CN109920417B - Voice processing method, device, equipment and storage medium - Google Patents
Voice processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN109920417B CN109920417B CN201910123621.1A CN201910123621A CN109920417B CN 109920417 B CN109920417 B CN 109920417B CN 201910123621 A CN201910123621 A CN 201910123621A CN 109920417 B CN109920417 B CN 109920417B
- Authority
- CN
- China
- Prior art keywords
- thread
- cpu core
- time
- online cpu
- consuming
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 27
- 238000012545 processing Methods 0.000 claims abstract description 44
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000006870 function Effects 0.000 claims description 18
- 230000015654 memory Effects 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 4
- 238000010411 cooking Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 241001672694 Citrus reticulata Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Landscapes
- Selective Calling Equipment (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a voice processing method, a voice processing device, voice processing equipment and a storage medium. The method comprises the following steps: searching an online CPU core and a corresponding running thread list when voice information is received; determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core; and calling a target online CPU core to create a voice recognition thread and performing recognition processing on voice information. The invention realizes that the voice recognition thread is dynamically established according to the occupation condition of the on-line CPU core in the intelligent household appliance so as to adjust the calculation strategy of the voice recognition algorithm, fully utilizes the CPU resource, optimizes the calculation speed of the voice recognition algorithm and further improves the voice recognition speed.
Description
Technical Field
Embodiments of the present invention relate to data processing technologies, and in particular, to a method, an apparatus, a device, and a storage medium for processing speech.
Background
With the rapid development of science and technology, more and more intelligent household appliances are added with a voice recognition function. In order to increase the speed of speech recognition, a multi-thread parallel running speech recognition algorithm can be generally adopted to perform recognition processing on speech information.
Currently, a Central Processing Unit (CPU) in an intelligent appliance dynamically adjusts the number of cores of the on-line CPU in consideration of power consumption of the intelligent appliance. However, when the number of the on-line CPU cores is small, a plurality of threads may work on the same on-line CPU core, and the voice recognition algorithm is executed in a multi-thread manner, because extra time is wasted in scheduling of the threads, the processing speed of the voice recognition algorithm is slowed down, and the voice recognition speed is further reduced.
Disclosure of Invention
In view of the above, the present invention provides a speech processing method, apparatus, device and storage medium, which can improve the speech recognition speed.
In a first aspect, an embodiment of the present invention provides a speech processing method, including:
searching an online CPU core and a corresponding running thread list when voice information is received;
determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core;
and calling the target online CPU core to create a voice recognition thread and carrying out recognition processing on voice information.
Further, before the receiving the voice message, the method further includes:
acquiring all pre-installed applications of the intelligent household appliance;
and analyzing the thread information corresponding to all the pre-installed applications to determine the time-consuming thread.
Further, the analyzing the thread information corresponding to all the pre-installed applications to determine a time-consuming thread includes:
acquiring the average continuous working time and the average restarting time interval of each thread;
taking the thread which at least meets one of the following conditions as a time-consuming thread: the average continuous operating time is greater than a preset operating time threshold, and the average restart time interval is less than a preset time interval threshold.
Further, at most one of the on-line CPU cores has no time-consuming thread included in its running thread list,
the determining that the online CPU core with the least number of time-consuming threads in the running thread list is a target online CPU core comprises:
acquiring time-consuming threads in the running thread list of each online CPU core, and counting the number of the time-consuming threads;
and determining the online CPU core with the least time-consuming thread number as a target online CPU core.
Furthermore, at least two running thread lists of the online CPU core do not contain time-consuming threads,
the determining that the online CPU core with the least number of time-consuming threads in the running threads is the target online CPU core comprises the following steps:
acquiring the total running time of all running threads in a running thread list of each online CPU core to determine the current occupancy rate of each online CPU core;
and determining the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as a target online CPU core.
Further, before the invoking the target online CPU core to create the speech recognition thread, the method further includes:
determining the number of target on-line CPU cores;
and determining the number of the voice recognition threads to be created according to the target online CPU core number.
Further, after the calling the target online CPU core to create the speech recognition thread, the method further includes:
determining a corresponding thread mode according to the number of the voice recognition threads, wherein the thread mode comprises a single thread mode and a multi-thread mode;
and running a voice recognition algorithm by adopting the thread mode.
Further, after the calling the target online CPU core to create the speech recognition thread, the method further includes:
and binding the voice recognition thread and the corresponding target online CPU core by utilizing a CPU affinity function.
In a second aspect, an embodiment of the present invention provides a speech processing apparatus, including:
the searching module is used for searching the online CPU core and the corresponding running thread list after receiving the voice information;
the first determining module is used for determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core;
and the calling module is used for calling the target online CPU core to create a voice recognition thread and carrying out recognition processing on voice information.
Further, the speech processing apparatus further includes:
the acquisition module is used for acquiring all pre-installed applications of the intelligent household appliance;
and the second determining module is used for analyzing the thread information corresponding to all the pre-installed applications to determine the time-consuming thread.
Further, the second determining module includes:
the first acquisition unit is used for acquiring the average continuous working time and the average restarting time interval of each thread;
a first determining unit, configured to take a thread that at least satisfies one of the following conditions as a time-consuming thread: the average continuous operating time is greater than a preset operating time threshold, and the average restart time interval is less than a preset time interval threshold.
Further, the first determining module includes:
a second obtaining unit, configured to obtain a time-consuming thread in the running thread list of each online CPU core and count the number of the time-consuming threads, where the running thread list of at most one online CPU core does not include the time-consuming thread;
and the second determining unit is used for determining the online CPU core with the least time-consuming thread number as a target online CPU core.
Further, the first determining module further includes:
a third obtaining unit, configured to obtain total running time of all running threads in the running thread list of each online CPU core to determine a current occupancy rate of each online CPU core, where the running thread list of at least two online CPU cores does not include a time-consuming thread;
and the third determining unit is used for determining the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as the target online CPU core.
Further, the speech processing apparatus further includes:
a third determining module, configured to determine the number of target online CPU cores before the target online CPU core is called to create the speech recognition thread;
and the fourth determining module is used for determining the number of the voice recognition threads to be created according to the target online CPU core number.
Further, the speech processing apparatus further includes:
a fifth determining module, configured to determine, after the target online CPU core is called to create the voice recognition thread, a corresponding thread mode according to the number of the voice recognition threads, where the thread mode includes a single thread mode and a multi-thread mode;
and the running module is used for running the voice recognition algorithm by adopting the thread mode.
Further, the speech processing apparatus further includes:
and the binding module is used for binding the voice recognition thread and the corresponding target online CPU core by utilizing a CPU affinity function after the target online CPU core is called to create the voice recognition thread.
In a third aspect, an embodiment of the present invention further provides an apparatus, including: a display screen, a memory, and one or more processors;
the display screen is used for displaying the state information of the intelligent household appliance;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the speech processing method of the first aspect.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the speech processing method according to the first aspect.
According to the method, when the voice information is received, the on-line CPU core and the corresponding running thread list are searched, then the on-line CPU core with the least time-consuming thread number in the running thread list is determined as the target on-line CPU core, the target on-line CPU core is called to establish the voice recognition thread, the voice recognition algorithm is run in the voice recognition thread to recognize the voice information, the voice recognition thread is dynamically established according to the occupation condition of the on-line CPU core in the intelligent household appliance, the calculation strategy of the voice recognition algorithm is adjusted, the CPU resource is fully utilized, the calculation speed of the voice recognition algorithm is optimized, and the voice recognition speed is further improved.
Drawings
Fig. 1 is a flowchart of a speech processing method according to an embodiment of the present invention;
fig. 2 is a flowchart of a speech processing method according to a second embodiment of the present invention;
fig. 3 is a flowchart of a speech processing method according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a voice-controlled intelligent home appliance according to a third embodiment of the present invention;
fig. 5 is a block diagram of a speech processing apparatus according to a fourth embodiment of the present invention;
fig. 6 is a schematic structural diagram of an apparatus according to a fifth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a speech processing method according to an embodiment of the present invention, where a speech recognition algorithm provided in the embodiment may be executed by an intelligent appliance, and the intelligent appliance includes two or more physical entities or is a physical entity. For example, intelligent appliances include, but are not limited to, intelligent range hoods, intelligent air conditioners, intelligent televisions, intelligent water heaters, and the like.
As shown in fig. 1, the speech processing method specifically includes the following steps:
and S110, searching an online CPU core and a corresponding running thread list when the voice information is received.
The voice information refers to a voice operation instruction sent to the intelligent household appliance by a user. In the embodiment, the language of the voice message is not limited, such as mandarin, english or dialects of various types; the sentence sequencing, the sentence type and the like of the voice information are not limited, for example, the voice information may be "turn off the fan", or the voice information may be "turn the air volume up", and the like.
An online CPU core may be understood to be a CPU core that remains operating online. In the embodiment, in order to improve the computing power of the intelligent home appliance, a dual-core processor, a quad-core processor, a six-core processor, or the like may be disposed in the intelligent home appliance, which is not limited. The dual-core processor refers to a processor integrating two operation cores, namely, a CPU chip integrating two CPU cores, and so on, the four-core processor refers to a CPU chip integrating four CPU cores, and the six-core processor refers to a CPU chip integrating six CPU cores. Of course, the functionality of the two or more processor cores integrated on each processor is the same. For example, a dual-core processor is a processor core having two identical functions on a single semiconductor-based processor, in other words, a dual-core processor is a core integrating two physical processor cores. Of course, to avoid the waste of CPU resources and reduce the power consumption of the smart appliance, part of the CPU core may be in an offline state. For example, the intelligent range hood is provided with the four-core processor, but only the light function of the intelligent range hood is used at the moment, only two or one CPU core is needed to be kept in an online working state, and other CPU cores are in an offline state, so that the power consumption of the intelligent range hood and the waste of CPU resources are reduced. The running thread list may be understood as a list including at least one running thread, and it should be understood that, when the user controls the intelligent household appliance by voice, after the intelligent household appliance receives voice information sent by the user, the intelligent household appliance scans the CPU cores configured therein to determine the state of each CPU core, and uses the CPU core in a working state as an online CPU core, and then scans and determines the running threads on each online CPU core, and counts all the running threads to form the running thread list. A running thread refers to a thread running on an online CPU core. Of course, the number of threads running on each on-line CPU core is not certain, and is not limited.
And S120, determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core.
The time-consuming threads are determined according to the amount of CPU resources occupied by each thread. In the embodiment, each online CPU core has a plurality of running threads, but in order to increase the speed of recognizing the voice information by the smart home appliance, when the voice recognition thread is created, an online CPU core that occupies the smallest CPU resource needs to be selected as a target online CPU core, where the target online CPU core may be understood as a CPU core that is in an idle state or is about to be idle. Wherein, in the idle state, the core can be understood as an online CPU core not containing a time-consuming thread; an on-line CPU core that does not include time-consuming threads but whose total CPU resource occupied by the running threads exceeds a certain threshold (e.g., 70% of the CPU resource) is therefore not idle. However, since the on-line CPU core does not include a time-consuming thread, the running thread in the on-line CPU core will be finished in a short time, that is, the on-line CPU core will be idle, and the on-line CPU core may also be considered as the target on-line CPU core. Of course, there are situations where there are multiple time-consuming threads running on each online CPU core in the intelligent appliance, and there are also situations where there is no time-consuming thread running. It can also be understood that when there are multiple on-line CPU cores of the intelligent appliance, there may be a case where all on-line CPU cores have time-consuming threads to run, and there may also be a case where all on-line CPU cores have no time-consuming threads to run. For the convenience of illustration, the step can be divided into two cases, that is, no time-consuming thread is included in the running thread list of at most one on-line CPU core, and no time-consuming thread is included in the running thread lists of at least two on-line CPU cores.
In one embodiment, when the running thread list of at most one online CPU core does not include a time-consuming thread, step S120 may include steps S1201-S1202, where it is to be noted that, when there are multiple online CPU cores in the intelligent appliance and at most one online CPU core does not include a time-consuming thread, the number of time-consuming threads in the online CPU cores including the time-consuming thread may be counted to determine the target online CPU core. The method comprises the following specific steps:
s1201, acquiring the time-consuming threads in the running thread list of each online CPU core, and counting the number of the time-consuming threads.
The number of the time-consuming threads is the number of the time-consuming threads in the threads running on each on-line CPU core. In an embodiment, each of the on-line CPU cores runs a plurality of threads, and in order to increase the processing speed of the smart home appliance on the threads, all the on-line CPU cores may be used to create a plurality of threads for the same algorithm, so as to run a plurality of threads corresponding to the algorithm in parallel. Exemplarily, taking an intelligent household appliance as an intelligent range hood as an example, the determination of the target online CPU core is explained. Assuming that a quad-core processor is configured on the intelligent range hood, that is, the intelligent range hood includes four CPU cores, which are a CPU core 11, a CPU core 22, a CPU core 33 and a CPU core 44, and a voice recognition function is configured on the intelligent range hood at the same time, if the intelligent range hood is using an application program corresponding to a video player, and the four CPU cores are all in an online working state, wherein the four online CPU cores, namely the CPU core 11, the CPU core 22, the CPU core 33 and the CPU core 44, are all running audio and video encoding and decoding threads, and the CPU core 11, the CPU core 22 and the CPU core 33 are also running network communication threads. Since the playing time of each video is at least several seconds, several minutes or even longer, it can be considered that the four online CPU cores cannot release the CPU resources in a short time, and the network communication thread needs to establish a communication connection with the server for a long time, it can be considered that the CPU resources cannot be released in a short time, at this time, the number of the time-consuming threads on each online CPU core can be determined, that is, the running thread lists of the three online CPU cores, i.e., CPU core 11, CPU core 22 and CPU core 33, all include two time-consuming threads, and the number of the time-consuming threads in the running thread list of CPU core 44 is one.
S1202, the online CPU core with the least time-consuming thread number is determined to be the target online CPU core.
In an embodiment, after determining the number of time-consuming threads running on each online CPU core, the intelligent appliance directly determines the online CPU core with the smallest number of time-consuming threads as the target online CPU core. For example, if it is determined in step S1201 that the number of time-consuming threads in the running thread list of the CPU core 44 is one and less than the number of time-consuming threads in the other three online CPU cores, the CPU core 44 may be determined as the target online CPU core.
In one embodiment, when the running thread lists of at least two on-line CPU cores do not contain time-consuming threads, step S120 may include steps S1203-S1204. It should be noted that, when there are multiple online CPU cores in the intelligent appliance and at least two of the online CPU cores do not include a time-consuming thread, the total running time of all running threads in the online CPU cores that do not include the time-consuming thread may be counted to determine the current occupancy rate of each online CPU core, so as to determine the target online CPU core according to the current occupancy rate. The method comprises the following specific steps:
s1203, obtaining the total running time of all running threads in the running thread list of each online CPU core to determine the current occupancy rate of each online CPU core.
The total running time can be understood as the total time required by each thread to run from the beginning to the end on the on-line CPU core in a unit time. It should be understood that all the running threads in each online CPU core are run alternately, i.e., the total running time may be the running time per unit time required by all the threads on each online CPU core except the system idle thread. For example, assuming that three running threads, i.e., a thread a, a thread B and a thread C, run in the online CPU core 55 of one of the intelligent appliances, if the running time required by the thread a is 200ms, the running time required by the thread B is 300ms, and the running time required by the thread C is 300ms, the total running time per unit time of all the running threads in the running thread list of the online CPU core 55 is 800 ms. Wherein, the occupancy rate is the ratio of the total running time of all the threads except the system idle thread on the CPU core in the unit time to the unit time. For example, if the unit time is 1 second, the current occupancy rate of the online CPU core 55 is 800ms/1s — 0.8, and the current occupancy rate of the online CPU core 55 is 80%.
And S1204, determining the on-line CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as a target on-line CPU core.
In the embodiment, after the total running time of all running threads in each online CPU core in unit time is determined, the intelligent household appliance performs statistical analysis on the total running time to determine the current occupancy rate of each online CPU core, and selects the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as the target online CPU core. The preset occupancy rate threshold is a parameter for measuring whether the on-line CPU core is idle or not. For example, the preset occupancy threshold may be 70%. Certainly, the size of the preset occupancy rate threshold is not limited, and developers can set the occupancy rate threshold according to different equipment parameters of the intelligent household appliance. Exemplarily, assuming that there are four online CPU cores on the intelligent range hood, which are the CPU core 11, the CPU core 22, the CPU core 33 and the CPU core 44, respectively, wherein there is no time-consuming thread on the two online CPU cores of the CPU core 11 and the CPU core 22, and the current occupancy rates of the CPU core 11 and the CPU core 22 in a unit time are 60% and 40%, respectively, since the current occupancy rates of the CPU core 11 and the CPU core 22 are both less than the preset occupancy rate threshold value of 70%, both the CPU core 11 and the CPU core 22 are determined as the target online CPU core. For another example, assuming that there are no time-consuming threads on both the CPU core 77 and the CPU core 88, and the current occupancy rate of the CPU core 77 is 80%, the current occupancy rate of the CPU core 88 is 50%, and since the current occupancy rate of the CPU core 77 is greater than the preset occupancy rate of 70%, only the CPU core 88 is determined as the target online CPU core.
S130, calling a target online CPU core to create a voice recognition thread and carrying out recognition processing on voice information.
The speech recognition thread is a thread for running a speech recognition algorithm. In an embodiment, each target online CPU core has a plurality of threads, and in order to perform recognition processing on voice information, a voice recognition thread needs to be created on the target online CPU core to run a voice recognition algorithm, so as to perform recognition processing on the voice information. Specifically, when at least two target online CPU cores are assumed, a speech recognition thread is created on each target online CPU core to run a speech recognition algorithm in parallel, thereby fully utilizing CPU resources and accelerating speech recognition speed. Of course, if there is only one target online CPU core, only one speech recognition thread is created on the target online CPU core to run the speech recognition algorithm, so as to avoid the problem of thread scheduling caused by multiple speech recognition threads on the same target online CPU core.
According to the technical scheme of the embodiment, when voice information is received, an online CPU core and a corresponding running thread list are searched, then the online CPU core with the least time-consuming thread number in the running thread list is determined as a target online CPU core, the target online CPU core is called to create a voice recognition thread, a voice recognition algorithm is run in the voice recognition thread to recognize and process the voice information, the voice recognition thread is dynamically created according to the occupation condition of the online CPU core in the intelligent household appliance to adjust the calculation strategy of the voice recognition algorithm, CPU resources are fully utilized, the calculation speed of the voice recognition algorithm is optimized, and further the voice recognition speed is improved.
Example two
Fig. 2 is a flowchart of a speech processing method according to a second embodiment of the present invention. The present embodiment is further described with reference to the above embodiments. Referring to fig. 2, the speech processing method specifically includes the following steps:
s210, all pre-installed applications of the intelligent household appliance are obtained.
Here, since the user cannot download and install the application program of the smart home appliance again after the smart home appliance is shipped from the factory, the developer needs to configure the application program corresponding to the smart home appliance in the smart home appliance before the smart home appliance is shipped from the factory in order to allow the user to intelligently control and use the smart home appliance. In the research and development stage of the intelligent household appliance, all pre-installed applications in the intelligent household appliance are acquired so as to determine thread information of each pre-installed application, which can occupy CPU resources for a long time.
S220, analyzing the thread information corresponding to all the pre-installed applications to determine the time-consuming threads.
In an embodiment, after all pre-installed applications in the intelligent appliance are acquired, thread information corresponding to each pre-installed application is analyzed to determine a time-consuming thread, where the step may specifically include steps S2201 to S2202:
s2201, acquiring the average continuous working time and the average restarting time interval of each thread.
Wherein, the average continuous working time can be understood as the continuous working time of each thread on the CPU core; the average restart interval may be understood as the interval between two consecutive restarts of each thread on the CPU core. In an embodiment, the continuous working time of each thread on the CPU core is not fixed, and the time interval between two restarts is also not fixed, so as to determine the time-consuming thread, the average continuous working time and the average restart time interval of each thread can be obtained. Illustratively, assuming that the first continuous working time of the thread a on the CPU core 11 is 4 seconds, the second continuous working time is 8 seconds, and the third continuous working time is 6 seconds, the average continuous working time of the thread a can be considered to be 6 seconds. Similarly, assuming that the time interval between the first and second boots of thread B on CPU core 22 is 6 milliseconds, the time interval between the second and third boots is 10 milliseconds, and the time interval between the third and fourth boots is 5 milliseconds, the average restart time interval for thread B may be considered to be 7 milliseconds.
S2202, taking the thread which at least meets one of the following conditions as a time-consuming thread: the average continuous on-time is greater than a preset on-time threshold and the average restart time interval is greater than a preset time interval threshold.
The preset working time threshold refers to a critical value for judging whether the average continuous working time of the time-consuming thread is a critical value; the preset time interval threshold refers to a critical value for determining whether the average restart time interval of the time-consuming threads is zero. For example, the preset on-time threshold may be set to 5 seconds, and the preset time interval threshold may be set to 10 milliseconds. It should be understood that when the average continuous working time of a thread is greater than the preset working time threshold, the thread is considered to be a time-consuming thread; similarly, a thread may be considered to be a time-consuming thread when its average restart time interval is less than a preset time interval threshold. Exemplarily, in step S2201, the average continuous operating time of the thread a in the CPU core 11 is 6S, and if the average continuous operating time is greater than 5 seconds, the thread a is a time-consuming thread; and the average restart time interval of thread B in CPU core 22 is 7 milliseconds, which is less than 10 milliseconds, then thread B is a time consuming thread. Of course, the specific values of the preset working time threshold and the preset time interval threshold are not limited, and developers can limit the specific actual conditions of the intelligent household appliance.
Of course, in the embodiment, the time-consuming thread may also be determined by the work scene of the intelligent appliance. Exemplarily, assuming that the intelligent range hood is in a cooking scene, a user does not watch a video for a long time in the working scene, although the video encoding and decoding thread occupies CPU resources for a long time in other scenes, the video encoding and decoding thread is quickly closed by the user in the cooking scene, so that the CPU resources are not occupied, and at this time, in establishing the time-consuming thread database, the average continuous working time of the running thread and the working scene can be used as the judgment standard to determine the time-consuming thread.
And S230, determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core.
And S240, determining the target on-line CPU core number.
The target number of CPU cores may be understood as the number of target online CPU cores. In an embodiment, the target number of online CPU cores is not fixed, which is related to whether there is a time consuming thread in the running thread list of the current online CPU core. Specifically, when there is no time-consuming thread on one online CPU core, the online CPU core may be considered as a target online CPU core, and certainly, if there are multiple online CPU cores in the intelligent appliance that have no time-consuming thread, the multiple online CPU cores may be considered as target online CPU cores. The process for determining the number of target online CPU cores may refer to the description of the above embodiments, and is not described herein again.
And S250, determining the number of the voice recognition threads to be created according to the core number of the target online CPU.
The number of the voice recognition threads is the number of the voice recognition threads established on all the on-line CPU cores in the intelligent household appliance. It should be noted that, in order to increase the speed of recognizing voice information by the smart home appliance and avoid creating multiple voice recognition threads on the same online CPU core, and when performing recognition processing on voice information, the voice recognition speed is reduced due to scheduling of the voice recognition threads, in an embodiment, the number of voice recognition threads is the same as the number of target online CPU cores. It will be appreciated that only one speech recognition thread is created on each target online CPU core.
And S260, calling a target online CPU core to create a voice recognition thread.
In an embodiment, after the target online CPU cores are determined, a corresponding speech recognition thread is created on each target online CPU core. Exemplarily, assuming that target online CPU cores on the intelligent range hood are CPU core 11 and CPU core 22, one speech recognition thread is respectively created on CPU core 11 and CPU core 22, that is, there are two speech recognition threads on the intelligent range hood. For another example, assuming that the target online CPU core on the intelligent range hood only has the CPU core 88, a speech recognition thread is created on the CPU core 88, that is, only one speech recognition thread is on the intelligent range hood.
And S270, determining a corresponding thread mode according to the number of the voice recognition threads.
The thread mode comprises a single thread mode and a multi-thread mode.
In an embodiment, thread mode is used to determine the number of threads running in a process, which may include single-threaded and multi-threaded modes. The single thread mode is that only one thread exists in one process; and the multi-thread mode is that a process has a plurality of threads. It should be noted that, when the number of the voice recognition threads is one, a single-thread mode is adopted; and when the number of the voice recognition threads is two or more, the multi-thread mode is adopted.
And S280, operating the voice recognition algorithm in a thread mode.
In an embodiment, after determining the thread mode, the speech recognition thread on the corresponding online CPU core runs the speech recognition algorithm in the thread mode. Specifically, when a single-thread mode is adopted, the number of target on-line CPU cores is indicated to be one, and a corresponding voice recognition thread is directly established on the target on-line CPU cores so as to run a voice recognition algorithm; when a multithreading mode is adopted, the number of the target on-line CPU cores is two or more, a voice recognition thread is created on each target on-line CPU core, and the two or more voice recognition threads are operated on the target on-line CPU cores in parallel to calculate the voice recognition algorithm, so that the voice recognition speed is improved.
S290, recognition processing is performed on the voice information.
In the embodiment, after the voice recognition algorithm is calculated through each voice recognition thread, the voice information is subjected to recognition processing so as to recognize a voice command sent by a user, and the intelligent household appliance is controlled according to the voice command.
According to the technical scheme, on the basis of the embodiment, the average continuous working time and the average restarting time interval of each thread are obtained, the threads with the average continuous working time larger than the preset working time threshold and/or the average restarting time interval smaller than the preset time interval threshold are used as time-consuming threads, the purpose that after voice information of a user is received, the running threads and the time-consuming threads in the intelligent household appliance are directly compared is achieved, the target online CPU core can be determined, the number of the voice recognition threads is determined according to the number of the target online CPU cores, then the corresponding thread mode is determined according to the number of the voice recognition threads, the voice recognition algorithm is operated, the CPU resources of the intelligent household appliance are fully utilized, and the calculation speed of the voice recognition algorithm is optimized.
On the basis of the foregoing embodiment, in order to improve the intimacy between the voice recognition thread and the target online CPU core, after step S260, the voice recognition processing method may further include: and binding the voice recognition thread and the corresponding target online CPU core by utilizing the CPU affinity function.
Wherein, the CPU Affinity function is a CPU Affinity function. In the embodiment, the process of binding the voice recognition thread and the corresponding target online CPU core by using the CPU affinity function is not described herein again. It should be noted that the reason for binding the voice recognition thread with the corresponding target online CPU core is to improve the probability that the voice recognition thread runs on the target online CPU core. For example, after the speech recognition thread a and the target online CPU core 11 are bound, the affinity between the two is 80; and the speech recognition thread a and the target online CPU core 22 are not bound, and the affinity between the two is 20, the speech recognition thread a is automatically run on the target online CPU core 11 after the speech recognition thread a is created.
EXAMPLE III
Fig. 3 is a flowchart of a speech processing method according to a third embodiment of the present invention. The technical solution of the present embodiment is to explain a speech processing method as a preferred embodiment on the basis of the above-described embodiments. Referring to fig. 3, the voice processing method includes:
s310, acquiring all pre-installed applications of the intelligent household appliance, and establishing a time-consuming thread database.
The time-consuming thread database can be understood as a warehouse containing time-consuming threads corresponding to all pre-installed applications in the intelligent household appliance. In the embodiment, in a development stage of an intelligent household appliance, all preinstalled applications in the intelligent household appliance are acquired and analyzed to obtain thread information that may occupy CPU resources for a long time or frequently occupy the CPU resources at short intervals, such as audio and video software encoding and decoding threads, network communication threads with large data volume, and the like, which may be considered as time-consuming threads, and after the time-consuming threads are determined, all the time-consuming threads are stored in a preset time-consuming thread database so as to be convenient for calling and using. For specific judgment of the time-consuming thread, reference may be made to the description of the above embodiments, and details are not described herein.
And S320, receiving voice information.
In an embodiment, before the smart home appliance performs recognition processing on the voice information, the smart home appliance receives a piece of voice information sent by a user, so that the smart home appliance performs recognition processing on the voice information. The length of the audio data corresponding to the voice information is related to the complexity of the voice recognition algorithm. For example, the longer the length of the audio data, the more complex the speech recognition algorithm becomes. In an embodiment, the length of the audio data corresponding to the voice information is set to several hundred milliseconds, for example, 300 milliseconds.
Fig. 4 is a schematic structural diagram of a voice-controlled intelligent home appliance according to a third embodiment of the present invention. As shown in fig. 4, four terminal devices, namely, an intelligent range hood 301, an intelligent air conditioner 302, an intelligent television 303, and an intelligent water heater 304, are disposed in the home of a certain user, and the voice-controlled intelligent home appliance will be specifically described. After the user 300 sends out the voice information of "turning off the fan", the four terminal devices of the intelligent range hood 301, the intelligent air conditioner 302, the intelligent television 303 and the intelligent water heater 304 recognize the voice information to convert the voice information into the voice command word preset by the intelligent household appliance, because only the intelligent range hood 301 among the four terminal devices stores the voice command word corresponding to the voice information, the intelligent range hood 301 executes the corresponding operation according to the voice command word, namely, the fan of the intelligent range hood 301 is turned off, and other terminal devices do not execute the corresponding operation.
S330, searching an online CPU core and a corresponding running thread list.
In the embodiment, each CPU core in the intelligent appliance is subjected to polling scanning to determine the online CPU core, and then each online CPU core is subjected to polling scanning to obtain the running thread on each online CPU core and form a corresponding running thread list. After determining the running thread list, it is determined whether the running thread list includes the time-consuming threads in step S310 and the number of the time-consuming threads included in the running thread list. Each online CPU core has an operating thread list, and each operating thread list is independent, so that statistical analysis can be performed on the time-consuming threads on each online CPU core.
S340, determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core.
In the embodiment, after determining the running thread list on each online CPU core, determining the number of time-consuming threads on each online CPU core, and if the running thread list of one online CPU core does not contain the time-consuming threads, directly determining the online CPU core as a target online CPU core; and if each online CPU core in the intelligent household appliance has a time-consuming thread, selecting the online CPU core with the least number of time-consuming threads as a target online CPU core. Certainly, two or more online CPU cores do not include a time-consuming thread, the current occupancy rates of the two or more online CPU cores are obtained at this time, and the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold is determined as the target online CPU core, which can be understood as that, if the current occupancy rates of the two or more online CPU cores are both smaller than the preset occupancy rate threshold, the two or more online CPU cores are both determined as the target online CPU core; and if the current occupancy rate of only one online CPU core is smaller than the preset occupancy rate threshold value, determining the online CPU core as the target online CPU core.
And S350, calling the target online CPU core to create a voice recognition thread and performing recognition processing on the voice information.
In an embodiment, after the target online CPU core is determined, the number of the voice recognition threads is determined according to the number of the target online CPU cores, and a voice recognition algorithm is operated in each voice recognition thread so as to perform recognition processing on voice information.
According to the technical scheme of the embodiment, the voice recognition thread is dynamically established according to the occupation condition of the on-line CPU core, and then the voice recognition algorithm is calculated through the voice recognition thread so as to recognize and process the voice information, so that the CPU resource is fully utilized under the condition of not influencing the CPU power consumption, the calculation speed of the voice recognition algorithm is optimized, and the voice recognition speed is further optimized.
Example four
Fig. 5 is a block diagram of a speech processing apparatus according to a fourth embodiment of the present invention. The voice processing apparatus of this embodiment may be configured in an intelligent appliance, and referring to fig. 5, the voice processing apparatus includes: a lookup module 410, a first determination module 420, and a calling module 430.
The searching module 410 is configured to, upon receiving the voice message, search for an online CPU core and a corresponding running thread list;
a first determining module 420, configured to determine, as a target online CPU core, an online CPU core that consumes the least number of threads in a running thread list;
and the calling module 430 is used for calling the target online CPU core to create a voice recognition thread and perform recognition processing on the voice information.
According to the technical scheme of the embodiment, when voice information is received, an online CPU core and a corresponding running thread list are searched, then the online CPU core with the least time-consuming thread number in the running thread list is determined as a target online CPU core, the target online CPU core is called to create a voice recognition thread, a voice recognition algorithm is run in the voice recognition thread to recognize and process the voice information, the voice recognition thread is dynamically created according to the occupation condition of the online CPU core in the intelligent household appliance to adjust the calculation strategy of the voice recognition algorithm, CPU resources are fully utilized, the calculation speed of the voice recognition algorithm is optimized, and further the voice recognition speed is improved.
On the basis of the above embodiment, the speech processing apparatus further includes:
the acquisition module is used for acquiring all pre-installed applications of the intelligent household appliance;
and the second determining module is used for analyzing the thread information corresponding to all the pre-installed applications to determine the time-consuming thread.
On the basis of the above embodiment, the second determining module includes:
the first acquisition unit is used for acquiring the average continuous working time and the average restarting time interval of each thread;
a first determining unit, configured to take a thread that at least satisfies one of the following conditions as a time-consuming thread: the average continuous on-time is greater than a preset on-time threshold, and the average restart time interval is less than a preset time interval threshold.
On the basis of the above embodiment, the first determining module 420 includes:
the second acquisition unit is used for acquiring the time-consuming threads in the running thread list of each online CPU core and counting the number of the time-consuming threads, wherein the running thread list of at most one online CPU core does not contain the time-consuming threads;
and the second determining unit is used for determining the online CPU core with the least time-consuming thread number as the target online CPU core.
On the basis of the above embodiment, the first determining module 420 further includes:
a third obtaining unit, configured to obtain total running time of all running threads in the running thread list of each online CPU core to determine a current occupancy rate of each online CPU core, where the running thread list of at least two online CPU cores does not include a time-consuming thread;
and the third determining unit is used for determining the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as the target online CPU core.
On the basis of the above embodiment, the speech processing apparatus further includes:
the third determining module is used for determining the number of the target online CPU cores before the target online CPU cores are called to create the voice recognition thread;
and the fourth determining module is used for determining the number of the voice recognition threads to be created according to the number of the target online CPU cores.
On the basis of the above embodiment, the speech processing apparatus further includes:
the fifth determining module is used for determining a corresponding thread mode according to the number of the voice recognition threads after the target online CPU core is called to create the voice recognition threads, wherein the thread mode comprises a single thread mode and a multi-thread mode;
and the operation module is used for operating the voice recognition algorithm in a thread mode.
On the basis of the above embodiment, the speech processing apparatus further includes:
and the binding module is used for binding the voice recognition thread and the corresponding target online CPU core by utilizing the CPU affinity function after the target online CPU core is called to create the voice recognition thread.
The voice processing device can execute the voice processing method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
EXAMPLE five
Fig. 6 is a schematic structural diagram of an apparatus according to a fifth embodiment of the present invention. The apparatus comprises: a processor 510, a memory 520, an input device 530, an output device 540, and a display screen 550. The device may be an intelligent appliance, and the number of the processors 510 in the device may be one or more, and one processor 510 is taken as an example in fig. 6. The number of the memories 520 in the device may be one or more, and one memory 520 is taken as an example in fig. 6. The processor 510, the memory 520, the input device 530, the output device 540, and the display screen 550 of the apparatus may be connected by a bus or other means, as exemplified by the bus connection in fig. 6. In the embodiment, the intelligent household appliance can be an intelligent range hood, an intelligent air conditioner, an intelligent television, an intelligent water heater and the like.
The memory 520 is a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the intelligent appliance according to any embodiment of the present invention (e.g., the search module 410, the first determination module 420, and the call module 430 in the voice processing apparatus). The memory 520 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of the device, and the like. Further, the memory 520 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 520 may further include memory located remotely from processor 510, which may be connected to devices through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input means 530 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the intelligent appliance, and may also be a camera for acquiring images and a sound pickup device for acquiring audio data. The output device 540 may include an audio device such as a speaker. It should be noted that the specific composition of the input device 530 and the output device 540 can be set according to actual situations.
The display screen 550 may be a touch-enabled display screen 550, which may be a capacitive screen, an electromagnetic screen, or an infrared screen. Generally, the display screen 550 is configured to display data according to an instruction of the processor 510, for example, to display status information of the smart appliance, for example, to display information of power amount and operation mode, and to receive a touch operation applied to the display screen 550, for example, when the smart appliance is a range hood, the touch operation may be an option of increasing an air volume, switching operation modes, and sending a corresponding signal to the processor 510 or other devices.
The processor 510 implements the above-described voice processing method by executing software programs, instructions, and modules stored in the memory 520 to thereby execute various functional applications of the device and data processing.
The device provided above can be used to execute the speech processing method provided in any of the above embodiments, with corresponding functions and benefits.
EXAMPLE six
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a speech processing method, including:
searching an online CPU core and a corresponding running thread list when voice information is received;
determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core;
and calling a target online CPU core to create a voice recognition thread and performing recognition processing on voice information.
Of course, the storage medium provided by the embodiment of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the operations of the voice processing method described above, and may also perform related operations in the voice processing method provided by any embodiment of the present invention, and have corresponding functions and advantages.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a robot, a personal computer, a server, or a network device) to execute the voice processing method according to any embodiment of the present invention.
It should be noted that, in the above-mentioned speech processing apparatus, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above-mentioned division as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method of speech processing, comprising:
acquiring all pre-installed applications of the intelligent household appliance;
analyzing the thread information corresponding to all the pre-installed applications to determine time-consuming threads;
searching an online CPU core and a corresponding running thread list when voice information is received;
determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core;
and calling the target online CPU core to create a voice recognition thread and carrying out recognition processing on voice information.
2. The method of claim 1, wherein the analyzing the thread information corresponding to all the pre-installed applications to determine a time-consuming thread comprises:
acquiring the average continuous working time and the average restarting time interval of each thread;
taking the thread which at least meets one of the following conditions as a time-consuming thread: the average continuous operating time is greater than a preset operating time threshold, and the average restart time interval is less than a preset time interval threshold.
3. The method of claim 1, wherein no time-consuming threads are included in the running thread list of at most one of the on-line CPU cores,
the determining that the online CPU core with the least number of time-consuming threads in the running thread list is a target online CPU core comprises:
acquiring time-consuming threads in the running thread list of each online CPU core, and counting the number of the time-consuming threads;
and determining the online CPU core with the least time-consuming thread number as a target online CPU core.
4. The method of claim 1, wherein at least two of the on-line CPU cores have no time-consuming threads included in their thread lists,
the determining that the online CPU core with the least number of time-consuming threads in the running threads is the target online CPU core comprises the following steps:
acquiring the total running time of all running threads in a running thread list of each online CPU core to determine the current occupancy rate of each online CPU core;
and determining the online CPU core with the current occupancy rate smaller than the preset occupancy rate threshold value as a target online CPU core.
5. The method of claim 1, further comprising, prior to the invoking the target online CPU core to create a speech recognition thread:
determining the number of target on-line CPU cores;
and determining the number of the voice recognition threads to be created according to the target online CPU core number.
6. The method of claim 5, after the invoking the target online CPU core to create a speech recognition thread, further comprising:
determining a corresponding thread mode according to the number of the voice recognition threads, wherein the thread mode comprises a single thread mode and a multi-thread mode;
and running a voice recognition algorithm by adopting the thread mode.
7. The method of claim 1, after the invoking the target online CPU core to create a speech recognition thread, further comprising:
and binding the voice recognition thread and the corresponding target online CPU core by utilizing a CPU affinity function.
8. A speech processing apparatus, comprising:
the acquisition module is used for acquiring all pre-installed applications of the intelligent household appliance;
the second determining module is used for analyzing the thread information corresponding to all the pre-installed applications to determine a time-consuming thread;
the searching module is used for searching the online CPU core and the corresponding running thread list after receiving the voice information;
the first determining module is used for determining the online CPU core with the least time-consuming thread number in the running thread list as a target online CPU core;
and the calling module is used for calling the target online CPU core to create a voice recognition thread and carrying out recognition processing on voice information.
9. An apparatus, comprising: a display screen, a memory, and one or more processors;
the display screen is used for displaying the state information of the intelligent household appliance;
the memory for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the speech processing method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the speech processing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910123621.1A CN109920417B (en) | 2019-02-18 | 2019-02-18 | Voice processing method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910123621.1A CN109920417B (en) | 2019-02-18 | 2019-02-18 | Voice processing method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109920417A CN109920417A (en) | 2019-06-21 |
| CN109920417B true CN109920417B (en) | 2021-05-18 |
Family
ID=66961856
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910123621.1A Active CN109920417B (en) | 2019-02-18 | 2019-02-18 | Voice processing method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109920417B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112559269A (en) * | 2020-12-21 | 2021-03-26 | 北京安兔兔科技有限公司 | Processor performance testing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010157105A (en) * | 2008-12-26 | 2010-07-15 | Toshiba Corp | Program creation device for testing model |
| CN103365718A (en) * | 2013-06-28 | 2013-10-23 | 贵阳朗玛信息技术股份有限公司 | Thread scheduling method, thread scheduling device and multi-core processor system |
| CN107526640A (en) * | 2017-08-17 | 2017-12-29 | 广东欧珀移动通信有限公司 | Method for managing resource, device, mobile terminal and computer-readable recording medium |
-
2019
- 2019-02-18 CN CN201910123621.1A patent/CN109920417B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2010157105A (en) * | 2008-12-26 | 2010-07-15 | Toshiba Corp | Program creation device for testing model |
| CN103365718A (en) * | 2013-06-28 | 2013-10-23 | 贵阳朗玛信息技术股份有限公司 | Thread scheduling method, thread scheduling device and multi-core processor system |
| CN107526640A (en) * | 2017-08-17 | 2017-12-29 | 广东欧珀移动通信有限公司 | Method for managing resource, device, mobile terminal and computer-readable recording medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109920417A (en) | 2019-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113168304A (en) | Conditionally assigning various automatic assistant functions to interactions with peripheral assistant control devices | |
| CN109688036B (en) | Control method and device of intelligent household appliance, intelligent household appliance and storage medium | |
| CN110308782B (en) | Power consumption prediction and control method and device and computer readable storage medium | |
| JP7195343B2 (en) | Orchestrating the execution of a series of actions requested to be performed through the Automation Assistant | |
| CN110875041A (en) | Voice control method, device and system | |
| CN107704070B (en) | Application cleaning method and device, storage medium and electronic equipment | |
| CN107145329A (en) | Apparatus control method, device and smart machine | |
| CN106528290A (en) | Resource scheduling method for mobile terminal and mobile terminal | |
| JP7017598B2 (en) | Data processing methods, devices, devices and storage media for smart devices | |
| CN110808044B (en) | Voice control method and device for intelligent household equipment, electronic equipment and storage medium | |
| CN107894827B (en) | Application cleaning method and device, storage medium and electronic equipment | |
| CN111192590B (en) | Voice wake-up method, device, equipment and storage medium | |
| KR20220165777A (en) | Fulfillment of an actionable request before the user selects a specific autocomplete suggestion to complete the current user input | |
| CN110414582B (en) | Model training method and device, computing equipment and storage medium | |
| JP7262532B2 (en) | VOICE INTERACTIVE PROCESSING METHOD, APPARATUS, ELECTRONIC DEVICE, STORAGE MEDIUM AND PROGRAM | |
| CN111091813A (en) | Voice wakeup model updating method, device, equipment and medium | |
| CN112233676B (en) | Intelligent device awakening method and device, electronic device and storage medium | |
| CN112908321A (en) | Device control method, device, storage medium, and electronic apparatus | |
| CN118535073A (en) | Automated assistant controlled compressed spoken utterances for complex application GUIs | |
| CN109920417B (en) | Voice processing method, device, equipment and storage medium | |
| CN111782409A (en) | Task processing method, risk identification task processing device and electronic equipment | |
| CN112925634A (en) | Heterogeneous resource scheduling method and system | |
| CN107870809B (en) | Application closing method and device, storage medium and electronic equipment | |
| CN110837415A (en) | Thread scheduling method and device based on RISC-V multi-core processor | |
| CN112420043A (en) | Intelligent awakening method and device based on voice, electronic equipment 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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200403 Address after: 510530 Guangdong city of Guangzhou province YUNPU Whampoa district four Road No. 6 Applicant after: Guangzhou Shiyuan Electronic Technology Company Limited Address before: 510530, No. four, 6, Po Po Road, Yun Po Industrial Zone, Whampoa District, Guangzhou, Guangdong Applicant before: Guangzhou Shiyuan Electronic Technology Company Limited Applicant before: GUANGZHOU RUIXIN ELECTRONICS Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |