate range; if the current wake-up rate is greater than the maximum preset wake-up rate, determining that the current wake-up rate exceeds a preset wake-up rate range; if the current wake-up rate is greater than or equal to the minimum preset wake-up rate and the current wake-up rate is less than or equal to the maximum preset wake-up rate, determining that the current wake-up rate does not exceed the preset wake-up rate range.
Step S230: when the current wake-up rate of the target user is determined to be beyond the preset wake-up rate range, the standard wake-up threshold is adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold.
In the embodiment of the present application, step S230 may refer to the content of the foregoing embodiment, and will not be described herein.
Step S240: and identifying the detected wake-up audio based on the updated target wake-up threshold.
In the embodiment of the application, after updating the adjusted standard wake-up threshold value to the target wake-up threshold value, the intelligent device can identify the detected wake-up audio based on the updated target wake-up threshold value, so that the intelligent device can acquire the similarity of the wake-up audio.
In some embodiments, the intelligent device pre-stores at least one user's voiceprint information, and the pre-storing mode may be input by the user when the user uses the voiceprint information for the first time or may be self-stored by the intelligent device. It can be understood that the wake-up audio includes corresponding voiceprint information, which is briefly described as the voiceprint information corresponding to the wake-up audio. The intelligent device can acquire voiceprint information corresponding to the wake-up audio according to the detected wake-up audio, determine whether a user corresponding to the wake-up audio is a target user according to the prestored voiceprint information and the voiceprint information corresponding to the wake-up audio, and acquire similarity corresponding to the wake-up audio when determining that the user corresponding to the wake-up audio is the target user.
As an implementation manner, the intelligent device may acquire matching degree of voiceprint information corresponding to the wake-up audio and prestored voiceprint information, and determine whether the user corresponding to the wake-up audio is a target user according to the matching degree and a matching degree threshold. When the matching degree is greater than or equal to the matching degree threshold value, determining the user corresponding to the wake-up audio as a target user; and when the matching degree is smaller than the matching degree threshold value, determining that the user corresponding to the wake-up audio is not the target user.
Step S250: and when the similarity of the wake-up audios reaches a target wake-up threshold value, executing wake-up on the intelligent device.
In the embodiment of the application, after the intelligent device identifies the detected wake-up audio based on the updated target wake-up threshold, whether to execute wake-up on the intelligent device can be determined according to whether the acquired similarity corresponding to the wake-up audio reaches the target wake-up threshold.
In some embodiments, when the similarity of the wake-up audio reaches a target wake-up threshold, the intelligent device performs wake-up on the intelligent device; when the similarity of the wake-up audio frequency does not reach the target wake-up threshold value, the intelligent equipment does not execute wake-up on the intelligent equipment; therefore, the method and the device realize that the intelligent equipment is only awakened when the similarity of the awakening audios reaches the target awakening threshold value, so that the false awakening rate of the intelligent equipment can be reduced, and the user experience is improved.
According to the scheme provided by the application, the intelligent device acquires the current wake-up rate of the target user in the preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user based on the standard wake-up threshold value in the total wake-up times, when the current wake-up rate of the target user exceeds the preset wake-up rate range, the standard wake-up threshold value is adjusted, and the adjusted standard wake-up threshold value is updated to be the target wake-up threshold value, so that the current wake-up rate of the target user based on the target wake-up threshold value is in the preset wake-up rate range, and therefore, the standard wake-up threshold value of the target user is adjusted according to the current wake-up rate of the target user, the wake-up rates of different users tend to be consistent, and user experience is improved. And the detected wake-up audio is identified based on the updated target wake-up threshold, when the similarity of the wake-up audio reaches the target wake-up threshold, the wake-up is executed on the intelligent device, so that the wake-up is executed on the intelligent device only when the similarity of the wake-up audio reaches the target wake-up threshold, the false wake-up rate of the intelligent device can be reduced, and the user experience is improved.
Referring to fig. 3, which illustrates a device diagram of an apparatus 300 for optimizing wake-up of a smart device according to an embodiment of the present application, the apparatus 300 for optimizing wake-up of a smart device may be applied to a smart device, and the detailed description will be made with respect to the device diagram illustrated in fig. 3, where the apparatus 300 for optimizing wake-up of a smart device may include:
The current wake-up rate obtaining module 310 is configured to obtain a current wake-up rate of the target user in a preset time period, where the current wake-up rate is a proportion of wake-up success times of the target user based on a standard wake-up threshold in a total wake-up times;
The adjustment updating module 320 is configured to adjust the standard wake-up threshold when it is determined that the current wake-up rate of the target user exceeds the preset wake-up rate range, and update the adjusted standard wake-up threshold to the target wake-up threshold, so that the current wake-up rate of the target user based on the target wake-up threshold is within the preset wake-up rate range.
In some implementations, the adjustment update module 320 can include:
The first acquisition unit is used for acquiring a plurality of similarities of the multiple awakening audios of the target user in a preset time period, wherein the similarities are in one-to-one correspondence with the multiple awakening audios, and the similarities are the similarities of the awakening audios and the standard audios;
the first determining unit is used for determining a target wake-up threshold value based on the similarities corresponding to the multiple wake-up audios, wherein the proportion of the number of the similarities reaching the target wake-up threshold value in the multiple similarities to the total number of the multiple similarities is in a preset proportion range;
And the updating unit is used for updating the standard wake-up threshold value into a target wake-up threshold value, and the target wake-up threshold value is used for executing wake-up on the intelligent device when the similarity between the received audio data of the target user and the standard audio is greater than or equal to the target wake-up threshold value.
In some implementations, the current wake-up rate acquisition module 310 may include:
The second acquisition unit is used for acquiring a plurality of wake-up audios of the target user and the similarity corresponding to the wake-up audios, wherein the similarity is a similarity value of the wake-up audios and the standard wake-up audios;
The second determining unit is used for comparing the multiple similarities with a standard awakening threshold value and determining the number of awakening audios with the similarities being greater than or equal to the standard awakening threshold value;
and the obtaining unit is used for calculating the proportion of the number of the awakening audios with the similarity larger than the standard awakening threshold value to the total number of the awakening audios, and obtaining the current awakening rate.
As an embodiment, the current wake-up rate acquisition module 310 may further include:
The third acquisition unit is used for acquiring voiceprint information in the wake-up audio when the wake-up audio of the target user is detected;
And the third determining unit is used for determining the current wake-up rate corresponding to the target user based on the voiceprint information.
In some implementations, the adjustment update module 320 may further include:
The comparison unit is used for respectively comparing the current wake-up rate corresponding to the target user with the maximum preset wake-up rate and the minimum preset wake-up rate of the preset wake-up rate range;
A fourth determining unit, configured to determine that the current wake-up rate exceeds the preset wake-up rate range when the current wake-up rate is less than the minimum preset wake-up rate;
and a fifth determining unit, configured to determine that the current wake-up rate exceeds the preset wake-up rate range when the current wake-up rate is greater than the maximum preset wake-up rate.
In some embodiments, the optimizing apparatus 300 for smart device wakeup may further include:
The identifying module is configured to identify the detected wake-up audio based on the updated target wake-up threshold after the standard wake-up threshold is adjusted by the adjusting and updating module 320 and the adjusted standard wake-up threshold is updated to the target wake-up threshold;
And the execution wake-up module is used for executing wake-up on the intelligent device when the similarity of the wake-up audios reaches a target wake-up threshold value.
In some implementations, the current wake-up rate acquisition module 310 may further include:
And the fourth acquisition unit is used for acquiring the current wake-up rate in a preset time period before the current time at intervals of a preset period.
According to the scheme provided by the application, the intelligent device acquires the current wake-up rate of the target user in the preset time period, wherein the current wake-up rate is the proportion of the wake-up success times of the target user based on the standard wake-up threshold value in the total wake-up times, when the current wake-up rate of the target user exceeds the preset wake-up rate range, the standard wake-up threshold value is adjusted, and the adjusted standard wake-up threshold value is updated to be the target wake-up threshold value, so that the current wake-up rate of the target user based on the target wake-up threshold value is in the preset wake-up rate range, and therefore, the standard wake-up threshold value of the target user is adjusted according to the current wake-up rate of the target user, the wake-up rates of different users tend to be consistent, and user experience is improved.
Referring to fig. 4, a smart device 400 according to an embodiment of the present application is shown, where the smart device 400 may perform the foregoing optimization method for waking up the smart device, and the smart device 400 includes one or more (only one is shown in the figure) processors 410 and a memory 420 that are coupled to each other. The memory 420 stores therein a program capable of executing the contents of the foregoing embodiments, and the processor 410 may execute the program stored in the memory 420, and the memory 420 includes the apparatus described in the foregoing embodiments.
Processor 410 may include one or more processing cores. The processor 410 utilizes various interfaces and lines to connect various portions of the overall smart device 400, perform various functions of the smart device 400, and process data by executing or executing instructions, programs, code sets, or instruction sets stored in the memory 420, and invoking data stored in the memory 420. Alternatively, the processor 410 may be implemented in hardware in at least one of digital signal Processing (DIGITAL SIGNAL Processing, DSP), field-Programmable gate array (Field-Programmable GATE ARRAY, FPGA), programmable logic array (Programmable Logic Array, PLA). The processor 410 may integrate one or a combination of several of a central processing unit (Central Processing Unit, CPU), an image processor (Graphics Processing Unit, GPU), and a modem, etc. The CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for being responsible for rendering and drawing of display content; the modem is used to handle wireless communications. It will be appreciated that the modem may not be integrated into the processor 410 and may be implemented solely by a single communication chip.
Memory 420 may include random access Memory (Random Access Memory, RAM) or Read-Only Memory (ROM). Memory 420 may be used to store instructions, programs, code sets, or instruction sets. The memory 420 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a current wake rate, a standard wake threshold, a number of wake successes, a total number of wake ups, a preset wake rate range, a target wake threshold, etc.), instructions for implementing various method embodiments described below, and the like. The storage data area may also store data created by the smart device 400 in use (e.g., multiple similarities, multiple wake-up tones, similarities, voiceprint information, maximum preset wake-up rate, minimum preset wake-up rate), etc.
It will be appreciated by those skilled in the art that the configuration shown in fig. 4 is merely illustrative, and is not intended to limit the configuration of the smart device. For example, the smart device 400 may also include more or fewer components than shown in fig. 4, or have a different configuration than shown in fig. 4.
Referring now to FIG. 5, a computer readable storage medium is shown, in which a program code is stored 500, and which is adapted to be invoked by a processor to perform the method described in the foregoing method embodiments, according to one embodiment of the present application.
The computer readable storage medium 500 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer readable storage medium 500 comprises a non-volatile computer readable medium (non-transitory computer-readable storage medium). The computer readable storage medium 500 has storage space for program code 510 that performs any of the method steps described above. The program code can be read from or written to one or more computer program products. Program code 510 may be compressed, for example, in a suitable form.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting; although the application has been described in detail with reference to the foregoing embodiments, it will be appreciated by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not drive the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present application.
Claims (9)
1. An optimization method for waking up an intelligent device, which is applied to the intelligent device, the method comprising:
acquiring the current wake-up rate of a target user in a preset time period, wherein the current wake-up rate is the proportion of the number of wake-up successes to the total wake-up times under the condition that the target user is based on a standard wake-up threshold value;
When the current wake-up rate of the target user is determined to exceed a preset wake-up rate range, the standard wake-up threshold is adjusted, and the adjusted standard wake-up threshold is updated to be the target wake-up threshold, so that the current wake-up rate of the target user based on the target wake-up threshold is within the preset wake-up rate range;
The adjusting the standard wake-up threshold value and updating the adjusted standard wake-up threshold value to a target wake-up threshold value comprises:
Obtaining a plurality of similarities of the multiple wake-up audios of the target user in a preset time period, wherein the similarities are in one-to-one correspondence with the multiple wake-up audios, and the similarities are the similarities of the wake-up audios and standard audios;
Determining a target wake-up threshold based on the similarities corresponding to the multiple wake-up audios, wherein the proportion of the number of the similarities reaching the target wake-up threshold in the multiple similarities to the total number of the multiple similarities is within a preset proportion range;
Updating the standard wake-up threshold to the target wake-up threshold, wherein the target wake-up threshold is used for executing wake-up on the intelligent device when the similarity between the received audio data of the target user and the standard audio is greater than or equal to the target wake-up threshold.
2. The method of claim 1, wherein the obtaining the current wake-up rate of the target user for the preset time period comprises:
Obtaining a plurality of wake-up audios of a target user and the similarity corresponding to the wake-up audios, wherein the similarity is a similarity value between the wake-up audios and a standard wake-up audio;
comparing a plurality of the similarities with the standard wake-up threshold, and determining the number of wake-up audios for which the similarities are greater than or equal to the standard wake-up threshold;
And calculating the proportion of the number of the awakening audios of the similarity which is larger than or equal to a standard awakening threshold value to the total number of the awakening audios, and obtaining the current awakening rate.
3. The method of claim 2, wherein the obtaining the current wake-up rate of the target user for the preset time period comprises:
when wake-up audio of a target user is detected, voiceprint information in the wake-up audio is acquired;
And determining the current wake-up rate corresponding to the target user based on the voiceprint information.
4. The method of claim 1, wherein when determining that the current wake-up rate of the target user is outside a preset wake-up rate range, comprising:
respectively comparing the current wake-up rate corresponding to the target user with a maximum preset wake-up rate and a minimum preset wake-up rate of the preset wake-up rate range;
When the current wake-up rate is smaller than the minimum preset wake-up rate, determining that the current wake-up rate exceeds the preset wake-up rate range;
and when the current wake-up rate is larger than the maximum preset wake-up rate, determining that the current wake-up rate exceeds the preset wake-up rate range.
5. The method of claim 1, wherein after the adjusting the standard wake threshold and updating the adjusted standard wake threshold to a target wake threshold, the method further comprises:
identifying the detected wake-up audio based on the updated target wake-up threshold;
and when the similarity of the wake-up audio reaches the target wake-up threshold, executing wake-up on the intelligent equipment.
6. The method according to any one of claims 1 to 5, wherein the obtaining the current wake-up rate of the target user in the preset time period includes:
And acquiring the current wake-up rate in the preset time period before the current time every preset period.
7. An optimization device for waking up an intelligent device, which is applied to the intelligent device, the device comprising:
The current wake-up rate acquisition module is used for acquiring the current wake-up rate of the target user in a preset time period, wherein the current wake-up rate is the proportion of the number of wake-up successes to the total wake-up times under the condition that the target user is based on a standard wake-up threshold value;
the adjustment updating module is used for adjusting the standard awakening threshold value when the current awakening rate of the target user exceeds a preset awakening rate range, and updating the adjusted standard awakening threshold value into a target awakening threshold value so that the current awakening rate of the target user based on the target awakening threshold value is in the preset awakening rate range;
The adjustment updating module comprises:
The first acquisition unit is used for acquiring a plurality of similarities of the multiple wake-up audios of the target user in a preset time period, wherein the similarities are in one-to-one correspondence with the multiple wake-up audios, and the similarities are the similarities of the wake-up audios and the standard audio;
the first determining unit is used for determining a target wake-up threshold value based on the similarities corresponding to the multiple wake-up audios, wherein the proportion of the number of the similarities reaching the target wake-up threshold value in the multiple similarities to the total number of the multiple similarities is in a preset proportion range;
And the updating unit is used for updating the standard awakening threshold value into the target awakening threshold value, and the target awakening threshold value is used for executing awakening on the intelligent equipment when the similarity between the received audio data of the target user and the standard audio is greater than or equal to the target awakening threshold value.
8. A smart device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 6 when executing the computer program.
9. A computer readable storage medium having stored therein program code which is callable by a processor to perform the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010898830.6A CN111949323B (en) | 2020-08-31 | 2020-08-31 | Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010898830.6A CN111949323B (en) | 2020-08-31 | 2020-08-31 | Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111949323A CN111949323A (en) | 2020-11-17 |
| CN111949323B true CN111949323B (en) | 2024-06-11 |
Family
ID=73368141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010898830.6A Active CN111949323B (en) | 2020-08-31 | 2020-08-31 | Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111949323B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112951243A (en) * | 2021-02-07 | 2021-06-11 | 深圳市汇顶科技股份有限公司 | Voice awakening method, device, chip, electronic equipment and storage medium |
| CN113076144A (en) * | 2021-04-23 | 2021-07-06 | 深圳创维－Rgb电子有限公司 | Power amplifier adaptation method, terminal device, storage medium and computer program product |
| CN113628620A (en) * | 2021-08-12 | 2021-11-09 | 云知声（上海）智能科技有限公司 | Intelligent device awakening method and device, electronic device and storage medium |
| CN114821859B (en) * | 2022-03-15 | 2024-01-12 | 杭州遁甲科技有限公司 | Intelligent equipment control method and device |
| CN117711395A (en) * | 2023-06-30 | 2024-03-15 | 荣耀终端有限公司 | Voice interaction method and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110992962A (en) * | 2019-12-04 | 2020-04-10 | 珠海格力电器股份有限公司 | Wake-up adjusting method and device for voice equipment, voice equipment and storage medium |
| CN111128155A (en) * | 2019-12-05 | 2020-05-08 | 珠海格力电器股份有限公司 | Awakening method, device, equipment and medium for intelligent equipment |
| CN111223490A (en) * | 2020-03-12 | 2020-06-02 | Oppo广东移动通信有限公司 | Voiceprint awakening method and device, equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11232788B2 (en) * | 2018-12-10 | 2022-01-25 | Amazon Technologies, Inc. | Wakeword detection |
-
2020
- 2020-08-31 CN CN202010898830.6A patent/CN111949323B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110992962A (en) * | 2019-12-04 | 2020-04-10 | 珠海格力电器股份有限公司 | Wake-up adjusting method and device for voice equipment, voice equipment and storage medium |
| CN111128155A (en) * | 2019-12-05 | 2020-05-08 | 珠海格力电器股份有限公司 | Awakening method, device, equipment and medium for intelligent equipment |
| CN111223490A (en) * | 2020-03-12 | 2020-06-02 | Oppo广东移动通信有限公司 | Voiceprint awakening method and device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111949323A (en) | 2020-11-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111949323B (en) | Optimization method and device for waking up intelligent equipment, intelligent equipment and storage medium | |
| CN108847219B (en) | Awakening word preset confidence threshold adjusting method and system | |
| CN109378000B (en) | Voice wake-up method, device, system, equipment, server and storage medium | |
| CN111968644B (en) | Intelligent device awakening method and device and electronic device | |
| CN111223497B (en) | Nearby wake-up method and device for terminal, computing equipment and storage medium | |
| CN108735209B (en) | Wake-up word binding method, intelligent device and storage medium | |
| WO2017156925A1 (en) | Unlock method and mobile terminal | |
| CN111558120B (en) | Sleep wake-up method, device, computer equipment and computer readable storage medium | |
| CN112420041B (en) | Control method and device for wake-up threshold of intelligent device, intelligent device and storage medium | |
| CN110148405B (en) | Voice instruction processing method and device, electronic equipment and storage medium | |
| CN110070857B (en) | Model parameter adjusting method and device of voice awakening model and voice equipment | |
| CN108564948B (en) | Voice recognition method and electronic equipment | |
| CN111312222A (en) | Awakening and voice recognition model training method and device | |
| US11380131B2 (en) | Method and device for face recognition, storage medium, and electronic device | |
| CN110782891B (en) | Audio processing method and device, computing equipment and storage medium | |
| CN110634468B (en) | Voice wake-up method, device, equipment and computer readable storage medium | |
| CN111128155B (en) | Awakening method, device, equipment and medium for intelligent equipment | |
| CN111357267B (en) | Alarm clock ringing time setting method and device and electronic equipment | |
| CN112233676A (en) | Intelligent device awakening method and device, electronic device and storage medium | |
| CN113963695A (en) | Awakening method, awakening device, equipment and storage medium of intelligent equipment | |
| CN116386676B (en) | Voice awakening method, voice awakening device and storage medium | |
| CN109086017A (en) | Control method, device and computer readable storage medium based on multi-screen terminal | |
| CN113160815A (en) | Intelligent control method, device and equipment for voice awakening and storage medium | |
| EP3793275B1 (en) | Location reminder method and apparatus, storage medium, and electronic device | |
| CN110164431B (en) | Audio data processing method and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |