The system of the method for terminal wake-up, awakening method and correspondence is set
Technical field
The present embodiments relate to data processing technique, particularly relate to a kind of arrange the method for terminal wake-up, awakening method
And the system of correspondence.
Background technology
Along with constantly weeding out the old and bring forth the new of intelligent terminal, the emerging intelligent terminal such as VR has provided the user the most omnibearing virtual
Experience, although these intelligent terminal differ from one another, but the most all remain the energy saver mode in mobile phone.
In the power save mode, the equal power-off of peripheral hardware unit such as sensor of intelligent terminal, to save electric energy.Waking up up
Time, need user to wake up instruction up by pressing the button to send, indicate each hardware power on and run.But, in some cases,
Such as, worn the user of the VR helmet to be not easy to touch button and wake up up.For another example, inconvenient hand-held intelligent terminal when, the most not
Convenient by pressing the button wake-up device.Accordingly, it would be desirable to existing wake-up mode is improved.
Summary of the invention
The present invention provides a kind of system arranging the method for terminal wake-up, awakening method and correspondence, to solve existing intelligence
Terminal can only rely on the mode of machinery touch-control to wake up terminal up.
First aspect, embodiments provides a kind of method arranging terminal wake-up, including: when receiving default checking
Waking up voice signal, and default speech recognition thresholding up in length, wherein, described speech recognition thresholding for determining when waking up up is
The no boundary value starting the voice signal that identification is received;The voice signal that wakes up up received is saved in waking up up of described terminal
In processing module.
Second aspect, the embodiment of the present invention additionally provides a kind of awakening method for terminal, including: proceed to not in terminal
During dormancy state, start and terminal wakes up processing module up;Wake up up described in the voice signal received being delivered to during dormancy
Processing module;When described wake up up processing module monitor described voice signal meet default speech recognition thresholding time, verify institute
Whether the voice signal received can wake up terminal up.
The third aspect, the embodiment of the present invention additionally provides a kind of system arranging terminal wake-up, including: module is set, uses
Waking up voice signal, and default speech recognition thresholding in receiving to preset up in checking duration, wherein, described speech recognition thresholding is used
In determining whether to start the boundary value of the voice signal that identification is received when waking up up;Wake up voice signal memory module up, be used for
The voice signal that wakes up up received is saved in the waking up up in processing module of described terminal.
Fourth aspect, the embodiment of the present invention also provides for a kind of waking up system up for terminal, including: wake up processing module up;
Starting module, for when terminal proceeds to resting state, the loss started in terminal wakes up processing module up;Speech reception module,
Processing module is waken up up described in being delivered to by the voice signal received during dormancy;The described processing module that wakes up up is for working as
The described processing module that wakes up up monitors described voice signal when meeting default speech recognition thresholding, the voice letter that checking is received
Number whether can wake up terminal up.
What the present invention was provided by limit wakes up voice signal and speech recognition thresholding up, on the one hand can provide cleverer
The unlocking manner lived, moreover it is possible to improve the motility that voice signal is set.
Accompanying drawing explanation
Fig. 1 is the flow chart of the method arranging terminal wake-up in the embodiment of the present invention one;
Fig. 2 is the structural representation of the system arranging terminal wake-up in the embodiment of the present invention two;
Fig. 3 is the flow chart of the awakening method for terminal in the embodiment of the present invention three;
Fig. 4 is the structural representation of the system that wakes up up for terminal in the embodiment of the present invention four.
Detailed description of the invention
The present invention is described in further detail with embodiment below in conjunction with the accompanying drawings.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that, in order to just
Part related to the present invention is illustrate only rather than entire infrastructure in description, accompanying drawing.
Embodiment one
The flow chart of the method arranging terminal wake-up that Fig. 1 provides for the embodiment of the present invention one, the present embodiment is applicable to
Be provided for waking up up the situation of terminal to use voice mode to wake up terminal up, the method can be performed by the system of setting,
Wherein, the described system that arranges is installed in the terminal.Described terminal includes but not limited to: mobile phone, VR device, wearable device etc., its
In, it is excellent with VR device.Described method to set up specifically includes following steps:
Step S110, receive preset checking duration in wake up voice signal, and default speech recognition thresholding, wherein, institute up
Predicate sound recognitiion gate limits the use of in the boundary value determining whether to start the voice signal that identification is received when waking up up.
Specifically, user wakes up the button of voice signal up in the interface click input that arranges of terminal, and says for verifying
The statement waken up up, described arranges system by the speech reception module (i.e. comprising the circuit of mike) of terminal in default checking
Voice signal is received in duration.Wherein, wake up processing module up process the data volume of voice signal to reduce, described system is set
The citing of default checking duration is less than 3s.
Meanwhile, before reception wakes up voice signal up or after reception wakes up voice signal up, also by default voice
Identify thresholding.
Specifically, in one case, described speech recognition thresholding can be before reception wakes up voice signal up in advance
In fixed minimum energy thresholding (such as 30 decibels), highest energy thresholding, low-limit frequency thresholding, highest frequency thresholding at least one
Kind, described speech recognition thresholding is saved in non-volatile storage medium.To arrange system operation based on user true when described
It is set to setting when waking up voice signal up, the speech recognition thresholding stored is adjusted and deposits to waking up up in processing module.
In another case, the described system that arranges, when reception wakes up voice signal up, wakes up voice up described in display in real time
Signal place frequency rule is interval and/or energy is interval (as decibel is interval).User can arrange language according to shown each interval selection
Sound identification thresholding, set speech recognition thresholding is saved in non-volatile memories Jie of terminal by the most described system that arranges respectively
Matter and described wake up up in processing module.
In another case, described system is set wakes up voice signal up when reception wakes up voice signal up or in reception
Afterwards, arrange according to waking up the frequency range of voice signal and/or energy range up.
Such as, described arrange system by the frequency range waking up voice signal up received and/or energy range
At least one in little value, maximum is as speech recognition thresholding, and set speech recognition thresholding is saved in end respectively
The non-volatile memory medium of end and described wake up up in processing module.
And for example, described the system expansion ratio according to default lower limit and the upper limit is set, is waken up up voice letter receive
Number frequency range and/or energy range in minima, on the basis of at least one in maximum, speech recognition door is set
Limit, and set speech recognition thresholding is saved in respectively the non-volatile memory medium of terminal and described wakes up processing module up
In.
Wake up processing module described in here, up to be made up of the primary processor of terminal, internal memory etc..In order to reduce hardware fortune as far as possible
The power consumption that brought of row, described in wake up processing module up and be made up of the circuit comprising micro-control unit (MCU).Wherein, in MCU
Built-in storage medium, to deposit described speech recognition thresholding.It addition, during terminal wake-up state, primary processor takies voice
The resource of receiver module (such as mike etc.).The described primary processor arranged in system may indicate that MCU storage wakes up voice letter up
Number and speech recognition thresholding etc..
Step S120, the voice signal that wakes up up received is saved in the waking up up in processing module of described terminal.
Here, described arrange system the original awake voice signal received can be saved in described in wake up processing module up
In.A kind of alternative is, described arrange system extract described in wake up up in voice signal characteristic information, by described characteristic information
Being saved in and wake up up in processing module, wherein said characteristic information includes: wake up up the signature waveform in voice signal and/or according to
Wake up the text message that voice signal is translated up.
Such as, the described system that arranges is extracted and is waken up the frequency waveform of voice signal, the frequency change signature waveform such as the fastest up, and
The signature waveform extracted is saved in and wakes up up in processing module.
And for example, the described system that arranges, according to default voice-text translator, is waken up up voice signal conversion by receive
Become corresponding text message, and described text message is saved in as characteristic information wakes up up in processing module.
It should be noted that the characteristic information extracted in above-mentioned two examples can be maintained in by the described system that arranges
Wake up up in processing module.
In the present embodiment, be provided by limit wakes up voice signal and speech recognition thresholding up, on the one hand can have
Effect reduction terminal verifies complexity during voice signal during resting state, on the other hand can also reduce checking voice signal
Time power consumption.
Embodiment two
As in figure 2 it is shown, the structural representation of the system that terminal wake-up is set of the embodiment of the present invention two offer, the present embodiment
It is applicable to be provided for waking up up the situation of terminal in order to use voice mode to wake up terminal up.Wherein, the described system that arranges is pacified
Dress is in the terminal.Described terminal includes but not limited to: mobile phone, VR device, wearable device etc., wherein, be excellent with VR device.Described
Arrange system 1 to specifically include: arrange module 11, wake up voice signal memory module 12 up.
Described arrange module 11 for receive preset checking duration in wake up voice signal, and default speech recognition door up
Limit, wherein, described speech recognition thresholding for determining whether to start the boundary value of the voice signal that identification is received when waking up up.
Specifically, user wakes up the button of voice signal up in the interface click input that arranges of terminal, and says for verifying
The statement waken up up, the described module 11 that arranges is tested default by the speech reception module (i.e. comprising the circuit of mike) of terminal
Voice signal is received in card duration.Wherein, wake up processing module up process the data volume of voice signal to reduce, described mould is set
The citing of block 11 default checking duration is less than 3s.
Meanwhile, before reception wakes up voice signal up or after reception wakes up voice signal up, also by default voice
Identify thresholding.
Specifically, in one case, described speech recognition thresholding can be before reception wakes up voice signal up in advance
In fixed minimum energy thresholding (such as 30 decibels), highest energy thresholding, low-limit frequency thresholding, highest frequency thresholding at least one
Kind, described speech recognition thresholding is saved in non-volatile storage medium.When described, module 11 operation based on user is set
It is defined as setting when waking up voice signal up, the speech recognition thresholding stored is adjusted and deposits to waking up up in processing module.
In another case, the described module 11 that arranges, when reception wakes up voice signal up, wakes up language up described in display in real time
Tone signal place frequency rule is interval and/or energy is interval (as decibel is interval).User can be arranged according to shown each interval selection
Speech recognition thresholding, set speech recognition thresholding is saved in the non-volatile of terminal and deposits by the most described module 11 that arranges respectively
Storage media and described wake up up in processing module.
In another case, described arrange module 11 when reception wakes up voice signal up or reception wake up up voice letter
After number, arrange according to waking up the frequency range of voice signal and/or energy range up.
Such as, the described module 11 that arranges is by the frequency range waking up voice signal up received and/or energy range
At least one in minima, maximum is as speech recognition thresholding, and is saved in respectively by set speech recognition thresholding
The non-volatile memory medium of terminal and described wake up up in processing module.
And for example, described the module 11 expansion ratio according to default lower limit and the upper limit is set, is waken up up voice receive
On the basis of minima in the frequency range of signal and/or energy range, at least one in maximum, speech recognition is set
Thresholding, and set speech recognition thresholding is saved in respectively the non-volatile memory medium of terminal and described wakes up process mould up
In block.
Wake up processing module described in here, up to be made up of the primary processor of terminal, internal memory etc..In order to reduce hardware fortune as far as possible
The power consumption that brought of row, described in wake up processing module up and be made up of the circuit comprising micro-control unit (MCU).Wherein, in MCU
Built-in storage medium, to deposit described speech recognition thresholding.It addition, during terminal wake-up state, primary processor takies voice
The resource of receiver module (such as mike etc.).The described primary processor arranged in module 11 may indicate that MCU storage wakes up voice up
Signal and speech recognition thresholding etc..
The described voice signal memory module 12 that wakes up up is for being saved in described terminal by the voice signal that wakes up up received
Wake up up in processing module.
Wake up voice signal memory module 12 described in here, up and the original awake voice signal received can be saved in described
Wake up up in processing module.A kind of alternative is, described in wake up up voice signal memory module 12 extract described in wake up voice signal up
In characteristic information, described characteristic information is saved in and wakes up up in processing module, wherein said characteristic information includes: wake up voice up
Signature waveform in signal and/or according to waking up the text message that voice signal is translated up.
Such as, wake up voice signal memory module 12 described in up to extract and wake up the frequency waveform of voice signal, frequency change up
The fast signature waveform that waits, and the signature waveform extracted is saved in wakes up up in processing module.
And for example, voice signal memory module 12 is waken up up according to default voice-text translator, by calling out of being received described in
Awake voice signal is converted into corresponding text message, and is saved in as characteristic information by described text message and wakes up processing module up
In.
Waking up voice signal memory module 12 described in it should be noted that up can be by the spy extracted in above-mentioned two examples
Reference breath is maintained in waking up up in processing module.
Embodiment three
As it is shown on figure 3, the present embodiment provides a kind of awakening method for terminal.Described awakening method is in not in terminal
During dormancy state, waken up up the situation of terminal by monitoring and recognition of speech signals.Described awakening method is by installing in the terminal
The system that wakes up up perform.Described terminal includes but not limited to: mobile phone, VR device, wearable device etc., wherein, with VR device is
Excellent.Described awakening method comprises the steps:
Step S210, when terminal proceeds to resting state, start and terminal wake up processing module up.
Specifically, when terminal is not because operating for a long time or proceeding to resting state under dormancy instructs, described in wake up system up
Processing module is waken up up described in startup.
The system that wakes up up described in here, can be by the Main Processor Unit (comprising the hardware cell of primary processor and internal memory) of terminal
The checking wakeup process of processing module is waken up up described in performing.
In a kind of alternative, described in wake up processing module up and include MCU and peripheral circuit thereof.Described MCU connects terminal
In Main Processor Unit.
When described wake up up system monitoring proceed to resting state to terminal time, by the speech reception module in terminal from terminal
Main Processor Unit switch to wake up processing module up.
Specifically, when described wake up up system monitoring proceed to resting state to terminal time, control primary processor to MCU output open
Dynamic instruction so that wake up processing module up and perform default monitoring and wake up proof procedure up.
Step S220, during dormancy, the voice signal received is delivered to described in wake up processing module up.
Wake up system instruction primary processor described in here, up to transfer to wake up processing module up by the control of speech reception module.
During dormancy, described in wake up the voice signal that speech reception module (such as mike) received by system up and export to waking up process up
Module.The described processing module real-time reception voice signal that wakes up up, and perform to wake up proof procedure up.Wherein, wake up up authenticated described in
Journey is not necessarily software program, it is also possible to be combined with hardware circuit and software wakes up proof procedure up.
In waking up proof procedure up, described in wake up up processing module according to default speech recognition thresholding determine whether start
Wake up checking up.
Such as, it is not up to the minima of speech recognition thresholding or beyond language when the frequency of the voice signal of real-time reception
During the maximum of sound identification thresholding, described in wake up processing module up and not perform step S230, otherwise then perform step S230.
And for example, when the decibels of the voice signal of real-time reception is not up to the minima of speech recognition thresholding or has exceeded
During the maximum of speech recognition thresholding, described in wake up processing module up and not perform step S230, otherwise then perform step S230.
Step S230, monitor described voice signal and meet default speech recognition thresholding when the described processing module that wakes up up
Time, whether the voice signal that checking is received can wake up terminal up.
Wake up processing module described in here, up when monitoring described voice signal and meeting default speech recognition thresholding, test
The voice signal that card is received carries out mating of waveform with the voice signal prestored.
A kind of alternative is, before verifying whether the voice signal received can wake up terminal up, and described step S230
Also include: whether the duration that the voice signal of the satisfied speech recognition thresholding preset of judgement is lasting verifies duration beyond presetting,
The most whether checking voice signal can wake up terminal up, if it is not, then whether checking voice signal can wake up terminal up.
Such as, if the lasting duration of the voice signal meeting above-mentioned speech recognition thresholding is beyond presetting checking duration, then
Illustrating that this voice signal is not the command signal waking up terminal up, on this basis, not whether checking voice signal can wake up end up
End.Otherwise, then whether checking voice signal can wake up terminal up.
On the basis of above-mentioned each optional embodiment, described in wake up processing module up and extract the feature in described voice signal
Information, mates institute's characteristic information extraction with the characteristic information preserved, and determines whether to wake up end up based on matching result
End, wherein, the characteristic information preserved includes: wake up the signature waveform in voice signal up and/or according to waking up voice signal institute up
The text message of translation.
Processing module is waken up up according to embodiment one arranges feature information extraction side when waking up voice signal up described in here,
Formula extracts the characteristic information of the voice messaging received in the dormant state, and by the characteristic information extracted with preserved
Characteristic information mates, and determines whether to wake up terminal up according to matching result.
Such as, if the signature waveform in the voice signal received amplified/reduce and the signature waveform kissing that preserved
Close, it is determined that be proved to be successful.
In a kind of alternative, described step S230 includes: step S231, S232.(being all unillustrated)
S231, determine meet preset speech recognition thresholding time, receive preset checking duration voice signal.
Specifically, system is waken up up when determining that received voice signal meets speech recognition thresholding, when will determine described in
Carve as initial time, and receive the voice signal of the checking duration preset.
S232, the characteristic information extracted in described voice signal, by institute's characteristic information extraction and the characteristic information preserved
Mate, and determine whether to wake up terminal up based on matching result.
In the present embodiment, described in wake up system mode as described above up, from received checking duration voice letter
Characteristic information extraction in number.
If waking up processing module described in it should be noted that up to be performed to wake up proof procedure up by MCU, then in determining and waking up up
Time disconnected, described in wake up system up when waking up terminal up, by described speech reception module by wake up up processing module switch to main process single
Unit.
The system that wakes up up described in here, wakes up, based on described, the information the being proved to be successful main process of startup that processing module is provided up
Unit and other hardware cells, and it is single to indicate Main Processor Unit that by MCU, the control of speech reception module is switched to main process
Unit.
The present embodiment is by when voice signal meets default speech recognition thresholding, and the voice signal that checking is received is
No can wake up terminal up, it is possible to effectively prevent the power consumption problems of too of terminal during dormancy.It addition, utilize MCU to reduce monitoring
The power consumption of voice signal.It addition, when the duration that the voice signal judging the satisfied speech recognition thresholding preset is lasting surpasses
When going out to preset checking duration, just whether checking voice signal matches with default characteristic information, then can effectively reduce checking
Number of times, to reduce the energy loss during dormancy as far as possible.It addition, the voice signal only accepting checking duration carries out waking up up testing
Card, it is possible to be effectively improved verification efficiency.
Embodiment four
As shown in Figure 4, the present embodiment provides a kind of and wakes up system up for terminal.The described system that wakes up up is in not in terminal
During dormancy state, waken up up the situation of terminal by monitoring and recognition of speech signals.The described system that wakes up up is installed in the terminal.Institute
State terminal to include but not limited to: mobile phone, VR device, wearable device etc., wherein, be excellent with VR device.The described system 2 that wakes up up is wrapped
Include: start module 21, speech reception module 22, wake up processing module 23 up.
Described startup module 21 for when terminal proceeds to resting state, starts and wakes up processing module 23 up in terminal.
Specifically, when terminal is not because operating for a long time or proceeding to resting state under dormancy instructs, described startup module
Processing module 23 is waken up up described in 21 startups.
Here, described startup module 21 (can be comprised the hardware list of primary processor and internal memory by the Main Processor Unit of terminal
Unit) perform described in wake up the checking wakeup process of processing module 23 up.
In a kind of alternative, described in wake up processing module 23 up and include MCU and peripheral circuit thereof.Described MCU connects eventually
Main Processor Unit in end.
When described startup module 21 monitor terminal proceed to resting state time, by the speech reception module 22 in terminal from end
Main Processor Unit in end switches to wake up processing module 23 up.
Specifically, when described startup module 21 monitor terminal proceed to resting state time, control primary processor to MCU export
Enabled instruction so that wake up processing module 23 up and perform default monitoring and wake up proof procedure up.
Described speech reception module 22 wakes up process up described in being delivered to by the voice signal received during dormancy
Module 23.
Here, described speech reception module 22 indicates primary processor to transfer to wake up place up by the control of speech reception module 22
Reason module 23.During dormancy, the voice that speech reception module 22 (such as mike) is received by described speech reception module 22
Signal exports to waking up processing module 23 up.The described processing module 23 real-time reception voice signal that wakes up up, and perform wake up up authenticated
Journey.Wherein, proof procedure is waken up described in up not necessarily for software program, it is also possible to be combined with waking up up of hardware circuit and software
Proof procedure.
In waking up proof procedure up, described in wake up processing module 23 up and determine whether to open according to default speech recognition thresholding
Move and wake up checking up.
Such as, it is not up to the minima of speech recognition thresholding or beyond language when the frequency of the voice signal of real-time reception
During the maximum of sound identification thresholding, described in wake up processing module 23 not execution up and wake up proof procedure up, otherwise then perform to wake up up to test
Card process.
And for example, when the decibels of the voice signal of real-time reception is not up to the minima of speech recognition thresholding or has exceeded
During the maximum of speech recognition thresholding, described in wake up processing module 23 not execution up and wake up proof procedure up, otherwise then perform to wake up up
Proof procedure.
When described wake up up processing module 23 monitor described voice signal meet default speech recognition thresholding time, verify institute
Whether the voice signal received can wake up terminal up.
Processing module 23 is waken up up when monitoring described voice signal and meeting default speech recognition thresholding described in here,
The voice signal that checking is received carries out mating of waveform with the voice signal prestored.
A kind of alternative is, before whether the voice signal received in checking can wake up terminal up, described in wake up process up
Module 23 is additionally operable to judge whether to meet the lasting duration of the voice signal of speech recognition thresholding preset beyond presetting checking
Duration, the most whether checking voice signal can wake up terminal up, if it is not, then whether checking voice signal can wake up terminal up.
Such as, if the lasting duration of the voice signal meeting above-mentioned speech recognition thresholding is beyond presetting checking duration, then
Illustrating that this voice signal is not the command signal waking up terminal up, on this basis, not whether checking voice signal can wake up end up
End.Otherwise, then whether checking voice signal can wake up terminal up.
On the basis of above-mentioned each optional embodiment, described in wake up processing module 23 up and extract the spy in described voice signal
Reference ceases, and is mated with the characteristic information preserved by institute's characteristic information extraction, and determines whether to wake up up based on matching result
Terminal, wherein, the characteristic information preserved includes: wake up the signature waveform in voice signal up and/or according to waking up voice signal up
The text message translated.
Processing module 23 is waken up up according to embodiment one arranges feature information extraction when waking up voice signal up described in here,
Mode extracts the characteristic information of the voice messaging received in the dormant state, and by the characteristic information extracted with preserved
Characteristic information mate, and determine whether to wake up terminal up according to matching result.
Such as, if the signature waveform in the voice signal received amplified/reduce and the signature waveform kissing that preserved
Close, it is determined that be proved to be successful.
In a kind of alternative, described in wake up processing module 23 up and be additionally operable to meet the speech recognition thresholding preset determining
Time, receive the voice signal of the checking duration preset;And for extracting the characteristic information in described voice signal, will be extracted
Characteristic information mates with the characteristic information preserved, and determines whether to wake up terminal up based on matching result.
Specifically, described in wake up processing module 23 up when determining that received voice signal meets speech recognition thresholding, will
Determine that the moment as initial time, and receives the voice signal of the checking duration preset.
In the present embodiment, described in wake up processing module 23 mode as described above up, from the checking duration received
Characteristic information extraction in voice signal.
If waking up processing module 23 described in it should be noted that up to be performed to wake up proof procedure up by MCU, then wake up up determining
During interruption, described in wake up processing module 23 up when waking up terminal up, described speech reception module 22 is cut by waking up processing module 23 up
Shift to Main Processor Unit.
The information being proved to be successful that waking up processing module 23 described in here, up is provided starts Main Processor Unit and other hardware
Unit, and indicate Main Processor Unit that by MCU, the control of speech reception module 22 is switched to Main Processor Unit.
Note, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious change,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although by above example, the present invention is carried out
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other Equivalent embodiments more can be included, and the scope of the present invention is determined by scope of the appended claims.