CN106297777A - A kind of method and apparatus waking up voice service up - Google Patents
A kind of method and apparatus waking up voice service up Download PDFInfo
- Publication number
- CN106297777A CN106297777A CN201610656147.5A CN201610656147A CN106297777A CN 106297777 A CN106297777 A CN 106297777A CN 201610656147 A CN201610656147 A CN 201610656147A CN 106297777 A CN106297777 A CN 106297777A
- Authority
- CN
- China
- Prior art keywords
- voice
- threshold
- acoustic features
- score value
- waking
- 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.)
- Granted
Links
- 230000002618 waking effect Effects 0.000 title claims abstract description 81
- 238000000034 method Methods 0.000 title claims abstract description 43
- 230000008878 coupling Effects 0.000 claims abstract description 97
- 238000010168 coupling process Methods 0.000 claims abstract description 97
- 238000005859 coupling reaction Methods 0.000 claims abstract description 97
- 239000000284 extract Substances 0.000 claims abstract description 5
- 230000013011 mating Effects 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 238000010030 laminating Methods 0.000 description 1
- 238000011022 operating instruction Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Electric Clocks (AREA)
Abstract
The invention discloses a kind of method and apparatus waking up voice service up, described method includes: obtain the voice of input；Extract the acoustic features of described voice；The acoustic features that the acoustic features of described voice wakes up word up with the standard pre-set is mated, obtains the first coupling score value；If described first coupling score value is more than the first threshold pre-set, then wake up voice service up；If described first coupling score value is less than described first threshold, but more than the Second Threshold pre-set, then described voice is carried out record, and the acoustic features of the acoustic features of this record voice with last record voice is mated, obtain the second coupling score value；If described second coupling score value is more than the 3rd threshold value pre-set, then increases and repeatedly input counting；When described repeatedly input counting more than pre-set four threshold values time, wake up voice service up.Use the present invention can improve the accuracy that voice service wakes up up, improve Consumer's Experience.
Description
Technical field
The present invention relates to field of computer technology, particularly relate to a kind of method and apparatus waking up voice service up.
Background technology
In recent years, the smart machine with voice operating function has progressed into the life of people, and this kind equipment is using
Before voice operating, it is generally required to wake up voice operating function up by specifically waking up word up.
In the awakening method of existing voice service, the generally phonetic entry to user is calculated by specific algorithm,
Wake up word up to carry out mating marking with default standard, if final calculated mark is more than the threshold value preset, wake up language up
Sound services, and does not wakes up up.
But, due to the difference of user life living environment, often with certain accent, or due to disease
Impact, cause cacology true, use above-mentioned existing awakening method, the coupling marking nothing waking up word sound up of user's input
Method exceedes default threshold value, even if user is iteratively repeated wakes up word up, it is also difficult to the voice service of wake-up device, user exactly
Experience poor.
Summary of the invention
The purpose of the embodiment of the present invention is to provide a kind of method and apparatus waking up voice service up, improves voice service and calls out
The accuracy waken up, improves Consumer's Experience.
For achieving the above object, embodiments provide a kind of method waking up voice service up, including:
Obtain the voice of input；
Extract the acoustic features of described voice；
The acoustic features that the acoustic features of described voice wakes up word up with the standard pre-set is mated, obtains first
Coupling score value；
If described first coupling score value is more than the first threshold pre-set, then wake up voice service up；
If described first coupling score value is less than described first threshold, but more than the Second Threshold pre-set, then to described
Voice carries out record, and the acoustic features of the acoustic features of this record voice with last record voice is mated, and obtains
Take the second coupling score value；
If described second coupling score value is more than the 3rd threshold value pre-set, then increases and repeatedly input counting；
When described repeatedly input counting more than pre-set four threshold values time, wake up voice service up.
The voice that the present invention is between first threshold and Second Threshold by the coupling score value repeatedly inputted user enters
Row special handling, mates this input voice with last input voice, identical to know that user inputs the most repeatedly
Or close voice, when user repeatedly input same or like voice exceed certain number of times time, the language of normal wakeup equipment
Sound services, and improves user and wakes up the accuracy of voice service up, is conducive to improving Consumer's Experience.
Further, described when described repeatedly input counting more than pre-set four threshold values time, wake up up voice clothes
After business, also include:
Whether it is successfully made voice operating after judging to wake up voice service up；If being successfully made voice operating, then increase
Successful Operation counts；If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting；
When described Successful Operation counting is more than the 5th threshold value pre-set, and the acoustic features of described voice is not deposited
It is stored in alternative when waking up dictionary up, the acoustic features of described voice is stored in alternative waking up up in dictionary；
When described Successful Operation counting is less than described 5th threshold value, and the acoustic features of described voice has been stored in institute
State alternative when waking up dictionary up, the acoustic features of described voice is deleted from described alternative waking up up dictionary.
Preferably, if described first coupling score value is less than described first threshold, but more than the second threshold pre-set
Value, then carry out record to described voice, and by the acoustic features of the acoustic features of this record voice with last record voice
Mate, obtain the second coupling score value, particularly as follows:
If described first coupling score value is less than the first threshold pre-set, but more than the Second Threshold pre-set, then
The acoustic features of described voice is mated with the described alternative acoustic features waken up up in dictionary, obtains the 3rd coupling score value；
If described 3rd coupling score value is more than the 6th threshold value pre-set, then wake up voice service up；
If described 3rd coupling score value is less than described 6th threshold value, then described voice is carried out record, and by this record
The acoustic features of voice is mated with the acoustic features of last record voice, obtains described second coupling score value.
The present invention, after waking up voice service up, updates according to follow-up voice operating situation and alternative wakes up dictionary up, when with
When family is successfully made voice operating, is regarded as effective voice and waken up up, and by described voice after the most effectively waking up up
Add alternative waking up up in dictionary to；And unsuccessful carry out voice operating time, regarded as false wake-up, reduced the one-tenth of this voice
Merit operation count, and when described Successful Operation counting is less than certain value, described voice is deleted from alternative waking up up dictionary, carries
The high alternative effectiveness waking up word up.In follow-up voice wakes up up, the present invention also utilizes the described alternative history waken up up in dictionary
Record, the voice being in coupling score value between first threshold and Second Threshold carries out match cognization, is conducive to improving identification
Efficiency and accuracy.
Further, if mating score value more than the 3rd threshold value pre-set described second, then repetition is increased
After input counting, also include:
Repeatedly input described in if to count and do not change within the time limited, then the described counting that repeatedly inputs is reset.
Preferably, described method also includes:
If described first coupling score value is less than described Second Threshold, then judges to wake up voice service up, and reacquire
The voice of input；
If described second coupling score value is less than described 3rd threshold value, then judges to wake up voice service up, and reacquire
The voice of input；
Repeatedly input counting described in and if be less than described 4th threshold value, then judge to wake up voice service up, and reacquire
The voice of input.
Correspondingly, present invention also offers a kind of device waking up voice service up, including:
Voice acquisition module, for obtaining the voice of input；
Characteristic extracting module, for extracting the acoustic features of described voice；
First matching module, for waking up the acoustic features of word up by the acoustic features of described voice and the standard pre-set
Mate, obtain the first coupling score value；
First wake module, if for described first coupling score value more than the first threshold pre-set, then waking up voice up
Service；
Second matching module, if for described first coupling score value less than described first threshold, but more than pre-setting
Second Threshold, then carry out record to described voice, and by the sound of the acoustic features of this record voice with last record voice
Learn feature to mate, obtain the second coupling score value；
Repeat count module, if for described second coupling score value more than the 3rd threshold value pre-set, then increasing repetition
Input counting；
Second wake module, for when described repeatedly input counting more than pre-set four threshold values time, wake up voice up
Service.
Further, described device also includes:
Whether operation count module, be successfully made voice operating after judgement wakes up voice service up；If successfully carrying out
Voice operating, then increase Successful Operation counting；If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting；
Dictionary increases module, for being more than the 5th threshold value pre-set, and institute's predicate when described Successful Operation counting
The acoustic features of sound is not stored in alternative when waking up dictionary up, and the acoustic features of described voice is stored in alternative waking up up in dictionary；
Module deleted in dictionary, for being less than described 5th threshold value, and the sound of described voice when described Successful Operation counting
Learn feature and be stored in described alternative when waking up dictionary up, the acoustic features of described voice is deleted from described alternative waking up up dictionary
Remove.
Preferably, described second matching module includes:
First matching unit, if for described first coupling score value less than the first threshold pre-set, but more than in advance
The Second Threshold arranged, then mate the acoustic features of described voice with the described alternative acoustic features waken up up in dictionary,
Obtain the 3rd coupling score value；
First wakeup unit, if for described 3rd coupling score value more than the 6th threshold value pre-set, then waking up voice up
Service；
Second matching unit, if for described 3rd coupling score value less than described 6th threshold value, then carrying out described voice
Record, and the acoustic features of this record voice being mated with the last acoustic features recording voice, obtains described the
Two coupling score values.
Further, described device also includes:
Counting dump block, does not changes, then by described within the time limited if repeatedly inputting described in Yong Yu to count
Repeatedly input counting to reset.
Preferably, described device also includes:
First retries module, if for described first coupling score value less than described Second Threshold, then judging to wake up language up
Sound services, and reacquires the voice of input；
Second retries module, if for described second coupling score value less than described 3rd threshold value, then judging to wake up language up
Sound services, and reacquires the voice of input；
Triple die trial blocks, if repeatedly inputting counting described in Yong Yu to be less than described 4th threshold value, then judge to wake up language up
Sound services, and reacquires the voice of input.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of an embodiment of the method waking up voice service up that the present invention provides；
Fig. 2 is the method flow diagram of another embodiment of the method waking up voice service up that the present invention provides；
Fig. 3 is the method flow diagram of another embodiment of the method waking up voice service up that the present invention provides；
Fig. 4 is the structure drawing of device of an embodiment of the method waking up voice service up that the present invention provides；
Fig. 5 is the structure drawing of device of another embodiment of the method waking up voice service up that the present invention provides；
Fig. 6 is the structure drawing of device of another embodiment of the method waking up voice service up that the present invention provides.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
See Fig. 1, be the method flow diagram of an embodiment of the method waking up voice service up that the present invention provides.
The method waking up voice service described in as shown in Figure 1 up, including step S11～S17:
S11, obtains the voice of input.
In the middle of being embodied as, sound pick-up outfit (such as mike etc.) can be passed through and obtain the phonetic entry of user.
S12, extracts the acoustic features of described voice.
Described acoustic features is the feature with certain distinction extracted based on human hearing characteristic, as chosen voice
MFCC (Mel-Frequency Cepstrum Coefficient, the Mel frequency cepstrum coefficient) feature used in identification, this
Skilled person can choose different acoustic featuress according to actual needs, it is possible to extracts being combined of multiple acoustic features,
This is not construed as limiting by the present invention.
S13, mates the acoustic features that the acoustic features of described voice wakes up word up with the standard pre-set, and obtains
First coupling score value.
S14, if described first coupling score value is more than the first threshold pre-set, then wakes up voice service up.
S15 is if described first coupling score value is less than described first threshold, but more than the Second Threshold pre-set, the most right
Described voice carries out record, and the acoustic features of the acoustic features of this record voice with last record voice is carried out
Join, obtain the second coupling score value.
In the middle of being embodied as, described first threshold and described Second Threshold are two threshold values chosen in advance, described
Second Threshold is less than described first threshold, and if first threshold can be 90, described Second Threshold can be 80.When described first
When partition value is more than first threshold, illustrates that user has said exactly and wake up word up, can directly wake up voice service up, after waiting user
Continuous voice operating instruction.When the first coupling score value is less than first threshold, continue to judge whether this score value is more than Second Threshold,
If described first coupling score value is less than first threshold but more than Second Threshold, illustrate that input voice wakes up word up with standard and more connects
Closely, but also it is not enough to directly wake up voice service up.It is defeated that coupling score value is between first threshold and Second Threshold by the present invention
Enter voice and carry out special handling, this input voice was inputted voice with last time and carries out characteristic matching, to judge whether that user is
No at the same voice of typing repeatedly.
Especially, when this voice inputted is for inputting first, the most there is not a record voice and (be equivalent to upper one
Secondary record voice is empty), then this input voice carries out with last registration voice mating the second coupling score value obtained is 0, this
Time can be set to 1 by repeatedly inputting counting, with show described voice for inputting first, and to this input voice carry out record, with
Mate with follow-up input voice, it is judged that whether subsequent voice is for repeatedly inputting.
S16, if described second coupling score value is more than the 3rd threshold value pre-set, then increases and repeatedly inputs counting.
S17, when described repeatedly input counting more than pre-set four threshold values time, wake up voice service up.
Repeatedly input counting and repeatedly input the number of times of same or like voice for recording user, when described second mates
When score value is more than three threshold value pre-set, it is believed that voice of this input of user identical with the last voice recorded or
Person is closer to, and repeatedly inputs counting and adds one.If described second coupling score value is less than described 3rd threshold value, then described in repeatedly input
Counting resets.When user repeatedly inputs four threshold value that the number of times of same voice reaches default, though i.e. explanation user pronunciation has
Deviation, but user wakes up voice service up making repeated attempts, and the present invention will normally wake up voice service up in the case, be conducive to
Improve Consumer's Experience.
Such as, the word that wakes up up of the voice service function of certain smart machine is " Intelligent life ", certain user due to accent problem,
Said into " self energy god fire " wake up word up, according to existing awakening method, cannot owing to waking up the score value that mates of word up with standard
Reach first threshold, even if user's repeated multiple times input voice also cannot wake up voice service up.And provide according to the present embodiment
The method waking up voice service up, when in he short time repeated multiple times attempt to wake up up time, when the number of times repeated reaches four threshold values,
Equipment just can be with normal wakeup.
It should be noted that, the present invention for coupling score value equal with threshold value in the case of, both can enter by situation about being more than
Row processes, it is also possible to processing by situation about being less than, those skilled in the art can select according to practical situation, the present invention
This is not construed as limiting.In like manner, repeatedly input counting to exist with the 5th threshold value with the 4th threshold value and the counting of Successful Operation hereafter
Also can select one by situation about being more than or less than in the case of equal to process.More excellent, can take by choosing suitable threshold value
Value is to avoid the situation appearance being equal to.
See Fig. 2, be the method flow diagram of another embodiment of the method waking up voice service up that the present invention provides.This
Embodiment adds step S21～S23 on the basis of embodiment illustrated in fig. 1 further.
As in figure 2 it is shown, the present embodiment is after step S17, also include:
S21, it is judged that whether be successfully made voice operating after waking up voice service up.If being successfully made voice operating, then
Increase Successful Operation counting.If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting.
S22, when described Successful Operation counting is more than the 5th threshold value pre-set, and the acoustic features of described voice is not
It is stored in alternative when waking up dictionary up, the acoustic features of described voice is stored in alternative waking up up in dictionary.
S23, when described Successful Operation counting is less than described 5th threshold value, and the acoustic features of described voice is stored
In described alternative wake up dictionary up time, by the acoustic features of described voice from described alternative wake up up dictionary delete.
In the middle of being embodied as, when voice service is waken up and user is successfully made voice operating, can be by this
Wake up up and regard as effective voice and wake up up, and add described voice to alternative waking up up in dictionary after the most effectively waking up up.And
When voice service is waken up but user does not carry out voice operating or have input voice command that system cannot understand, can
This is waken up up and regards as false wake-up, reduce the Successful Operation counting of this voice, and at Successful Operation counting less than certain value
Time, described voice is deleted from alternative waking up up dictionary.By the way the use history of user is carried out record, waking up up
After voice service, update according to the voice operating situation that user is follow-up and alternative wake up dictionary up, be conducive to improving and wake up having of word up
Effect property, improves the efficiency of follow-up identification, the use habit of laminating user.
Preferred embodiment central in one, described S15 step includes:
S151, if described first coupling score value is less than the first threshold pre-set, but more than the second threshold pre-set
Value, then mate the acoustic features of described voice with the described alternative acoustic features waken up up in dictionary, obtains the 3rd coupling
Score value.
S152, if described 3rd coupling score value is more than the 6th threshold value pre-set, then wakes up voice service up.
S153, if described 3rd coupling score value is less than described 6th threshold value, then carries out record to described voice, and by this
The acoustic features of record voice is mated with the acoustic features of last record voice, obtains described second coupling score value.
When the acoustic features of user input voice is present in alternative waking up up in dictionary, user calls out without making repeated attempts
Wake up, even if deviation of pronouncing also can wake up voice service directly, rapidly up.Meanwhile, the dynamic of dictionary is waken up up in conjunction with alternative
Update, the use habit of the user that can fit, be greatly improved the efficiency waking up voice service up.
And when the first coupling score value of input voice is less than the first threshold pre-set, but more than second pre-set
Threshold value, and when being not present in alternative waking up up in dictionary, then this voice is recorded (voice of record for and the last time
Record voice and record next time voice comparison coupling), and enter judge whether to be repeated several times the handling process of input, with
Obtain the number of times that described voice repeatedly inputs, when its number of times repeatedly input reaches corresponding threshold value, the voice of wake-up device
Service.The present embodiment only wakes up word up and cannot be waken up up the voice of voice service carried out by the alternative dictionary that wakes up up being near the mark
Record, and calculate it and repeatedly input number of times, advantageously reduce program internal memory and take and amount of calculation, improve the sound of voice call service
Answer speed.
See Fig. 3, be the method flow diagram of another embodiment of the method waking up voice service up that the present invention provides.This
Embodiment, on the basis of Fig. 1 or embodiment illustrated in fig. 2, further increases step S41.
As it is shown on figure 3, the present embodiment is after described step S16, also include:
S41, if described in repeatedly input to count and do not change within the time limited, then to repeatedly input counting clear by described
Zero.
Generally, when repeatedly inputting same or like voice at short notice, just think that user is making repeated attempts
Wake up up.Repeatedly input described in if to count and do not change within the time limited, illustrate that current speech and last time input voice
Time interval is relatively big, does not constitute and repeatedly inputs, and can be reset by the described counting that repeatedly inputs, be conducive to improving further waking up voice up
The accuracy of service.
Though being not drawn in figure, but in the middle of being embodied as, what Fig. 1, Fig. 2 or embodiment illustrated in fig. 3 provided wakes up voice clothes up
The method of business also includes:
If described first coupling score value is less than described Second Threshold, then judges to wake up voice service up, and reacquire
The voice of input.
If described second coupling score value is less than described 3rd threshold value, then judges to wake up voice service up, and reacquire
The voice of input.
Repeatedly input counting described in and if be less than described 4th threshold value, then judge to wake up voice service up, and reacquire
The voice of input.
If described first coupling score value is less than described Second Threshold, illustrate that the input of user is the most inaccurate, it is not necessary to wake up up
Voice service, and terminate the process to this section of voice, continue to gather new voice.If described second coupling score value is less than described the
Three threshold values, illustrate the voice of twice input before and after user and differ, user unused same or like voice make repeated attempts
Wake up up, now can determine that and can not wake up voice service up, and terminate the process to this section of voice, continue to gather new voice.If institute
State and repeatedly input counting less than described 4th threshold value, illustrate that user repeats the number of times of typing voice and not yet reaches to wake up voice service up
Threshold value, now without waking up voice service up, and terminate the process to this section of voice, continue to gather new voice.
See Fig. 4, be the structure drawing of device of an embodiment of the device waking up voice service up that the present invention provides.This reality
The ultimate principle executing example is consistent with embodiment illustrated in fig. 1, part is not described in detail in detail in the present embodiment and can be found in embodiment illustrated in fig. 1
Associated description.
As shown in Figure 4, the device waking up voice service described in up includes:
Voice acquisition module 11, for obtaining the voice of input.
Characteristic extracting module 12, for extracting the acoustic features of described voice.
First matching module 13, special for the acoustic features of described voice and the standard pre-set being waken up up the acoustics of word
Levy and mate, obtain the first coupling score value.
First wake module 14, if for described first coupling score value more than the first threshold pre-set, then waking up language up
Sound services.
Second matching module 15, if for described first coupling score value less than described first threshold, but more than pre-setting
Second Threshold, then described voice is carried out record, and by the acoustic features of this record voice and last record voice
Acoustic features is mated, and obtains the second coupling score value.
Repeat count module 16, if for described second coupling score value more than the 3rd threshold value pre-set, then increasing weight
Input counting again.
Second wake module 17, for when described repeatedly input counting more than pre-set four threshold values time, wake up language up
Sound services.
The voice that the present invention is between first threshold and Second Threshold by the coupling score value repeatedly inputted user enters
Row special handling, inputted voice with last time mate this input voice, with know user the most repeatedly input identical or
Close voice, when user repeatedly input same or like voice exceed certain number of times time, the voice of normal wakeup equipment
Service, improves user and wakes up the accuracy of voice service up, be conducive to improving Consumer's Experience.
See Fig. 5, be the structure drawing of device of another embodiment of the device waking up voice service up that the present invention provides.This
Embodiment farther includes on the basis of embodiment illustrated in fig. 4:
Whether operation count module 21, be successfully made voice operating after judgement wakes up voice service up.If successfully entering
Go voice operating, then increase Successful Operation counting.If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting.
Dictionary increases module 22, for being more than, when described Successful Operation counting, the 5th threshold value pre-set and described
The acoustic features of voice is not stored in alternative when waking up dictionary up, the acoustic features of described voice is stored in and alternative wakes up dictionary up
In.
Module 23 deleted in dictionary, for being less than described 5th threshold value when described Successful Operation counts, and described voice
Acoustic features has been stored in described alternative when waking up dictionary up, the acoustic features of described voice alternative is waken up up dictionary from described
Delete.
Preferred embodiment central in one, described second matching module 15 includes:
First matching unit 151, if for described first coupling score value less than the first threshold pre-set, but more than pre-
The Second Threshold first arranged, then carried out the acoustic features of described voice and the described alternative acoustic features waken up up in dictionary
Join, obtain the 3rd coupling score value.
First wakeup unit 152, if for described 3rd coupling score value more than the 6th threshold value pre-set, then waking up language up
Sound services.
Second matching unit 153, if for described 3rd coupling score value less than described 6th threshold value, then entering described voice
Row record, and the acoustic features of the acoustic features of this record voice with last record voice is mated, obtain described
Second coupling score value.
The present invention, after waking up voice service up, updates according to follow-up voice operating situation and alternative wakes up dictionary up, when with
When family is successfully made voice operating, is regarded as effective voice and waken up up, and by described voice after the most effectively waking up up
Add alternative waking up up in dictionary to.And unsuccessful carry out voice operating time, regarded as false wake-up, reduced the one-tenth of this voice
Merit operation count, and when Successful Operation counting is less than certain value, described voice is deleted from alternative waking up up dictionary, improves
The alternative effectiveness waking up word up.In follow-up voice wakes up up, the present invention also utilizes the described alternative history note waken up up in dictionary
Record, the voice being in coupling score value between first threshold and Second Threshold carries out match cognization, is conducive to improving the effect identified
Rate and accuracy.
See Fig. 6, be the structure drawing of device of another embodiment of the device waking up voice service up that the present invention provides.This
Embodiment farther includes on the basis of Fig. 4 or embodiment illustrated in fig. 5:
Counting dump block 41, does not changes, then by institute within the time limited if repeatedly inputting described in Yong Yu to count
State and repeatedly input counting clearing.
When the acoustic features of user input voice is present in alternative waking up up in dictionary, user calls out without making repeated attempts
Wake up, even if deviation of pronouncing also can wake up voice service directly, rapidly up.Meanwhile, the dynamic of dictionary is waken up up in conjunction with alternative
Update, be greatly improved the efficiency waking up voice service up.
Though figure is not drawn into, but in the middle of being embodied as, wake up voice up such as what Fig. 4, Fig. 5 or embodiment illustrated in fig. 6 provided
The device of service also includes:
First retries module, if for described first coupling score value less than described Second Threshold, then judging to wake up language up
Sound services, and reacquires the voice of input.
Second retries module, if for described second coupling score value less than described 3rd threshold value, then judging to wake up language up
Sound services, and reacquires the voice of input.
Triple die trial blocks, if repeatedly inputting counting described in Yong Yu to be less than described 4th threshold value, then judge to wake up language up
Sound services, and reacquires the voice of input.
It should be noted that, device embodiment described above is only schematically, wherein said as separating component
The unit illustrated can be or may not be physically separate, and the parts shown as unit can be or can also
It not physical location, i.e. may be located at a place, or can also be distributed on multiple NE.Can be according to reality
Need to select some or all of module therein to realize the purpose of the present embodiment scheme.It addition, the device that the present invention provides
In embodiment accompanying drawing, the annexation between module represents have communication connection between them, specifically can be implemented as one or
A plurality of communication bus or holding wire.Those of ordinary skill in the art, in the case of not paying creative work, are i.e. appreciated that
And implement.
Through the above description of the embodiments, those skilled in the art is it can be understood that can borrow to the present invention
The mode helping software to add required common hardware realizes, naturally it is also possible to include special IC, specially by specialized hardware
Realize with CPU, private memory, special components and parts etc..Generally, all functions completed by computer program can
Realize with corresponding hardware easily, and, the particular hardware structure being used for realizing same function can also be multiple many
Sample, such as analog circuit, digital circuit or special circuit etc..But, the most more in the case of software program real
It it is now more preferably embodiment.Based on such understanding, prior art is made by technical scheme the most in other words
The part of contribution can embody with the form of software product, and this computer software product is stored in the storage medium that can read
In, such as the floppy disk of computer, USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory), random access memory
Device (RAM, Random Access Memory), magnetic disc or CD etc., including some instructions with so that a computer sets
Standby (can be personal computer, server, or the network equipment etc.) performs the method described in each embodiment of the present invention.
The above, the only detailed description of the invention of the present invention, but protection scope of the present invention is not limited thereto, and any
Those familiar with the art, in the technical scope that the invention discloses, can readily occur in change or replace, should contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.
Claims (10)
1. the method waking up voice service up, it is characterised in that including:
Obtain the voice of input；
Extract the acoustic features of described voice；
The acoustic features that the acoustic features of described voice wakes up word up with the standard pre-set is mated, obtains the first coupling
Score value；
If described first coupling score value is more than the first threshold pre-set, then wake up voice service up；
If described first coupling score value is less than described first threshold, but more than the Second Threshold pre-set, then to described voice
Carry out record, and the acoustic features of the acoustic features of this record voice with last record voice is mated, acquisition the
Two coupling score values；
If described second coupling score value is more than the 3rd threshold value pre-set, then increases and repeatedly input counting；
When described repeatedly input counting more than pre-set four threshold values time, wake up voice service up.
The method waking up voice service the most as claimed in claim 1 up, it is characterised in that repeatedly inputting counting described in described working as
During more than four threshold value pre-set, after waking up voice service up, also include:
Whether it is successfully made voice operating after judging to wake up voice service up；If being successfully made voice operating, then increase successfully
Operation count；If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting；
When described Successful Operation counting is more than the 5th threshold value pre-set, and the acoustic features of described voice is not stored in
Alternative when waking up dictionary up, the acoustic features of described voice is stored in alternative waking up up in dictionary；
When described Successful Operation counting is less than described 5th threshold value, and the acoustic features of described voice has been stored in described standby
When choosing wakes up dictionary up, the acoustic features of described voice is deleted from described alternative waking up up dictionary.
The method waking up voice service the most as claimed in claim 2 up, it is characterised in that if described first coupling score value is little
In described first threshold, but more than the Second Threshold pre-set, then described voice is carried out record, and this is recorded voice
Acoustic features mate with the acoustic features of last record voice, obtain the second coupling score value, particularly as follows:
If described first coupling score value is less than the first threshold pre-set, but more than the Second Threshold pre-set, then by institute
The acoustic features of predicate sound is mated with the described alternative acoustic features waken up up in dictionary, obtains the 3rd coupling score value；
If described 3rd coupling score value is more than the 6th threshold value pre-set, then wake up voice service up；
If described 3rd coupling score value is less than described 6th threshold value, then described voice is carried out record, and this is recorded voice
Acoustic features mate with the acoustic features of last record voice, obtain described second coupling score value.
The method waking up voice service the most as claimed in claim 1 up, it is characterised in that if mating score value described second
More than the 3rd threshold value pre-set, then increase after repeatedly inputting counting, also include:
Repeatedly input described in if to count and do not change within the time limited, then the described counting that repeatedly inputs is reset.
5. the method waking up voice service up as described in any one of claim 1 to 4, it is characterised in that described method also includes:
If described first coupling score value is less than described Second Threshold, then judges to wake up voice service up, and reacquire input
Voice；
If described second coupling score value is less than described 3rd threshold value, then judges to wake up voice service up, and reacquire input
Voice；
Repeatedly input counting described in and if be less than described 4th threshold value, then judge to wake up voice service up, and reacquire input
Voice.
6. the device waking up voice service up, it is characterised in that including:
Voice acquisition module, for obtaining the voice of input；
Characteristic extracting module, for extracting the acoustic features of described voice；
First matching module, for carrying out the acoustic features that the acoustic features of described voice and the standard pre-set wake up word up
Coupling, obtains the first coupling score value；
First wake module, if for described first coupling score value more than the first threshold pre-set, then waking up voice service up；
Second matching module, if for described first coupling score value less than described first threshold, but more than second pre-set
Threshold value, then carry out record to described voice, and the acoustic features of this record voice is special with the acoustics of last record voice
Levy and mate, obtain the second coupling score value；
Repeat count module, if for described second coupling score value more than the 3rd threshold value pre-set, then increasing and repeatedly input
Counting；
Second wake module, for when described repeatedly input counting more than pre-set four threshold values time, wake up voice service up.
Wake up the device of voice service the most as claimed in claim 6 up, it is characterised in that described device also includes:
Whether operation count module, be successfully made voice operating after judgement wakes up voice service up；If being successfully made language
Sound operates, then increase Successful Operation counting；If the unsuccessful voice operating that carries out, then reduce described Successful Operation counting；
Dictionary increases module, for being more than, when described Successful Operation counts, the 5th threshold value pre-set, and described voice
Acoustic features is not stored in alternative when waking up dictionary up, and the acoustic features of described voice is stored in alternative waking up up in dictionary；
Module deleted in dictionary, and for being less than described 5th threshold value when described Successful Operation counting, and the acoustics of described voice is special
Levy and be stored in described alternative when waking up dictionary up, the acoustic features of described voice is deleted from described alternative waking up up dictionary.
Wake up the device of voice service the most as claimed in claim 7 up, it is characterised in that described second matching module includes:
First matching unit, if for described first coupling score value less than the first threshold pre-set, but more than pre-setting
Second Threshold, then the acoustic features of described voice is mated with the described alternative acoustic features waken up up in dictionary, obtain
3rd coupling score value；
First wakeup unit, if for described 3rd coupling score value more than the 6th threshold value pre-set, then waking up voice service up；
Second matching unit, if for described 3rd coupling score value less than described 6th threshold value, then described voice being carried out record,
And the acoustic features of the acoustic features of this record voice with last record voice is mated, obtain described second coupling
Score value.
Wake up the device of voice service the most as claimed in claim 6 up, it is characterised in that described device also includes:
Counting dump block, does not changes, then by described repetition within the time limited if repeatedly inputting described in Yong Yu to count
Input counting resets.
10. the device waking up voice service up as described in any one of claim 6 to 9, it is characterised in that described device also wraps
Include:
First retries module, if for described first coupling score value less than described Second Threshold, then judging to wake up voice clothes up
Business, and reacquire the voice of input；
Second retries module, if for described second coupling score value less than described 3rd threshold value, then judging to wake up voice clothes up
Business, and reacquire the voice of input；
Triple die trial blocks, if repeatedly inputting counting described in Yong Yu to be less than described 4th threshold value, then judge to wake up voice clothes up
Business, and reacquire the voice of input.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610656147.5A CN106297777B (en) | 2016-08-11 | 2016-08-11 | A kind of method and apparatus waking up voice service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610656147.5A CN106297777B (en) | 2016-08-11 | 2016-08-11 | A kind of method and apparatus waking up voice service |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106297777A true CN106297777A (en) | 2017-01-04 |
| CN106297777B CN106297777B (en) | 2019-11-22 |
Family
ID=57669603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610656147.5A Active CN106297777B (en) | 2016-08-11 | 2016-08-11 | A kind of method and apparatus waking up voice service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106297777B (en) |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106910496A (en) * | 2017-02-28 | 2017-06-30 | 广东美的制冷设备有限公司 | Intelligent electrical appliance control and device |
| CN107358954A (en) * | 2017-08-29 | 2017-11-17 | 成都启英泰伦科技有限公司 | It is a kind of to change the device and method for waking up word in real time |
| CN107622770A (en) * | 2017-09-30 | 2018-01-23 | 百度在线网络技术（北京）有限公司 | voice awakening method and device |
| CN107919124A (en) * | 2017-12-22 | 2018-04-17 | 北京小米移动软件有限公司 | Equipment awakening method and device |
| CN108320733A (en) * | 2017-12-18 | 2018-07-24 | 上海科大讯飞信息科技有限公司 | Voice data processing method and device, storage medium, electronic equipment |
| CN108509225A (en) * | 2018-03-28 | 2018-09-07 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN108538293A (en) * | 2018-04-27 | 2018-09-14 | 青岛海信电器股份有限公司 | Voice awakening method, device and smart machine |
| CN108766422A (en) * | 2018-04-02 | 2018-11-06 | 青岛海尔科技有限公司 | Response method, device, storage medium and the computer equipment of speech ciphering equipment |
| CN108777142A (en) * | 2018-06-05 | 2018-11-09 | 上海木木机器人技术有限公司 | A kind of interactive voice recognition methods and interactive voice robot based on airport environment |
| CN108899028A (en) * | 2018-06-08 | 2018-11-27 | 广州视源电子科技股份有限公司 | Voice awakening method, searching method, device and terminal |
| CN109410952A (en) * | 2018-10-26 | 2019-03-01 | 北京蓦然认知科技有限公司 | A kind of voice awakening method, apparatus and system |
| CN110046045A (en) * | 2019-04-03 | 2019-07-23 | 百度在线网络技术（北京）有限公司 | The data package processing method and device that voice wakes up |
| CN110111789A (en) * | 2019-05-07 | 2019-08-09 | 百度国际科技（深圳）有限公司 | Voice interactive method, calculates equipment and computer-readable medium at device |
| CN110166985A (en) * | 2019-05-28 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of service awaking method and applications client |
| CN110211605A (en) * | 2019-05-24 | 2019-09-06 | 珠海多士科技有限公司 | Smart machine speech sensitivity adjusting method, device, equipment and storage medium |
| CN110265018A (en) * | 2019-07-01 | 2019-09-20 | 成都启英泰伦科技有限公司 | A kind of iterated command word recognition method continuously issued |
| CN110444210A (en) * | 2018-10-25 | 2019-11-12 | 腾讯科技（深圳）有限公司 | A kind of method of speech recognition, the method and device for waking up word detection |
| CN110600029A (en) * | 2019-09-17 | 2019-12-20 | 苏州思必驰信息科技有限公司 | User-defined awakening method and device for intelligent voice equipment |
| CN110706703A (en) * | 2019-10-16 | 2020-01-17 | 珠海格力电器股份有限公司 | Voice wake-up method, device, medium and equipment |
| CN111081251A (en) * | 2019-11-27 | 2020-04-28 | 云知声智能科技股份有限公司 | Voice wake-up method and device |
| CN111192588A (en) * | 2019-12-31 | 2020-05-22 | 深圳市瑞讯云技术有限公司 | System awakening method and device |
| CN111235820A (en) * | 2018-11-29 | 2020-06-05 | 青岛海尔洗衣机有限公司 | Control method for laundry treating apparatus |
| CN111309283A (en) * | 2020-03-25 | 2020-06-19 | 北京百度网讯科技有限公司 | Voice control method and device for user interface, electronic equipment and storage medium |
| CN111524528A (en) * | 2020-05-28 | 2020-08-11 | Oppo广东移动通信有限公司 | Voice awakening method and device for preventing recording detection |
| CN112086096A (en) * | 2019-06-14 | 2020-12-15 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium |
| CN112102821A (en) * | 2019-06-18 | 2020-12-18 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium applied to electronic equipment |
| US10931814B2 (en) | 2019-05-28 | 2021-02-23 | Advanced New Technologies Co., Ltd. | Service recommendation |
| CN112908330A (en) * | 2021-03-04 | 2021-06-04 | 深圳市云希谷科技有限公司 | Voice wake-up method and device for terminal equipment and computer readable storage medium |
| CN113113007A (en) * | 2021-03-30 | 2021-07-13 | 北京金山云网络技术有限公司 | Voice data processing method and device, electronic equipment and storage medium |
| CN113590207A (en) * | 2021-07-30 | 2021-11-02 | 思必驰科技股份有限公司 | Method and device for improving awakening effect |
| CN113963695A (en) * | 2021-10-13 | 2022-01-21 | 深圳市欧瑞博科技股份有限公司 | Awakening method, awakening device, equipment and storage medium of intelligent equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102999161A (en) * | 2012-11-13 | 2013-03-27 | 安徽科大讯飞信息科技股份有限公司 | Implementation method and application of voice awakening module |
| CN103021409A (en) * | 2012-11-13 | 2013-04-03 | 安徽科大讯飞信息科技股份有限公司 | Voice activating photographing system |
| CN103714815A (en) * | 2013-12-09 | 2014-04-09 | 何永 | Voice control method and device thereof |
| US8918319B2 (en) * | 2011-06-03 | 2014-12-23 | National Chiao University | Speech recognition device and speech recognition method using space-frequency spectrum |
| CN105141919A (en) * | 2015-09-01 | 2015-12-09 | 武汉同迅智能科技有限公司 | Monitoring terminal device remotely controlled by voice |
| CN105261368A (en) * | 2015-08-31 | 2016-01-20 | 华为技术有限公司 | Voice wake-up method and apparatus |
| US20160086603A1 (en) * | 2012-06-15 | 2016-03-24 | Cypress Semiconductor Corporation | Power-Efficient Voice Activation |
| CN105575395A (en) * | 2014-10-14 | 2016-05-11 | 中兴通讯股份有限公司 | Voice wake-up method and apparatus, terminal, and processing method thereof |
| CN105741838A (en) * | 2016-01-20 | 2016-07-06 | 百度在线网络技术（北京）有限公司 | Voice wakeup method and voice wakeup device |
-
2016
- 2016-08-11 CN CN201610656147.5A patent/CN106297777B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8918319B2 (en) * | 2011-06-03 | 2014-12-23 | National Chiao University | Speech recognition device and speech recognition method using space-frequency spectrum |
| US20160086603A1 (en) * | 2012-06-15 | 2016-03-24 | Cypress Semiconductor Corporation | Power-Efficient Voice Activation |
| CN102999161A (en) * | 2012-11-13 | 2013-03-27 | 安徽科大讯飞信息科技股份有限公司 | Implementation method and application of voice awakening module |
| CN103021409A (en) * | 2012-11-13 | 2013-04-03 | 安徽科大讯飞信息科技股份有限公司 | Voice activating photographing system |
| CN103714815A (en) * | 2013-12-09 | 2014-04-09 | 何永 | Voice control method and device thereof |
| CN105575395A (en) * | 2014-10-14 | 2016-05-11 | 中兴通讯股份有限公司 | Voice wake-up method and apparatus, terminal, and processing method thereof |
| CN105261368A (en) * | 2015-08-31 | 2016-01-20 | 华为技术有限公司 | Voice wake-up method and apparatus |
| CN105141919A (en) * | 2015-09-01 | 2015-12-09 | 武汉同迅智能科技有限公司 | Monitoring terminal device remotely controlled by voice |
| CN105741838A (en) * | 2016-01-20 | 2016-07-06 | 百度在线网络技术（北京）有限公司 | Voice wakeup method and voice wakeup device |
Cited By (47)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018157542A1 (en) * | 2017-02-28 | 2018-09-07 | 广东美的制冷设备有限公司 | Smart home appliance control method and device |
| CN106910496A (en) * | 2017-02-28 | 2017-06-30 | 广东美的制冷设备有限公司 | Intelligent electrical appliance control and device |
| CN107358954A (en) * | 2017-08-29 | 2017-11-17 | 成都启英泰伦科技有限公司 | It is a kind of to change the device and method for waking up word in real time |
| CN107622770B (en) * | 2017-09-30 | 2021-03-16 | 百度在线网络技术（北京）有限公司 | Voice wake-up method and device |
| CN107622770A (en) * | 2017-09-30 | 2018-01-23 | 百度在线网络技术（北京）有限公司 | voice awakening method and device |
| CN108320733B (en) * | 2017-12-18 | 2022-01-04 | 上海科大讯飞信息科技有限公司 | Voice data processing method and device, storage medium and electronic equipment |
| CN108320733A (en) * | 2017-12-18 | 2018-07-24 | 上海科大讯飞信息科技有限公司 | Voice data processing method and device, storage medium, electronic equipment |
| CN107919124A (en) * | 2017-12-22 | 2018-04-17 | 北京小米移动软件有限公司 | Equipment awakening method and device |
| CN108509225A (en) * | 2018-03-28 | 2018-09-07 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
| CN108509225B (en) * | 2018-03-28 | 2021-07-16 | 联想(北京)有限公司 | Information processing method and electronic equipment |
| US11194378B2 (en) | 2018-03-28 | 2021-12-07 | Lenovo (Beijing) Co., Ltd. | Information processing method and electronic device |
| CN108766422A (en) * | 2018-04-02 | 2018-11-06 | 青岛海尔科技有限公司 | Response method, device, storage medium and the computer equipment of speech ciphering equipment |
| CN108766422B (en) * | 2018-04-02 | 2020-09-15 | 青岛海尔科技有限公司 | Voice equipment response method and device, storage medium and computer equipment |
| CN108538293A (en) * | 2018-04-27 | 2018-09-14 | 青岛海信电器股份有限公司 | Voice awakening method, device and smart machine |
| CN108777142A (en) * | 2018-06-05 | 2018-11-09 | 上海木木机器人技术有限公司 | A kind of interactive voice recognition methods and interactive voice robot based on airport environment |
| CN108899028A (en) * | 2018-06-08 | 2018-11-27 | 广州视源电子科技股份有限公司 | Voice awakening method, searching method, device and terminal |
| CN110444210A (en) * | 2018-10-25 | 2019-11-12 | 腾讯科技（深圳）有限公司 | A kind of method of speech recognition, the method and device for waking up word detection |
| CN110444210B (en) * | 2018-10-25 | 2022-02-08 | 腾讯科技（深圳）有限公司 | Voice recognition method, awakening word detection method and device |
| CN109410952A (en) * | 2018-10-26 | 2019-03-01 | 北京蓦然认知科技有限公司 | A kind of voice awakening method, apparatus and system |
| CN109410952B (en) * | 2018-10-26 | 2020-02-28 | 北京蓦然认知科技有限公司 | Voice awakening method, device and system |
| CN111235820A (en) * | 2018-11-29 | 2020-06-05 | 青岛海尔洗衣机有限公司 | Control method for laundry treating apparatus |
| CN110046045A (en) * | 2019-04-03 | 2019-07-23 | 百度在线网络技术（北京）有限公司 | The data package processing method and device that voice wakes up |
| CN110111789B (en) * | 2019-05-07 | 2022-02-08 | 阿波罗智联(北京)科技有限公司 | Voice interaction method and device, computing equipment and computer readable medium |
| CN110111789A (en) * | 2019-05-07 | 2019-08-09 | 百度国际科技（深圳）有限公司 | Voice interactive method, calculates equipment and computer-readable medium at device |
| CN110211605A (en) * | 2019-05-24 | 2019-09-06 | 珠海多士科技有限公司 | Smart machine speech sensitivity adjusting method, device, equipment and storage medium |
| TWI717190B (en) * | 2019-05-28 | 2021-01-21 | 開曼群島商創新先進技術有限公司 | Service arousal method and application client |
| US10931814B2 (en) | 2019-05-28 | 2021-02-23 | Advanced New Technologies Co., Ltd. | Service recommendation |
| CN110166985A (en) * | 2019-05-28 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of service awaking method and applications client |
| CN112086096A (en) * | 2019-06-14 | 2020-12-15 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium |
| CN112086096B (en) * | 2019-06-14 | 2024-04-05 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium |
| CN112102821A (en) * | 2019-06-18 | 2020-12-18 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium applied to electronic equipment |
| CN112102821B (en) * | 2019-06-18 | 2024-01-12 | 北京京东尚科信息技术有限公司 | Data processing method, device, system and medium applied to electronic equipment |
| CN110265018A (en) * | 2019-07-01 | 2019-09-20 | 成都启英泰伦科技有限公司 | A kind of iterated command word recognition method continuously issued |
| CN110600029A (en) * | 2019-09-17 | 2019-12-20 | 苏州思必驰信息科技有限公司 | User-defined awakening method and device for intelligent voice equipment |
| CN110706703A (en) * | 2019-10-16 | 2020-01-17 | 珠海格力电器股份有限公司 | Voice wake-up method, device, medium and equipment |
| CN111081251B (en) * | 2019-11-27 | 2022-03-04 | 云知声智能科技股份有限公司 | Voice wake-up method and device |
| CN111081251A (en) * | 2019-11-27 | 2020-04-28 | 云知声智能科技股份有限公司 | Voice wake-up method and device |
| CN111192588A (en) * | 2019-12-31 | 2020-05-22 | 深圳市瑞讯云技术有限公司 | System awakening method and device |
| CN111192588B (en) * | 2019-12-31 | 2022-09-09 | 深圳市瑞讯云技术有限公司 | System awakening method and device |
| CN111309283B (en) * | 2020-03-25 | 2023-12-05 | 北京百度网讯科技有限公司 | Voice control method and device of user interface, electronic equipment and storage medium |
| CN111309283A (en) * | 2020-03-25 | 2020-06-19 | 北京百度网讯科技有限公司 | Voice control method and device for user interface, electronic equipment and storage medium |
| CN111524528A (en) * | 2020-05-28 | 2020-08-11 | Oppo广东移动通信有限公司 | Voice awakening method and device for preventing recording detection |
| CN112908330A (en) * | 2021-03-04 | 2021-06-04 | 深圳市云希谷科技有限公司 | Voice wake-up method and device for terminal equipment and computer readable storage medium |
| CN112908330B (en) * | 2021-03-04 | 2022-08-09 | 深圳市云希谷科技有限公司 | Voice wake-up method and device for terminal equipment and computer readable storage medium |
| CN113113007A (en) * | 2021-03-30 | 2021-07-13 | 北京金山云网络技术有限公司 | Voice data processing method and device, electronic equipment and storage medium |
| CN113590207A (en) * | 2021-07-30 | 2021-11-02 | 思必驰科技股份有限公司 | Method and device for improving awakening effect |
| CN113963695A (en) * | 2021-10-13 | 2022-01-21 | 深圳市欧瑞博科技股份有限公司 | Awakening method, awakening device, equipment and storage medium of intelligent equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106297777B (en) | 2019-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106297777A (en) | A kind of method and apparatus waking up voice service up | |
| CN107818798B (en) | Customer service quality evaluation method, device, equipment and storage medium | |
| CN108538298B (en) | Voice wake-up method and device | |
| US10270736B2 (en) | Account adding method, terminal, server, and computer storage medium | |
| CN109326289A (en) | Exempt to wake up voice interactive method, device, equipment and storage medium | |
| JP2020523643A (en) | Voice identification feature optimization and dynamic registration method, client, and server | |
| CN108766418A (en) | Sound end recognition methods, device and equipment | |
| CN109741754A (en) | A kind of conference voice recognition methods and system, storage medium and terminal | |
| CN108735204A (en) | Equipment for executing task corresponding with user spoken utterances | |
| CN107545897A (en) | Conversation activity presumption method, conversation activity estimating device and program | |
| CN107316635B (en) | Voice recognition method and device, storage medium and electronic equipment | |
| WO2021151310A1 (en) | Voice call noise cancellation method, apparatus, electronic device, and storage medium | |
| CN109086276B (en) | Data translation method, device, terminal and storage medium | |
| CN105210146A (en) | Method and apparatus for controlling voice activation | |
| CN110459222A (en) | Sound control method, phonetic controller and terminal device | |
| WO2020098414A1 (en) | Data processing method for terminal, device, and terminal | |
| CN111083678B (en) | Playing control method and system of Bluetooth sound box and intelligent device | |
| CN109801636A (en) | Training method, device, electronic equipment and the storage medium of Application on Voiceprint Recognition model | |
| CN111508474A (en) | Voice interruption method, electronic equipment and storage device | |
| CN105469789A (en) | Voice information processing method and voice information processing terminal | |
| CN108320738A (en) | Voice data processing method and device, storage medium, electronic equipment | |
| CN110634468B (en) | Voice wake-up method, device, equipment and computer readable storage medium | |
| CN111831795B (en) | Multi-round dialogue processing method and device, electronic equipment and storage medium | |
| CN109242555B (en) | Voice-based advertisement playing method and related product | |
| CN110070857A (en) | The model parameter method of adjustment and device, speech ciphering equipment of voice wake-up model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |