CN108986813A - Wake up update method, device and the electronic equipment of word - Google Patents
Wake up update method, device and the electronic equipment of word Download PDFInfo
- Publication number
- CN108986813A CN108986813A CN201811014878.5A CN201811014878A CN108986813A CN 108986813 A CN108986813 A CN 108986813A CN 201811014878 A CN201811014878 A CN 201811014878A CN 108986813 A CN108986813 A CN 108986813A
- Authority
- CN
- China
- Prior art keywords
- wake
- word
- voice messaging
- sequence
- terminal device
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; 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 OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Abstract
The embodiment of the invention discloses a kind of update method, device and electronic equipments for waking up word.Wherein, method includes: to determine the processing mode for being directed to the first voice messaging when receiving the first voice messaging;If the processing mode for the first voice messaging is that when updating to wake up word, target speaker sequence is determined from the first voice messaging;The original wake-up word for waking up terminal device is updated according to target speaker sequence.The embodiment of the present invention solves wake-up word in the prior art and fixes, user can not customized terminal device the technical issues of waking up word, it ensure that the wake-up lexicon of updated terminal device shares family habit, prevent from not meeting the risk that user's habit causes user to forget that terminal device can not be waken up caused by waking up word or can not accurately saying wake-up word because waking up word.
Description
Technical field
The present embodiments relate to technical field of information processing, more particularly to a kind of update method for waking up word, device
And electronic equipment.
Background technique
With the development of speech recognition technology, the wide range of applications that voice wakes up, such as robot, mobile phone, can
Wearable device, smart home, vehicle-mounted etc..However, in the related technology, voice wakes up terminal device and is primarily present following problem: one
The wake-up word for being terminal device is typically all that can not modify, and when the wake-up word of terminal device does not meet user's habit, is existed
The technical issues of terminal device easily can not be waken up due to user forgets wake-up word or can not accurately say and wake up word.
Summary of the invention
In view of this, can be solved the embodiment of the invention provides a kind of update method, device and electronic equipment for waking up word
Certainly above-mentioned technical problem.
To solve the above-mentioned problems, the embodiment of the present invention mainly provides the following technical solutions:
In a first aspect, the embodiment of the invention provides a kind of update methods for waking up word, this method comprises:
When receiving the first voice messaging, the processing mode for being directed to the first voice messaging is determined;
If the processing mode for the first voice messaging is that when updating to wake up word, target hair is determined from the first voice messaging
Sound sequence;
The original wake-up word for waking up terminal device is updated according to target speaker sequence.
Second aspect, the embodiment of the present invention also provide a kind of updating device for waking up word, which includes:
Processing mode determining module when for receiving the first voice messaging, determines the processing for being directed to the first voice messaging
Mode;
Target sequence determining module, if being when updating to wake up word, from for the processing mode for the first voice messaging
Target speaker sequence is determined in one voice messaging;
Wake-up sequence update module, for updating the original wake-up word for waking up terminal device according to target speaker sequence.
The third aspect, the embodiment of the present invention also provide a kind of electronic equipment, comprising:
At least one processor;
And at least one processor connected to the processor, bus;Wherein,
Processor, memory complete mutual communication by bus;
Processor is used to call the program instruction in memory, to execute the update method for waking up word.
Fourth aspect, the embodiment of the present invention also provide a kind of non-transient computer readable storage medium, non-transient computer
Readable storage medium storing program for executing stores computer instruction, and computer instruction makes computer execute the update method for waking up word.
By above-mentioned technical proposal, technical solution provided in an embodiment of the present invention is at least had the advantage that
In the embodiment of the present invention, when receiving the first voice messaging, the processing mode for being directed to the first voice messaging is determined,
To determine that the processing mode for the first voice messaging is that when updating to wake up word, target speaker is determined from the first voice messaging
Sequence, to update the purpose that terminal device wakes up word to realize, solve as the pronunciation sequence for waking up terminal device is updated
Word is waken up in the prior art to fix, user can not customized terminal device the technical issues of waking up word, ensure that updated end
The wake-up lexicon of end equipment shares family habit, and preventing from not meeting user's habit because waking up word causes user to forget to wake up word or can not
Accurately say the risk that terminal device can not be waken up caused by waking up word.
Above description is only the general introduction of technical solution of the embodiment of the present invention, in order to better understand the embodiment of the present invention
Technological means, and can be implemented in accordance with the contents of the specification, and in order to allow above and other mesh of the embodiment of the present invention
, feature and advantage can be more clearly understood, the special specific embodiment for lifting the embodiment of the present invention below.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
The limitation of embodiment.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows a kind of flow diagram of update method for waking up word provided in an embodiment of the present invention;
Fig. 2 shows the flow diagrams of another update method for waking up word provided in an embodiment of the present invention;
Fig. 3 shows the flow diagram of the update method of another wake-up word provided in an embodiment of the present invention
Fig. 4 shows a kind of structural schematic diagram of updating device for waking up word provided in an embodiment of the present invention;
Fig. 5 shows the structural schematic diagram of another updating device for waking up word provided in an embodiment of the present invention;
Fig. 6 shows the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, the example of embodiment is shown in the accompanying drawings, wherein identical from beginning to end
Or similar label indicates same or similar element or element with the same or similar functions.It is retouched below with reference to attached drawing
The embodiment stated is exemplary, and for explaining only the invention, and is not construed as limiting the claims.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, " one
It is a ", " described " and "the" may also comprise plural form.It is to be further understood that being arranged used in specification of the invention
Diction " comprising " refer to that there are the feature, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition
Other one or more features, integer, step, operation, element, component and/or their group.It should be understood that when we claim member
Part is " connected " or when " coupled " to another element, it can be directly connected or coupled to other elements, or there may also be
Intermediary element.In addition, " connection " used herein or " coupling " may include being wirelessly connected or wirelessly coupling.It is used herein to arrange
Diction "and/or" includes one or more associated wholes for listing item or any cell and all combinations.
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
In the related technology, terminal device there is technical issues that wake up word can not, thus user forget wake up word
Or there is the terminal device that can not be waken up when can not accurately say the wake-up word of terminal device in user.For example, terminal
When the wake-up word of equipment is " helloKatrina ", if user does not remember, or when will not express the wake-up word, terminal device language
The function that sound wakes up just cannot achieve, and influence user's use.
Update method, device and the electronic equipment provided in an embodiment of the present invention for waking up word, it is intended to solve the prior art
Technical problem as above.
Embodiment one
As shown in Figure 1, including the following steps: the present invention provides a kind of update method for waking up word
Step S101, when receiving the first voice messaging, the processing mode for being directed to the first voice messaging is determined.
For the embodiment of the present invention, after terminal device booting, the first voice messaging is received by terminal device, so as to basis
First voice messaging realizes that modification terminal device wakes up the purpose of word.
Specifically, the first voice messaging can be what terminal device received, be also possible to connect voice with terminal device
Acquisition equipment detects.When the first voice messaging is detected by the voice capture device connecting with terminal device, voice is adopted
The first voice messaging that collection equipment will test is sent to terminal device.When practical application, voice capture device can be Mike
The devices such as wind, sound pick-up.
Specifically, terminal device can be mobile phone, pad, notebook, wearable device and such as smartwatch, intelligent sound
The smart machines such as case.
If being step S102, when updating to wake up word, from the first voice messaging for the processing mode of the first voice messaging
Determine target speaker sequence;
For the embodiment of the present invention, determine that the processing mode for the first voice messaging is to update the purpose for waking up word to be
Update original wake-up word of terminal device.
For the embodiment of the present invention, determine that the purpose of target speaker sequence is target speaker from the first voice messaging
Wake-up word of the sequence as new wake-up terminal device is realized with the new wake-up word for the sequence characterization wake-up terminal device that pronounces
Purpose.
For example, processing mode may wake up word, next track, last track, increase volume, reduction volume etc. to update, true
It surely is that when updating to wake up word, target speaker sequence is determined from the first voice messaging for the processing mode of the first voice messaging.
Step S103, the original wake-up word for waking up terminal device is updated according to target speaker sequence.
For the embodiment of the present invention, when receiving the first voice messaging, the processing side for being directed to the first voice messaging is determined
Formula, to determine that the processing mode for the first voice messaging is that when updating to wake up word, target is determined from the first voice messaging
Sequence of pronouncing updates the purpose that terminal device wakes up word to realize to wake up the pronunciation sequence of terminal device as update,
Solve in the prior art wake up word fix, user can not customized terminal device wake-up word the technical issues of, ensure that more
The wake-up lexicon of terminal device after new shares family habit, prevents from causing user to forget to wake up because wake-up word does not meet user's habit
Word or can not accurately say wake up word caused by the risk that can not be waken up of terminal device.
In some embodiments, the original wake-up word for waking up terminal device is updated according to target speaker sequence, comprising: step
S1031 (not shown) or step S1032 (not shown).
Step S1031, the corresponding text information of original wake-up word for waking up terminal device is updated according to target speaker sequence.
For the embodiment of the present invention, target speaker sequence is updated to wake up original wake-up word of terminal device, is realized
By the modification of original wake-up word of terminal device corresponding text information for sequence of pronouncing, so as in later period speech recognition process,
The voice messaging that will test is converted directly into corresponding pronunciation sequence, without accurately confirmly detecting voice messaging
Corresponding wake-up word text not only simplifies the calculation amount for waking up speech recognition in terminal device, also accelerates voice and wakes up terminal
The speed of equipment improves wake-up efficiency, solves the technical issues of being accustomed to the wake-up word of setting terminal equipment according to user.
For example, the former word that wakes up of terminal device is text information " you are well happy ", method provided in an embodiment of the present invention can
The wake-up word of terminal device is revised as " your good small pleasure " by " you ask well " to realize, i.e., original is waken up into word by text information
" you are well happy " is revised as " your good small pleasure " corresponding pronunciation sequence " nihaoxiaole ", to receive when waking up in the later period
To voice be converted to corresponding pronunciation sequence, without determine the corresponding wake-up word of voice messaging that receives whether with
" your good small pleasure " is completely the same, thus reduces the expense for the calculation amount that voice wakes up during terminal device.Meanwhile it receiving
To the corresponding sequence of voice be determined as " nihaoxiaole " when, terminal device can be waken up, both improved calling out for terminal device
The rate of waking up, has also speeded up the wake-up speed of terminal device.
Step S1032, the corresponding pronunciation sequence of original wake-up word for waking up terminal device is updated according to target speaker sequence.
For the embodiment of the present invention, when original wake-up word stored in terminal device is pronunciation sequence, by by mesh
Mark pronunciation sequence updates the corresponding pronunciation sequence of original wake-up word of terminal device, realizes in later period speech recognition, is sent out with target
Purpose of the sound sequence as the pronunciation sequence for waking up terminal device.
For example, the corresponding pronunciation sequence of original wake-up word of terminal device is " nihaolele ", target speaker sequence is
" nihaoxiaole " wakes up terminal after the corresponding pronunciation sequence of original wake-up word of target speaker sequence update terminal device
The corresponding pronunciation sequence modification of the wake-up word of equipment is for " nihaoxiaole ", that is to say, that detects that user says that " hello
When the pinyin pronunciation that little Bai ", " your good Xiao Le " etc. can be indicated with " nihaoxiaole ", terminal device can be waken up, in turn
Improve the wake-up rate of terminal device.
Further, for the embodiment of the present invention, the processing mode for being directed to the first voice messaging, packet are determined in step S101
Include S1011 (not shown) or S1012 (not shown).
Wherein, the processing mode for being directed to the first voice messaging step S1011, is determined based on the user instruction received.
For the embodiment of the present invention, determine that the processing mode for the first voice messaging is the user instruction by receiving
Determining.When practical application, user instruction, which can be, to be received by the user interface of terminal device.
For example, the corresponding option of different user instructions can be set in the user interface of terminal device, such as " updates and wake up
Word " option, " next track " option etc., terminal device receives user after the option that user interface selects, and is selected according to user
Option determine be directed to the first voice messaging processing mode.
Step S1012, speech recognition is carried out to the first voice messaging, to determine the processing side for being directed to the first voice messaging
Formula.
It may include the processing mode for being directed to the first voice messaging in the first voice messaging for the embodiment of the present invention,
By the speech recognition to the first voice messaging, to determine the processing for the first voice messaging for including in the first voice messaging
Mode.
The processing mode for the first voice messaging can be only included when practical application, in the first voice messaging, it can also be with
It simultaneously include the processing mode for the first voice messaging and the wake-up word for updating terminal device.Specifically, for the first voice
The processing mode of information and the wake-up word for updating terminal device can be disposable input, can also input in two times.For example, defeated
Enter " update and wake up word, update are as follows: your good lulu " or input " update and wake up word " and " your good lulu ", wherein " you reveal well
Dew " can be after determining processing mode according to " update wake up word ", receives and inputs after the prompt information inputted again.
It in some implementations, can simultaneously include step S1011 and step S1012, realization can both pass through step
S1011 determines the processing mode for being directed to the first voice messaging, can also be determined by step S1012 for the first voice messaging
Processing mode.When concrete application, it can be determined in the processing mode for the first voice messaging at both and select a kind of conduct
Default behavior.
Further, it is being determined in the embodiment for the processing mode of the first voice messaging with step S1012, from the first language
Target speaker sequence is determined in message breath, comprising: carries out the recognition result that speech recognition obtains according to the first voice messaging, really
Set the goal sequence.
For the embodiment of the present invention, the wake-up word of processing mode and update terminal device for the first voice messaging, i.e.,
First voice messaging is that the embodiment of the present invention can disposably be determined by speech recognition is directed to the first voice when disposably inputting
Target speaker sequence is determined in the processing mode of information and the first voice messaging.For example, the first voice messaging is " to update and wake up
Word updates are as follows: your good lulu " when, determine that the processing mode for the first voice messaging is " update and wake up word ", target speaker sequence
It is classified as " nihaolulu ".
For the embodiment of the present invention, the processing mode for the first voice messaging and update the wake-up word of terminal device can
To be when inputting by several times, to determine the corresponding content of voice of each input respectively by speech recognition.For example, the first voice messaging
It inputs in two times, input for the first time is " update and wake up word ", is inputted and is determined for the first voice messaging according to first time
Processing mode is that user can be prompted to carry out second and inputted after updating wake-up word, if inputting for second is " your good lulu ", really
The sequence that sets the goal is " nihaolulu ", is updated to " nihaolulu " to wake up the pronunciation sequence of terminal device to realize
The purpose of (i.e. wake-up word), so that the wake-up word of the user's input in later period speech recognition, detected is " nihaolulu "
When, fast wake-up terminal device.
Embodiment two
The embodiment of the invention provides alternatively possible implementations, as shown in Fig. 2, the method comprising the steps of S201,
Step S202, step S203, step S204, step S205 and step S206, wherein step S201, step S202 and step
The implementation procedure of S203 in embodiment one step S101, step S102 and step S103 it is same or similar, it is no longer superfluous herein
It states.
Step S204, after updating the original wake-up word for waking up terminal device according to target speaker sequence, after update
Target speaker sequence construct wake up terminal device wake-up identification model;
Step S205, after the second voice messaging for detecting user's input, the second voice messaging of user's input is inputted
To wake-up identification model;
If step S206, being identified by, the wake operation to terminal device is triggered.
It is corresponding in the original wake-up word for updating wake-up terminal device according to target speaker sequence for the embodiment of the present invention
In the implementation for sequence of pronouncing, after the pronunciation sequence that target speaker sequence is updated to wake up terminal device, according to target
The wake-up identification model of pronunciation sequence construct can targetedly identify the updated pronunciation sequence for waking up terminal device, if the
The pronunciation sequence for including in two voice messagings is consistent with the pronunciation sequence waken up in identification model, and determination is identified by, triggering pair
The wake operation of terminal device, is not otherwise triggered.In addition, the embodiment of the present invention is by the wake-up identification model of building because being to be directed to
Property the model that can only identify target speaker sequence, therefore reduce speech recognition excessively in wake up identification model memory it is big
It is small, the purpose quickly identified can be played, the wake-up rate of terminal device is improved, and accelerates the wake-up speed of terminal device.
For example, the corresponding sequence of original wake-up word of terminal device is " nihaolele ", target speaker sequence is
" nihaoxiaobai " is waken up eventually after the corresponding pronunciation sequence of original wake-up word of target speaker sequence update terminal device
The corresponding pronunciation sequence modification of original wake-up word of end equipment is for " nihaoxiaobai ", that is to say, that detects that user says
When " your good little Bai ", " your good Xiao one hundred " etc. can be with the wake-up words of the pinyin pronunciation of " nihaoxiaobai " characterization out, it is ok
Wake up terminal device.
Target speaker sequence is being updated to wake up the corresponding pronunciation of original wake-up word of terminal device by the embodiment of the present invention
After sequence, by constructing the wake-up identification model of target speaker sequence, so that in speech recognition, using waking up, identification model is fast
Speed identification, the wake-up speed of increasing terminal equipment improve the wake-up rate of terminal device.
It is corresponding in the original wake-up word for updating wake-up terminal device according to target speaker sequence for the embodiment of the present invention
In the implementation of text information, target speaker sequence is updated to wake up the pronunciation sequence of terminal device, is realized terminal
The corresponding text information of original wake-up word of equipment is revised as the purpose of pronunciation sequence, and in the wake-up of building target speaker sequence
After identification model, the later period is carried out in speech recognition process, is waken up the voice messaging that will test of identification model and is converted directly into pair
The pronunciation sequence answered not only simplifies wake-up terminal and sets without accurately confirmly detecting the corresponding text of voice messaging
The calculation amount of standby middle speech recognition also accelerates the speed that voice wakes up terminal device, improves wake-up efficiency, solve basis
User is accustomed to the technical issues of wake-up word of setting terminal equipment.
For example, original wake-up word of terminal device is text information " you are well happy ", method provided in an embodiment of the present invention
It may be implemented the wake-up word of terminal device being revised as " your good small pleasure " by " you are well happy ", " your good small pleasure " corresponding pronunciation sequence
It is classified as " nihaoxiaole ", and after the wake-up identification model of building target speaker sequence " nihaoxiaole ", the later period is carried out
In speech recognition process, the voice messaging that waking up identification model will test is converted directly into corresponding pronunciation sequence, and
Without accurately confirmly detecting the corresponding text of voice messaging, that is to say, that without confirm that user says be that " you are good small
It is happy ", the word that still " your good Xiao Le " etc. can be indicated with " nihaoxiaole ", therefore the embodiment of the present invention not only simplifies wake-up
The calculation amount of speech recognition in terminal device also accelerates the speed that voice wakes up terminal device, improves wake-up efficiency, solves
The technical issues of being accustomed to the wake-up word of setting terminal equipment according to user.
Embodiment three
The embodiment of the invention provides alternatively possible implementations, further include implementing on the basis of example 1
Method shown in example three, as shown in figure 3, the method comprising the steps of S301, step S302, step S303, step S304, step
S305 and step S306.
Wherein, step S301, step S101, step S102 and the step in step S302 and step S303 and embodiment one
The implementation procedure of S103 in embodiment one step S101, step S102 and step S103 it is same or similar, it is no longer superfluous herein
It states.
After updating the original wake-up word for waking up terminal device according to target speaker sequence, user step S304, is detected
The third voice messaging of input identifies third voice messaging using preset general pronunciation recognition sequence model, obtains
The corresponding pronunciation sequence of third voice messaging;
Step S305, judge the corresponding pronunciation sequence of third voice messaging and the pronunciation sequence of updated terminal device is
It is no consistent;
If step S306, consistent, wake-up processing is carried out to terminal device.
For the embodiment of the present invention, third voice messaging is known by preset general hair voice sequence identification model
Not, after obtaining the corresponding pronunciation sequence of third voice messaging, whether judgement is consistent with the pronunciation sequence of updated terminal device,
If consistent, wake-up processing is carried out to terminal device.
For the embodiment of the present invention, target speaker sequence is updated to wake up the pronunciation sequence of terminal device, realize by
The corresponding text information of original wake-up word of terminal device is revised as the purpose of pronunciation sequence, in later period speech recognition process,
The third voice messaging detected is identified using general hair voice sequence identification model, it is corresponding to obtain third voice messaging
Pronunciation sequence after, judge the pronunciation sequence for including in third voice messaging and it is updated wake up terminal device pronunciation sequence
It is whether consistent, if unanimously, triggering the wake operation to terminal device, otherwise not triggering.
It is all using general hair voice sequence identification model to inspection in speech recognition process for the embodiment of the present invention
The third voice messaging measured is identified to obtain corresponding pronunciation sequence, without accurately confirmly detecting voice messaging pair
The wake-up word text answered not only simplifies the calculation amount for waking up speech recognition in terminal device, also accelerates voice wake-up terminal and sets
Standby speed improves wake-up efficiency, solves the technical issues of being accustomed to the wake-up word of setting terminal equipment according to user.
For example, it is assumed that the content of third voice messaging is " I input be your well happy pleasure ", through the knowledge of general hair voice sequence
After other model identification, the corresponding pronunciation sequence of third voice messaging is " woshurudeshinihaolele ", and updated end
The pronunciation sequence of end equipment be " nihaoxiaole ", because in " woshurudeshinihaolele " and be not present with
" nihaoxiaole " consistent pronunciation sequence, therefore wake-up processing is not carried out to terminal device.If the content of third voice messaging
For " I input be your good small pleasure ", the corresponding pronunciation sequence of third voice messaging is " woshurudeshinihaoxiaole ",
Because including " nihaoxiaole " in " woshurudeshinihaoxiaole ", wake-up processing is carried out to terminal device.Or the
The content of three voice messagings is " your good small pleasure ", and the corresponding pronunciation sequence of third voice messaging is " nihaoxiaole ", passes through ratio
Compared with the two is consistent, carries out wake-up processing to terminal device.
It is to carry out waking up processing using wake-up identification model, and the present embodiment is then using logical in the embodiment of the present invention two
With pronunciation recognition sequence model will receive speech signal analysis be corresponding pronunciation sequence after, then with updated terminal device
Pronunciation sequence be compared, determine whether wake up terminal device purpose.It should be noted that when practical application, this
Invention can be arranged simultaneously referring concurrently to embodiment two and embodiment three and wake up identification model and universal identification model, realize
Wake-up processing is carried out according to the voice messaging that terminal device receives.
Example IV
The structural schematic diagram of a kind of updating device waking up word provided in an embodiment of the present invention, as shown in figure 4, the present invention is real
The updating device 40 for applying the wake-up word of example may include: processing mode determining module 401, target sequence determining module 402 and call out
Awake sequence update module 403.
Wherein, processing mode determining module 401 when for receiving the first voice messaging, determining and believing for the first voice
The processing mode of breath;
Target sequence determining module 402, if be that update wakes up word for the processing mode for the first voice messaging, from
Target speaker sequence is determined in first voice messaging;
Wake-up sequence update module 403, for updating the original wake-up word for waking up terminal device according to target speaker sequence.
For the embodiment of the present invention, compare the prior art, when receiving the first voice messaging, determines and is directed to the first language
The processing mode of message breath, to determine that the processing mode for the first voice messaging is when updating to wake up word, from the first voice
Target speaker sequence is determined in information, to pass through processing mode to realize as the pronunciation sequence for waking up terminal device is updated
Setting realize update terminal device wake up word purpose, solution can not adjust terminal device wake up word the technical issues of, with protect
The wake-up lexicon for demonstrate,proving updated terminal device shares family habit, prevents from causing user to forget because wake-up word does not meet user's habit
It wakes up word or can not accurately say and wake up word and the risk that can not wake up of the terminal device that generates.
Further, wake-up sequence update module 403 wakes up the original of terminal device for updating according to target speaker sequence
Wake up the corresponding text information of word;Or the corresponding pronunciation of original wake-up word for waking up terminal device is updated according to target speaker sequence
Sequence.
The embodiment of the present invention may be used also by can both update the corresponding text information of original wake-up word with target speaker sequence
To update the corresponding pronunciation sequence of original wake-up word with target speaker sequence, the mesh for updating original wake-up word of terminal device is realized
, so that in later period speech recognition process, the voice that will test is converted to corresponding pronunciation sequence, without accurate
The corresponding wake-up word of voice messaging is confirmly detected, the calculation amount for waking up speech recognition in terminal device is not only simplified, also accelerates
Voice wakes up the speed of terminal device, improves wake-up efficiency, solves the wake-up for being accustomed to setting terminal equipment according to user
The technical issues of word.
Further, processing mode determining module includes: 4011 (not shown) of the first determination unit, wherein
First determination unit 4011 is believed with determining for the first voice for carrying out speech recognition to the first voice messaging
The processing mode of breath.
Further, processing mode determining module includes: 4012 (not shown) of the second determination unit, wherein
Second determination unit 4012 is believed with determining for the first voice for carrying out speech recognition to the first voice messaging
The processing mode of breath.
In one implementation, processing mode determining module can only include the first determination unit, can also only include
Second determination unit, in another implementation, processing mode determining module can include the first determination unit and the simultaneously
Two determination units.
In the implementation that processing mode determining module includes the second determination unit, target sequence determining module 4012,
It is also used to carry out the recognition result that speech recognition obtains to the first voice messaging according in the second determining module, determines target speaker
Sequence.
Further, as shown in figure 5, the device 40 further include: model construction module 404, wherein
Model construction module 504, for according to target speaker sequence update wake up terminal device pronunciation sequence after,
The wake-up identification model of terminal device is waken up according to updated target speaker sequence construct.
Further, as shown in figure 5, the device 40 further include: wake up processing module 405, wherein
Processing module 405 is waken up, in the wake-up for waking up terminal device according to updated target speaker sequence construct
After identification model, after the second voice messaging for detecting user's input, the second voice messaging that user inputs is input to wake-up
Identification model;If being identified by, the wake operation to terminal device is triggered.
In some implementations, processing module 405 is waken up, is also used to updating wake-up terminal according to target speaker sequence
After the pronunciation sequence of equipment, detects the third voice messaging of user's input, utilize preset general pronunciation recognition sequence mould
Type identifies third voice messaging, obtains the corresponding pronunciation sequence of third voice messaging;Judge that third voice messaging is corresponding
Pronunciation sequence it is whether consistent with the pronunciation sequence of updated terminal device;If consistent, wake-up processing is carried out to terminal device.
Any reality into embodiment three of the embodiment of the present invention one can be performed in the updating device of the wake-up word of the embodiment of the present invention
The update method that word is waken up shown in example is applied, realization principle is similar, and details are not described herein again.
Embodiment five
The embodiment of the invention provides a kind of electronic equipment, as shown in fig. 6, electronic equipment shown in fig. 6 500 includes: place
Manage device 5001 and memory 5003.Wherein, processor 5001 is connected with memory 5003, is such as connected by bus 5002.Into one
Step ground, electronic equipment 500 can also include transceiver 5004.It should be noted that transceiver 5004 is not limited in practical application
One, the structure of the electronic equipment 500 does not constitute the restriction to the embodiment of the present application.
Processor 5001 can be CPU, general processor, DSP, ASIC, FPGA or other programmable logic device, crystalline substance
Body pipe logical device, hardware component or any combination thereof.It, which may be implemented or executes, combines described by the disclosure of invention
Various illustrative logic blocks, module and circuit.Processor 5001 is also possible to realize the combination of computing function, such as wraps
It is combined containing one or more microprocessors, DSP and the combination of microprocessor etc..
Bus 5002 may include an access, and information is transmitted between said modules.Bus 5002 can be pci bus or
Eisa bus etc..Bus 5002 can be divided into address bus, data/address bus, control bus etc..Only to be used in Fig. 6 convenient for indicating
One thick line indicates, it is not intended that an only bus or a type of bus.
Memory 5003 can be ROM or can store the other kinds of static storage device of static information and instruction, RAM
Or the other kinds of dynamic memory of information and instruction can be stored, it is also possible to EEPROM, CD-ROM or other CDs
Storage, optical disc storage (including compression optical disc, laser disc, optical disc, Digital Versatile Disc, Blu-ray Disc etc.), magnetic disk storage medium
Or other magnetic storage apparatus or can be used in carry or store have instruction or data structure form desired program generation
Code and can by any other medium of computer access, but not limited to this.
Memory 5003 is used to store the application code for executing the present invention program, and is held by processor 5001 to control
Row.Processor 5001 is for executing the application code stored in memory 5003, to realize Fig. 4 and embodiment illustrated in fig. 5
The updating device of the wake-up word of offer.
For the embodiment of the present invention, compare the prior art, when receiving the first voice messaging, determines and is directed to the first language
The processing mode of message breath, to determine that the processing mode for the first voice messaging is when updating to wake up word, from the first voice
Target speaker sequence is determined in information, to pass through processing mode to realize as the pronunciation sequence for waking up terminal device is updated
Setting realize update terminal device wake up word purpose, solution can not adjust terminal device wake up word the technical issues of, with protect
The wake-up lexicon for demonstrate,proving updated terminal device shares family habit, prevents from causing user to forget because wake-up word does not meet user's habit
It wakes up word or can not accurately say and wake up word and the risk that can not wake up of the terminal device that generates.
Embodiment six
The embodiment of the present invention provides a kind of non-transient computer readable storage medium, non-transient computer readable storage medium
Computer instruction is stored, computer instruction makes computer execute any embodiment in above-mentioned each method embodiment one and embodiment three
Shown in wake up processing method.
For the embodiment of the present invention, compare the prior art, when receiving the first voice messaging, determines and is directed to the first language
The processing mode of message breath, to determine that the processing mode for the first voice messaging is when updating to wake up word, from the first voice
Target speaker sequence is determined in information, to pass through processing mode to realize as the pronunciation sequence for waking up terminal device is updated
Setting realize update terminal device wake up word purpose, solution can not adjust terminal device wake up word the technical issues of, with protect
The wake-up lexicon for demonstrate,proving updated terminal device shares family habit, prevents from causing user to forget because wake-up word does not meet user's habit
It wakes up word or can not accurately say and wake up word and the risk that can not wake up of the terminal device that generates.
It should be understood that although each step in the flow chart of attached drawing is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, can execute in the other order.Moreover, at least one in the flow chart of attached drawing
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, execution sequence, which is also not necessarily, successively to be carried out, but can be with other
At least part of the sub-step or stage of step or other steps executes in turn or alternately.
The above is only some embodiments of the invention, it is noted that those skilled in the art are come
It says, various improvements and modifications may be made without departing from the principle of the present invention, these improvements and modifications also should be regarded as
Protection scope of the present invention.
Claims (10)
1. a kind of update method for waking up word characterized by comprising
When receiving the first voice messaging, the processing mode for being directed to first voice messaging is determined;
If the processing mode for first voice messaging is that when updating to wake up word, mesh is determined from first voice messaging
Mark pronunciation sequence;
The original wake-up word for waking up terminal device is updated according to the target speaker sequence.
2. the method according to claim 1, wherein the determination is directed to the processing side of first voice messaging
Formula, comprising:
Based on the user instruction received, the processing mode for being directed to first voice messaging is determined.
3. the method according to claim 1, wherein the determination is directed to the processing side of first voice messaging
Formula, comprising:
Speech recognition is carried out to first voice messaging, to determine the processing mode for being directed to first voice messaging.
4. according to the method described in claim 3, it is characterized in that, described determine target speaker from first voice messaging
Sequence, comprising:
The recognition result that speech recognition obtains is carried out according to first voice messaging, determines target speaker sequence.
5. the method according to claim 1, wherein described update according to the target speaker sequence wakes up terminal
After the pronunciation sequence of equipment, the method also includes:
The wake-up identification model of the terminal device is waken up according to updated target speaker sequence construct;
After waking up the wake-up identification model of the terminal device according to updated target speaker sequence construct, the method is also
Include:
After the second voice messaging for detecting user's input, the second voice messaging that user inputs is input to the wake-up and is identified
Model;
If being identified by, the wake operation to the terminal device is triggered.
6. the method according to claim 1, wherein described update according to the target speaker sequence wakes up terminal
After the pronunciation sequence of equipment, the method also includes:
The third voice messaging for detecting user's input, using preset general pronunciation recognition sequence model to the third voice
Information is identified, the corresponding pronunciation sequence of the third voice messaging is obtained;
Judge the third voice messaging it is corresponding pronunciation sequence and the updated terminal device pronunciation sequence whether one
It causes;
If consistent, wake-up processing is carried out to the terminal device.
7. the method according to claim 1, wherein described update according to the target speaker sequence wakes up terminal
Original wake-up word of equipment, comprising:
The corresponding text information of original wake-up word for waking up terminal device is updated according to the target speaker sequence;Or
The corresponding pronunciation sequence of original wake-up word for waking up terminal device is updated according to the target speaker sequence.
8. a kind of updating device for waking up word characterized by comprising
Processing mode determining module when for receiving the first voice messaging, determines the processing for being directed to first voice messaging
Mode;
Target sequence determining module, if being when updating to wake up word, from institute for the processing mode for first voice messaging
It states and determines target speaker sequence in the first voice messaging;
Wake-up sequence update module, for updating the original wake-up word for waking up terminal device according to the target speaker sequence.
9. a kind of electronic equipment characterized by comprising
At least one processor;
And at least one processor, the bus being connected to the processor;Wherein,
The processor, memory complete mutual communication by the bus;
The processor is used to call the program instruction in the memory, any into claim 7 with perform claim requirement 1
The update method of wake-up word described in.
10. a kind of non-transient computer readable storage medium, which is characterized in that the non-transient computer readable storage medium is deposited
Store up computer instruction, the computer instruction requires the computer perform claim 1 to described in any one of claim 7
Wake up the update method of word.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811014878.5A CN108986813A (en) | 2018-08-31 | 2018-08-31 | Wake up update method, device and the electronic equipment of word |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811014878.5A CN108986813A (en) | 2018-08-31 | 2018-08-31 | Wake up update method, device and the electronic equipment of word |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108986813A true CN108986813A (en) | 2018-12-11 |
Family
ID=64548625
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811014878.5A Pending CN108986813A (en) | 2018-08-31 | 2018-08-31 | Wake up update method, device and the electronic equipment of word |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108986813A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109767763A (en) * | 2018-12-25 | 2019-05-17 | 苏州思必驰信息科技有限公司 | It is customized wake up word determination method and for determine it is customized wake up word device |
CN110890093A (en) * | 2019-11-22 | 2020-03-17 | 腾讯科技(深圳)有限公司 | Intelligent device awakening method and device based on artificial intelligence |
CN112185382A (en) * | 2020-09-30 | 2021-01-05 | 北京猎户星空科技有限公司 | Method, device, equipment and medium for generating and updating wake-up model |
CN114071200A (en) * | 2022-01-17 | 2022-02-18 | 北京智象信息技术有限公司 | Method and system for dynamically updating TV pickup peripheral awakening words |
WO2023109129A1 (en) * | 2021-12-13 | 2023-06-22 | 海信视像科技股份有限公司 | Speech data processing method and apparatus |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160135047A1 (en) * | 2014-11-12 | 2016-05-12 | Samsung Electronics Co., Ltd. | User terminal and method for unlocking same |
CN107134279A (en) * | 2017-06-30 | 2017-09-05 | 百度在线网络技术(北京)有限公司 | A kind of voice awakening method, device, terminal and storage medium |
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 |
CN206977772U (en) * | 2017-07-30 | 2018-02-06 | 成都启英泰伦科技有限公司 | A kind of novel intelligent light fixture |
US20180108343A1 (en) * | 2016-10-14 | 2018-04-19 | Soundhound, Inc. | Virtual assistant configured by selection of wake-up phrase |
-
2018
- 2018-08-31 CN CN201811014878.5A patent/CN108986813A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160135047A1 (en) * | 2014-11-12 | 2016-05-12 | Samsung Electronics Co., Ltd. | User terminal and method for unlocking same |
US20180108343A1 (en) * | 2016-10-14 | 2018-04-19 | Soundhound, Inc. | Virtual assistant configured by selection of wake-up phrase |
CN107134279A (en) * | 2017-06-30 | 2017-09-05 | 百度在线网络技术(北京)有限公司 | A kind of voice awakening method, device, terminal and storage medium |
CN206977772U (en) * | 2017-07-30 | 2018-02-06 | 成都启英泰伦科技有限公司 | A kind of novel intelligent light fixture |
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 |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109767763A (en) * | 2018-12-25 | 2019-05-17 | 苏州思必驰信息科技有限公司 | It is customized wake up word determination method and for determine it is customized wake up word device |
CN110890093A (en) * | 2019-11-22 | 2020-03-17 | 腾讯科技(深圳)有限公司 | Intelligent device awakening method and device based on artificial intelligence |
CN110890093B (en) * | 2019-11-22 | 2024-02-09 | 腾讯科技(深圳)有限公司 | Intelligent equipment awakening method and device based on artificial intelligence |
CN112185382A (en) * | 2020-09-30 | 2021-01-05 | 北京猎户星空科技有限公司 | Method, device, equipment and medium for generating and updating wake-up model |
CN112185382B (en) * | 2020-09-30 | 2024-03-08 | 北京猎户星空科技有限公司 | Method, device, equipment and medium for generating and updating wake-up model |
WO2023109129A1 (en) * | 2021-12-13 | 2023-06-22 | 海信视像科技股份有限公司 | Speech data processing method and apparatus |
CN114071200A (en) * | 2022-01-17 | 2022-02-18 | 北京智象信息技术有限公司 | Method and system for dynamically updating TV pickup peripheral awakening words |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108986813A (en) | Wake up update method, device and the electronic equipment of word | |
CN107256707B (en) | Voice recognition method, system and terminal equipment | |
JP5996783B2 (en) | Method and terminal for updating voiceprint feature model | |
CN105493180B (en) | Electronic device and method for speech recognition | |
CN111261144B (en) | Voice recognition method, device, terminal and storage medium | |
EP3611724A1 (en) | Voice response method and device, and smart device | |
JP2019117623A (en) | Voice dialogue method, apparatus, device and storage medium | |
CN101795323B (en) | Electronic alarm operation method, electronic alarm and mobile communication terminal | |
US11200899B2 (en) | Voice processing method, apparatus and device | |
CN104407834A (en) | Message input method and device | |
CN106297801A (en) | Method of speech processing and device | |
US11587560B2 (en) | Voice interaction method, device, apparatus and server | |
CN103077721A (en) | Voice memorandum method of mobile terminal and mobile terminal | |
CN109688269A (en) | The filter method and device of phonetic order | |
CN110225386A (en) | A kind of display control method, display equipment | |
US20200265843A1 (en) | Speech broadcast method, device and terminal | |
CN109272995A (en) | Audio recognition method, device and electronic equipment | |
KR20130068303A (en) | Apparatus for performing voice command, mobile terminal having the same and method for performing voice command | |
CN113096647B (en) | Voice model training method and device and electronic equipment | |
KR20210032875A (en) | Voice information processing method, apparatus, program and storage medium | |
CN105161095A (en) | Method and device for picture composition of speech recognition syntax tree | |
US20180090126A1 (en) | Vocal output of textual communications in senders voice | |
US20190304455A1 (en) | Electronic device for processing user voice | |
KR20150104930A (en) | Method and system of supporting multitasking of speech recognition service in in communication device | |
KR100612477B1 (en) | A method of encoding text data to include enhanced speech data for use in a text to speech tts system, a method of decoding, a tts system and a mobile phone including said tts system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |