CN109767763A - It is customized wake up word determination method and for determine it is customized wake up word device - Google Patents
It is customized wake up word determination method and for determine it is customized wake up word device Download PDFInfo
- Publication number
- CN109767763A CN109767763A CN201811593641.7A CN201811593641A CN109767763A CN 109767763 A CN109767763 A CN 109767763A CN 201811593641 A CN201811593641 A CN 201811593641A CN 109767763 A CN109767763 A CN 109767763A
- Authority
- CN
- China
- Prior art keywords
- word
- customized
- wake
- chinese character
- user
- 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
Landscapes
- Machine Translation (AREA)
Abstract
The present invention discloses a kind of customized determination method for waking up word, comprising: receives the first user instruction;Custom content is determined according to the first user instruction;Custom content is carried out to wake up word assessment;Customized wake-up word is determined according to assessment result.The invention also discloses a kind of for determining the customized device for waking up word, the method and apparatus provided according to the present invention may be implemented to the customized of wake-up word, and customized wake-up word more accurate, that wake-up rate is high, false wake-up rate is low can be obtained, the process for generating wake-up word is more efficiently quick.Simultaneously as the quality of its wake-up word that can be good at guaranteeing customized part, so that improving the playability of speech production, the user experience of entire product is greatly improved.
Description
Technical field
The present invention relates to technical field of voice interaction, more particularly to a kind of customized determination method for waking up word and for true
The fixed customized device for waking up word.
Background technique
With the daily increasing and monthly benefiting of interactive voice technology, the voice of mainstream is waken up there are mainly two types of models at present, and one is bases
In language model, another kind is that the wake-up model based on no language model, based on language model includes acoustic model and language mould
Type needs the checking treatment by two models, although the accuracy rate of verification is higher, obtained wake-up word utilization rate and availability
It is higher, but great calculation amount is needed, so the treatment process of the model is slow, low efficiency.For based on no language model
It waking up model and only carries out acoustics verification, calculation amount is small, and processing speed is fast, but in user's wake-up word customized using the model
With regard to more troublesome, accuracy rate and availability can be reduced.
Summary of the invention
To solve the above-mentioned problems, it realizes and wakes up the customized setting of word, and guarantee the customized availability for waking up word more
High, more acurrate, inventor contemplates the method that assessment voice wakes up word in the standard of the prior art, for determining calling out for content
Awake word further assesses marking, and to waking up, word carries out sensitive word detection, the folded word of repetition detects, spoken word detection and sounding be not full
Word detection, and set up wake-up word threshold value for it and ensure that user using the customized wake-up rate liter for waking up word in this way
Height, false wake-up rate reduce.And due to the quality of its wake-up word that can be good at guaranteeing customized part, so that being promoted
The playability of speech production, the user experience of entire product are greatly improved.
In a first aspect, the embodiment of the invention provides a kind of customized determination methods for waking up word, comprising:
Receive the first user instruction;Custom content is determined according to the first user instruction;Custom content is waken up
Word assessment;Customized wake-up word is determined according to assessment result.
Second aspect, the embodiment of the invention provides a kind of for determining the customized device for waking up word, comprising: first connects
Module is received, for receiving the first user instruction;Custom content obtains module, customized for being determined according to the first user instruction
Content;Word evaluation module is waken up, wakes up word assessment for carrying out to custom content;Threshold wake-up value generation module is used for basis
Assessment result is to be determined as the customized custom content for waking up word to generate threshold wake-up value.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment comprising: at least one processor, and with extremely
The memory of few processor communication connection, wherein memory is stored with the instruction that can be executed by least one processor, refers to
It enables and being executed by least one described processor, so that at least one processor is able to carry out the customized determination method for waking up word
Step.
Fourth aspect, the embodiment of the present invention provide a kind of storage medium, are stored thereon with computer program, which is located
The step of reason device realizes the customized determination method for waking up word when executing.
The beneficial effect of the embodiment of the present invention is: based on the embodiment of the present invention customized wake-up word determination method and
Device may be implemented to obtain the low customized wake-up word of false wake-up rate, and the process for obtaining wake-up word is more efficiently rapid,
And the wake-up word of generation can also be assessed according to word threshold value is waken up, greatly improve the utilization rate for waking up word.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are some embodiments of the invention, for this field
For those of ordinary skill, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the determination method flow diagram of the customized wake-up word of an embodiment of the present invention;
Fig. 2 is the determination method flow diagram of the customized wake-up word of the present invention one and embodiment;
Fig. 3 is an embodiment of the present invention for determining the customized device block diagram for waking up word;
Fig. 4 is the electronic devices structure schematic diagram of an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.
The present invention can describe in the general context of computer-executable instructions executed by a computer, such as program
Module.Generally, program module includes routines performing specific tasks or implementing specific abstract data types, programs, objects, member
Part, data structure etc..The present invention can also be practiced in a distributed computing environment, in these distributed computing environments, by
Task is executed by the connected remote processing devices of communication network.In a distributed computing environment, program module can be with
In the local and remote computer storage media including storage equipment.
In the present invention, the fingers such as " module ", " device ", " system " are applied to the related entities of computer, such as hardware, hardware
Combination, software or software in execution with software etc..In detail, for example, element can with but be not limited to run on processing
Process, processor, object, executable element, execution thread, program and/or the computer of device.In addition, running on server
Application program or shell script, server can be element.One or more elements can be in the process and/or thread of execution
In, and element can be localized and/or be distributed between two or multiple stage computers on one computer, and can be by each
Kind computer-readable medium operation.Element can also according to the signal with one or more data packets, for example, from one with
Another element interacts in local system, distributed system, and/or the network in internet passes through signal and other system interactions
The signals of data communicated by locally and/or remotely process.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise", not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or equipment institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence " including ... ", it is not excluded that including described want
There is also other identical elements in the process, method, article or equipment of element.
The method and device for determining customized wake-up word in the embodiment of the present invention is applied to terminal device, the intelligence
Display interface, which can be projected out, configured with display screen or the terminal device in energy terminal interacts operation, example for user
Such as, any Intelligent hardware such as smart television, smart phone, tablet computer, PC, smart home, projector, the present invention do not make this
It limits.
Fig. 1 schematically shows the customized determination method flow diagram for waking up word according to the present invention.As shown in Figure 1,
The present embodiment includes the following steps:
Step S101: the first user instruction is received.It is implemented as, user inputs the wake-up word wanted to set up, with root
The first user instruction is generated according to the wake-up word content of user setting, i.e. the first user instruction includes in the wake-up word of user setting
Hold, which is Chinese character.
Step S102: custom content is determined according to the first user instruction.
For the implementation method of the step, illustratively implemented these as in the present embodiment:
Be provided with phonetic transcriptions of Chinese characters dictionary in advance, inside include the profession such as xinhua dictionary Chinese character dictionary, can make every
A Chinese character has corresponding phonetic.Due to the extensive knowledge and profound scholarship of Chinese character, many Chinese characters are all often polyphones, and some is that people is ripe
The Chinese character known often has many uncommon pronunciations, so in a preferred embodiment, can also by the phonetic transcriptions of Chinese characters dictionary into
Row optimization, the pronunciation of uncommon polyphone is screened out, common pronunciation is retained.Such as " open country " word is practical " ye " and " ya "
Pronunciation due to " ya " this pronunciation and is of little use, so can screen out " ya " this pronunciation when optimizing to it, can be improved in this way
The efficiency of subsequent processing.The wake-up word of user's input can be single word, phrase or sentence, for example user wants with " I arrives "
Phrase is as wake-up word.
In a particular application, it when the first user instruction for receiving user's sending, that is, receives user and is manually entered submission
When customized wake-up word, the customized Chinese character of setting can be first obtained, and each Chinese character in customized Chinese character is subjected to phonetic and is turned
It changes, determines optional pronunciation sequence, i.e., first customized Chinese character is split, determine single Chinese character, such as " I arrives ", torn open
It is divided into " I ", " arriving ", " " three words, the pronunciation of single Chinese character is determined according to preferred phonetic transcriptions of Chinese characters dictionary or phonetic transcriptions of Chinese characters library,
" wo ", " dao ", " le " are illustratively corresponded to, the pronunciation possible more than one of the single Chinese character determined in specific example
It is a, later optional pronunciation sequence can be generated according to the pronunciation of single Chinese character.Preferably, in order to avoid determining optional pronunciation sequence
Column are too many, and semantic parsing can also be carried out to customized Chinese character, optional pronunciation sequence is carried out according to the semanteme of customized Chinese character
Screening, the optional pronunciation sequence output more to be tallied with the actual situation.Only has one for the optional pronunciation sequence determined
Situation, it can the case where immediately arriving at unique pronunciation sequence, so that it may which directly determining customized content is that this is unique
Pronunciation.The case where having more than one for the optional pronunciation sequence determined, can then be determined unique according to semanteme parsing
Pronounce sequence, or using method shown in Fig. 2 carry out pronunciation sequence determination.Since phonetic transcriptions of Chinese characters dictionary is different from the prior art
To be stored in processing region, but store into external database, so being not take up data, the speed of processing also can be more
Fastly, more efficient.
Step S103: custom content is carried out to wake up word assessment.Specific implementation are as follows: accurately customized interior when getting
Rong Hou, in order to reduce false wake-up rate, it is necessary to the wake-up word to be assessed, the content of assessment includes sensitive word detection, such as
Include the word of state leader, political factor etc., repeat the not full word detection of folded word detection, spoken word detection and sounding,
The method wherein assessed the content for waking up word is referred to the prior art and realizes.
Step S104: customized wake-up word is determined according to assessment result.Specific implementation are as follows: to custom content according to upper
After the assessment content evaluation stated, assessment result can be obtained, if meeting when custom content contains the word for not meeting assessment content
It reminds user's modification or re-enters custom content appropriate.When custom content meets assessment content, i.e., this is customized
When content is suitable for waking up word, assessment result will be generated.In order to improve the utilization rate and reduction false wake-up rate that wake up word, just need
Threshold wake-up value is determined for the custom content, custom content and threshold wake-up value are determined as to final customized wake-up word.
It can be configured at least two Chinese character threshold value dictionaries in the database, it is right which can distribute its according to machine experience for each word
The threshold value answered generates threshold value dictionary, such as counts to the wake-up word in successive dynasties, the occurrence rate of " I " this word is higher, and score value is
0.6, " to " this word goes out occurrence height, and score value is " 0.7 ", and the appearance of " " this word is less high, and score value is " 0.3 ", by these
Score value is added to obtain joint fractional, as the threshold wake-up value for waking up word.
The determination method of the customized wake-up word provided according to the present invention may be implemented high according to the customized content of user
The determination of effect its if appropriate for doing wake-up word, and be also that it is configured with threshold value, greatly improve the customized wake-up of user
The utilization rate of word, and reduce false wake-up rate.
Fig. 2 schematically show according to the present invention one again embodiment customized wake-up word determination method flow
Figure.As shown in Fig. 2, the present embodiment includes the following steps:
Step S201: the first user instruction is received.Specific implementation is referred to step S101.
Step S202: obtaining the customized Chinese character of input according to the first user instruction, carries out phonetic to customized Chinese character and turns
It changes, determine optional pronunciation sequence and is presented to the user.Its specific implementation and step S102 are essentially identical, and difference exists
In, since the words and phrases of user's input may contain multiple polyphones, such as " who why ", " who " word contains there are two types of pronunciation,
" shui ", " shei ", " dry " word contain the sound and the four tones of standard Chinese pronunciation there are two types of pronunciation " gan ", word there are two types of pronunciation, " ma " softly and
Two sound.So this four word arrangement combinations have 6 kinds of pronunciations, in the case where the combination there are many pronunciation, cannot directly present
Give user determine custom content, need the preferred dictionary mentioned according to last embodiment, by this six kinds pronounce combination into
Row sequence, i.e., it is most common to put in the first place, it is presented to the user in the form of a list later.
Step S203: it receives user and is instructed according to the second user that the content presented issues, instructed according to second user
Determine the specified pronunciation sequence of the customized Chinese character of input.Specific implementation are as follows: presented in the form of a list when possible pronunciation
When to user, user can select the pronunciation for meeting the wake-up word of oneself setting according to list, can issue second user after determining
Instruction wakes up word using the selected voice as final.
Step S204 to step S205 is referred to shown in step S103 to step S104, herein without repeating.
According to the present embodiment can solve Chinese character involved in polyphone and lead to the problem of identification process inaccuracy.
The setting method of the customized wake-up word of above-described embodiment can be suitable for simultaneously language model and without language model,
When suitable for language model, it can be further ensured that the customized availability for waking up word, it is higher certainly that wake-up rate is provided in guarantee
Definition wakes up word, promotes user experience.And when suitable for no language model, by the customized wake-up word of the user received
Pronunciation and the detection of word order are carried out, to determine the orthoepy of the customized Chinese character of input, so that without customized under language model
The availability for waking up word is higher, more acurrate.In addition, since the embodiment of the present invention will test the dictionaries store of pronunciation and word order in number
According in library, and dictionary dictionary is optimized, the occupancy of internal resource can be thus substantially reduced, so that processing speed is more
Fastly.
In a preferred embodiment, Chinese character threshold value dictionary there are two being configured in database, one of Chinese character threshold value dictionary
For high threshold dictionary, another Chinese character threshold value dictionary is Low threshold dictionary, wherein is set in high threshold dictionary to the threshold value of each Chinese character
It sets higher, leads to be easy to the case where waking up since threshold value is lower to be applicable in practical application, to every in Low threshold dictionary
The threshold value setting of a Chinese character is lower, leads to be not easy the case where waking up since threshold value is higher to be applicable in practical application, thus
It can guarantee wake-up rate.
Fig. 3 schematically shows the device frame for being used to determine customized wake-up word according to an embodiment of the present invention
Figure.As shown in figure 3,
For determining that the customized device 1 for waking up word includes the first receiving module 2, custom content acquisition module 3, wakes up
Word evaluation module 4 and threshold wake-up value generation module 5.
First receiving module 2 may be implemented to receive user's input by user interface for the first user instruction of reception, by
User clicks determination after being manually entered customized content, that is, is considered as and issues the first user instruction, the first receiving module 2
Receive the instruction.
Custom content obtains module 3 and is used to determine that custom content, custom content are obtained according to the first user instruction
Module 3 includes pronunciation retrieval unit 301 and custom content determination unit 302.The retrieval unit 301 that pronounces is used for root
The optional pronunciation sequence of the customized Chinese character of input is determined according to the first user instruction and is presented to the user.It is referred to the reality of Fig. 1
When the customized Chinese character of existing mode, i.e. user's output only uniquely pronounces, it can confirm without user and directly present.It makes by oneself
Adopted content determining unit 302 is used to receive user and is instructed according to the second user that the content presented issues, according to second user
Instruction determines the specified pronunciation sequence of the customized Chinese character of input.It is referred to the implementation of Fig. 2, i.e., what user inputted makes by oneself
Adopted Chinese character will carry out priority arrangement according to the dictionary built in device, present in the form of a list there are many when pronunciation combination
To user, second of selection is carried out by user and is confirmed, that is, generates second user instruction.
In addition, device further includes built-in dictionary, specifically comprising the first phonetic transcriptions of Chinese characters dictionary 6 and preferred Chinese character are spelled
Sound dictionary 7, the first phonetic transcriptions of Chinese characters dictionary 6 are used to store the phonetic of Chinese character and all pronunciations being adapted to each Chinese character;It is preferred that the Chinese
Word pinyin lexicon 7 is used to store the phonetic of Chinese character and the common pronunciation being adapted to each Chinese character.Pronunciation retrieval unit 301
The optional pronunciation sequence of the customized Chinese character of input is determined according to the first phonetic transcriptions of Chinese characters dictionary 6 or the preferred phonetic transcriptions of Chinese characters dictionary 7.
The workload that processing can be substantially reduced according to preferred phonetic transcriptions of Chinese characters dictionary 7 can be fast for some pronunciations being of little use
Speed screens out.Also, the two dictionaries are not take up the resource space of device, store into external database, and can be with
Real-time update thus can be improved the processing speed of device, improve efficiency.
It wakes up word evaluation module 4 to be used to carry out waking up word assessment to custom content, wherein wake up word evaluation module 4
Customized Chinese character is carried out according to specified pronunciation sequence to wake up word assessment, the content of assessment includes sensitive word detection, such as includes
The word of state leader, political factor etc. repeats the not full word detection of folded word detection, spoken word detection and sounding.Assessment
Mode is referred to above method part, herein without repeating.
Threshold wake-up value generation module 5 is used to be to be determined as the customized custom content life for waking up word according to assessment result
At threshold wake-up value, custom content and threshold wake-up value are determined as customized wake-up word and exports or stores.Wherein it is determined that arousal threshold
Value can be by configuring Chinese character threshold value dictionary, and the description based on above method part is realized.Wake-up thus can be improved
The utilization rate of word simultaneously reduces false wake-up rate.
The device can not only realize it is customized wake up word setting, and be arranged customized wake-up word it is with higher can
With property and wake-up rate.And since the device can be good at guaranteeing the quality of the wake-up word of customized part, so that sharp
The product that wake-up word defines is carried out with it to be benefited, and improves the playability of speech production, the user experience of entire product obtains very
Big promotion.
In some embodiments, the embodiment of the present invention provides a kind of non-volatile computer readable storage medium storing program for executing, described to deposit
Being stored in storage media one or more includes the programs executed instruction, it is described execute instruction can by electronic equipment (including but
It is not limited to computer, server or the network equipment etc.) it reads and executes, to be made by oneself for executing any of the above-described of the present invention
Justice wakes up the determination method of word.
In some embodiments, the embodiment of the present invention also provides a kind of computer program product, and the computer program produces
Product include the computer program being stored on non-volatile computer readable storage medium storing program for executing, and the computer program includes that program refers to
It enables, when described program instruction is computer-executed, the computer is made to execute the customized determination for waking up word of any of the above-described
Method.
In some embodiments, the embodiment of the present invention also provides a kind of electronic equipment comprising: at least one processor,
And the memory being connect at least one described processor communication, wherein the memory is stored with can be by described at least one
The instruction that a processor executes, described instruction is executed by least one described processor, so that at least one described processor energy
Enough execute the customized determination method for waking up word.
In some embodiments, the embodiment of the present invention also provides a kind of storage medium, is stored thereon with computer program,
It is characterized in that, the program customized determination method for waking up word when being executed by processor.
The device for determining customized wake-up word of the embodiments of the present invention can be used for executing the embodiment of the present invention
The customized determination method for waking up word, and the customized determination side for waking up word of realization for reaching the embodiments of the present invention accordingly
Method technical effect achieved, which is not described herein again.Hardware processor (hardware can be passed through in the embodiment of the present invention
Processor) Lai Shixian related function module.
Fig. 4 is the hardware of the electronic equipment for the customized determination method for waking up word of execution that another embodiment of the application provides
Structural schematic diagram, as shown in figure 4, the equipment includes:
One or more processors 410 and memory 420, in Fig. 4 by taking a processor 410 as an example.
The equipment for executing the customized determination method for waking up word can also include: input unit 430 and output device 440.
Processor 410, memory 420, input unit 430 and output device 440 can pass through bus or other modes
It connects, in Fig. 4 for being connected by bus.
Memory 420 is used as a kind of non-volatile computer readable storage medium storing program for executing, can be used for storing non-volatile software journey
Sequence, non-volatile computer executable program and module, such as the determination method of the customized wake-up word in the embodiment of the present application
Corresponding program instruction/module.Processor 410 is by running the non-volatile software program being stored in memory 420, instruction
And module, thereby executing the various function application and data processing of server, i.e. realization above method embodiment is customized
Wake up the determination method of word.
Memory 420 may include storing program area and storage data area, wherein storing program area can store operation system
Application program required for system, at least one function;Storage data area can be stored to be created according to using for phonetic controller
Data etc..In addition, memory 420 may include high-speed random access memory, it can also include nonvolatile memory, example
Such as at least one disk memory, flush memory device or other non-volatile solid state memory parts.In some embodiments, it deposits
Optional reservoir 420 includes the memory remotely located relative to processor 410, these remote memories can pass through network connection
To phonetic controller.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, mobile radio communication
And combinations thereof.
Input unit 430 can receive the number or character information of input, and generates and fill with customized determining for wake-up word
The related signal of user setting and function control set.Output device 440 may include that display screen etc. shows equipment.
One or more of modules are stored in the memory 420, when by one or more of processors
When 410 execution, the determination method of the customized wake-up word in above-mentioned any means embodiment is executed.
Method provided by the embodiment of the present application can be performed in the said goods, has the corresponding functional module of execution method and has
Beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to method provided by the embodiment of the present application.
The electronic equipment of the embodiment of the present application exists in a variety of forms, including but not limited to:
(1) mobile communication equipment: the characteristics of this kind of equipment is that have mobile communication function, and to provide speech, data
Communication is main target.This Terminal Type includes: smart phone (such as iPhone), multimedia handset, functional mobile phone and low
Hold mobile phone etc..
(2) super mobile personal computer equipment: this kind of equipment belongs to the scope of personal computer, there is calculating and processing function
Can, generally also have mobile Internet access characteristic.This Terminal Type includes: PDA, MID and UMPC equipment etc., such as iPad.
(3) portable entertainment device: this kind of equipment can show and play multimedia content.Such equipment include: audio,
Video player (such as iPod), handheld device, e-book and intelligent toy and portable car-mounted navigation equipment.
(4) server: providing the equipment of the service of calculating, and the composition of server includes that processor, hard disk, memory, system are total
Line etc., server is similar with general computer architecture, but due to needing to provide highly reliable service, in processing energy
Power, stability, reliability, safety, scalability, manageability etc. are more demanding.
(5) other electronic devices with data interaction function.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of software plus general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, above-mentioned technology
Scheme substantially in other words can be embodied in the form of software products the part that the relevant technologies contribute, the computer
Software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions to
So that computer equipment (can be personal computer, server or the network equipment etc.) execute each embodiment or
Method described in certain parts of embodiment.
Finally, it should be noted that above embodiments are only to illustrate the technical solution of the application, rather than its limitations;Although
The application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (11)
1. a kind of customized determination method for waking up word, comprising:
Receive the first user instruction;
Custom content is determined according to first user instruction;
Custom content is carried out to wake up word assessment;
Customized wake-up word is determined according to assessment result.
2. described to determine custom content packet according to first user instruction according to the method described in claim 1, wherein
It includes:
The customized Chinese character of input is obtained according to the first user instruction;
Phonetic conversion is carried out to customized Chinese character, optional pronunciation sequence is determined and is presented to the user;
User is received to be instructed according to the second user that the content presented issues;
The specified pronunciation sequence for determining the customized Chinese character of input is instructed according to the second user.
3. according to the method described in claim 2, wherein, further includes:
Configure phonetic transcriptions of Chinese characters dictionary;
The phonetic transcriptions of Chinese characters dictionary of configuration is optimized, preferred phonetic transcriptions of Chinese characters dictionary is generated;
It is described that phonetic conversion is carried out to customized Chinese character, determine that optional pronunciation sequence includes:
Customized Chinese character is split, determines single Chinese character;
The pronunciation of single Chinese character is determined according to preferred phonetic transcriptions of Chinese characters dictionary;
According to the semanteme of customized Chinese character and the pronunciation of single Chinese character, optional pronunciation sequence is generated.
4. method according to any one of claims 1 to 3, wherein include to the wake-up word assessment that custom content carries out
Sensitive word detection repeats the not full word detection of folded word detection, spoken word detection and sounding.
5. according to the method described in claim 4, it is characterized in that, described determine customized wake-up word packet according to assessment result
It includes:
When assessment result is that the custom content is suitable for waking up word, threshold wake-up value is determined for the custom content, it will
The custom content and threshold wake-up value are determined as customized wake-up word.
6. according to the method described in claim 5, it is characterized by further comprising:
Configure at least two Chinese character threshold value dictionaries;
It is described to determine that threshold wake-up value includes: for the custom content
The threshold wake-up value of custom content is generated according to custom content and one of Chinese character threshold value dictionary.
7. a kind of for determining the customized device for waking up word characterized by comprising
First receiving module, for receiving the first user instruction;
Custom content obtains module, for determining custom content according to first user instruction;
Word evaluation module is waken up, wakes up word assessment for carrying out to custom content;
Threshold wake-up value generation module, for being to be determined as the customized custom content generation for waking up word to call out according to assessment result
Awake threshold value.
8. device according to claim 7, which is characterized in that the custom content obtains module and includes
Pronounce retrieval unit, the optional pronunciation sequence of the customized Chinese character for determining input according to first user instruction
It arranges and is presented to the user;
Custom content determination unit is instructed for receiving user according to the second user that the content presented issues, according to institute
State the specified pronunciation sequence that second user instruction determines the customized Chinese character of input;
The wake-up word evaluation module carries out customized Chinese character according to the specified pronunciation sequence to wake up word assessment.
9. device according to claim 8, wherein further include:
First phonetic transcriptions of Chinese characters dictionary, for storing the phonetic of Chinese character and all pronunciations being adapted to each Chinese character;
It is preferred that phonetic transcriptions of Chinese characters dictionary, for storing the phonetic of Chinese character and the common pronunciation being adapted to each Chinese character;
The pronunciation retrieval unit determines defeated according to the first phonetic transcriptions of Chinese characters dictionary or the preferred phonetic transcriptions of Chinese characters dictionary
The optional pronunciation sequence of the customized Chinese character entered.
10. a kind of electronic equipment comprising: at least one processor, and connect at least one described processor communication
Memory, wherein the memory be stored with can by least one described processor execute instruction, described instruction by it is described extremely
A few processor executes, so that at least one described processor is able to carry out any one of claim 1-6 the method
The step of.
11. a kind of storage medium, is stored thereon with computer program, which is characterized in that the realization when program is executed by processor
The step of any one of claim 1-6 the method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811593641.7A CN109767763B (en) | 2018-12-25 | 2018-12-25 | Method and device for determining user-defined awakening words |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811593641.7A CN109767763B (en) | 2018-12-25 | 2018-12-25 | Method and device for determining user-defined awakening words |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109767763A true CN109767763A (en) | 2019-05-17 |
CN109767763B CN109767763B (en) | 2021-01-26 |
Family
ID=66450263
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811593641.7A Active CN109767763B (en) | 2018-12-25 | 2018-12-25 | Method and device for determining user-defined awakening words |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109767763B (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN110838289A (en) * | 2019-11-14 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Awakening word detection method, device, equipment and medium based on artificial intelligence |
CN111128138A (en) * | 2020-03-30 | 2020-05-08 | 深圳市友杰智新科技有限公司 | Voice wake-up method and device, computer equipment and storage medium |
CN111292726A (en) * | 2020-03-10 | 2020-06-16 | 科通工业技术(深圳)有限公司 | Method and system for changing awakening words offline |
CN112009493A (en) * | 2020-09-03 | 2020-12-01 | 三一专用汽车有限责任公司 | Awakening method of vehicle-mounted control system, vehicle-mounted control system and vehicle |
CN112164395A (en) * | 2020-09-18 | 2021-01-01 | 北京百度网讯科技有限公司 | Vehicle-mounted voice starting method and device, electronic equipment and storage medium |
EP3832643A1 (en) * | 2019-12-05 | 2021-06-09 | SoundHound, Inc. | Dynamic wakewords for speech-enabled devices |
Citations (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003000571A (en) * | 2001-06-18 | 2003-01-07 | Mitsubishi Electric Corp | Device and method for estimating arousal level |
CN101067780A (en) * | 2007-06-21 | 2007-11-07 | 腾讯科技(深圳)有限公司 | Character inputting system and method for intelligent equipment |
US20080046250A1 (en) * | 2006-07-26 | 2008-02-21 | International Business Machines Corporation | Performing a safety analysis for user-defined voice commands to ensure that the voice commands do not cause speech recognition ambiguities |
CN101334704A (en) * | 2008-06-27 | 2008-12-31 | 中国科学院软件研究所 | Multichannel Chinese input method facing to mobile equipment |
CN102063900A (en) * | 2010-11-26 | 2011-05-18 | 北京交通大学 | Speech recognition method and system for overcoming confusing pronunciation |
CN103095911A (en) * | 2012-12-18 | 2013-05-08 | 苏州思必驰信息科技有限公司 | Method and system for finding mobile phone through voice awakening |
CN104620314A (en) * | 2012-04-26 | 2015-05-13 | 纽昂斯通讯公司 | Embedded system for construction of small footprint speech recognition with user-definable constraints |
US20150248882A1 (en) * | 2012-07-09 | 2015-09-03 | Nuance Communications, Inc. | Detecting potential significant errors in speech recognition results |
CN105009203A (en) * | 2013-03-12 | 2015-10-28 | 纽昂斯通讯公司 | Methods and apparatus for detecting a voice command |
CN105068987A (en) * | 2010-01-05 | 2015-11-18 | 谷歌公司 | Word-level correction of speech input |
CN105528404A (en) * | 2015-12-03 | 2016-04-27 | 北京锐安科技有限公司 | Establishment method and apparatus of seed keyword dictionary, and extraction method and apparatus of keywords |
CN105654785A (en) * | 2016-03-18 | 2016-06-08 | 上海语知义信息技术有限公司 | Personalized spoken foreign language learning system and method |
CN105654946A (en) * | 2014-12-02 | 2016-06-08 | 三星电子株式会社 | Method and apparatus for speech recognition |
US20160189716A1 (en) * | 2013-10-11 | 2016-06-30 | Apple Inc. | Speech recognition wake-up of a handheld portable electronic device |
CN106098059A (en) * | 2016-06-23 | 2016-11-09 | 上海交通大学 | customizable voice awakening method and system |
CN106233374A (en) * | 2014-04-17 | 2016-12-14 | 高通股份有限公司 | Generate for detecting the keyword model of user-defined keyword |
US9600231B1 (en) * | 2015-03-13 | 2017-03-21 | Amazon Technologies, Inc. | Model shrinking for embedded keyword spotting |
CN106611597A (en) * | 2016-12-02 | 2017-05-03 | 百度在线网络技术(北京)有限公司 | Voice wakeup method and voice wakeup device based on artificial intelligence |
CN106844343A (en) * | 2017-01-20 | 2017-06-13 | 上海傲硕信息科技有限公司 | Instruction results screening plant |
US20170186430A1 (en) * | 2013-12-05 | 2017-06-29 | Google Inc. | Promoting voice actions to hotwords |
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 |
CN104584119B (en) * | 2012-07-03 | 2017-10-17 | 谷歌公司 | Determine hot word grade of fit |
CN108536668A (en) * | 2018-02-26 | 2018-09-14 | 科大讯飞股份有限公司 | Wake up word appraisal procedure and device, storage medium, electronic equipment |
CN108564943A (en) * | 2018-04-27 | 2018-09-21 | 京东方科技集团股份有限公司 | voice interactive method and system |
CN108986813A (en) * | 2018-08-31 | 2018-12-11 | 出门问问信息科技有限公司 | Wake up update method, device and the electronic equipment of word |
-
2018
- 2018-12-25 CN CN201811593641.7A patent/CN109767763B/en active Active
Patent Citations (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003000571A (en) * | 2001-06-18 | 2003-01-07 | Mitsubishi Electric Corp | Device and method for estimating arousal level |
US20080046250A1 (en) * | 2006-07-26 | 2008-02-21 | International Business Machines Corporation | Performing a safety analysis for user-defined voice commands to ensure that the voice commands do not cause speech recognition ambiguities |
CN101067780A (en) * | 2007-06-21 | 2007-11-07 | 腾讯科技(深圳)有限公司 | Character inputting system and method for intelligent equipment |
CN101334704A (en) * | 2008-06-27 | 2008-12-31 | 中国科学院软件研究所 | Multichannel Chinese input method facing to mobile equipment |
CN105068987A (en) * | 2010-01-05 | 2015-11-18 | 谷歌公司 | Word-level correction of speech input |
CN102063900A (en) * | 2010-11-26 | 2011-05-18 | 北京交通大学 | Speech recognition method and system for overcoming confusing pronunciation |
CN104620314A (en) * | 2012-04-26 | 2015-05-13 | 纽昂斯通讯公司 | Embedded system for construction of small footprint speech recognition with user-definable constraints |
CN104584119B (en) * | 2012-07-03 | 2017-10-17 | 谷歌公司 | Determine hot word grade of fit |
US20150248882A1 (en) * | 2012-07-09 | 2015-09-03 | Nuance Communications, Inc. | Detecting potential significant errors in speech recognition results |
CN103095911A (en) * | 2012-12-18 | 2013-05-08 | 苏州思必驰信息科技有限公司 | Method and system for finding mobile phone through voice awakening |
CN105009203A (en) * | 2013-03-12 | 2015-10-28 | 纽昂斯通讯公司 | Methods and apparatus for detecting a voice command |
US20160189716A1 (en) * | 2013-10-11 | 2016-06-30 | Apple Inc. | Speech recognition wake-up of a handheld portable electronic device |
US20170186430A1 (en) * | 2013-12-05 | 2017-06-29 | Google Inc. | Promoting voice actions to hotwords |
CN106233374A (en) * | 2014-04-17 | 2016-12-14 | 高通股份有限公司 | Generate for detecting the keyword model of user-defined keyword |
CN105654946A (en) * | 2014-12-02 | 2016-06-08 | 三星电子株式会社 | Method and apparatus for speech recognition |
US9600231B1 (en) * | 2015-03-13 | 2017-03-21 | Amazon Technologies, Inc. | Model shrinking for embedded keyword spotting |
CN105528404A (en) * | 2015-12-03 | 2016-04-27 | 北京锐安科技有限公司 | Establishment method and apparatus of seed keyword dictionary, and extraction method and apparatus of keywords |
CN105654785A (en) * | 2016-03-18 | 2016-06-08 | 上海语知义信息技术有限公司 | Personalized spoken foreign language learning system and method |
CN106098059A (en) * | 2016-06-23 | 2016-11-09 | 上海交通大学 | customizable voice awakening method and system |
CN106611597A (en) * | 2016-12-02 | 2017-05-03 | 百度在线网络技术(北京)有限公司 | Voice wakeup method and voice wakeup device based on artificial intelligence |
CN106844343A (en) * | 2017-01-20 | 2017-06-13 | 上海傲硕信息科技有限公司 | Instruction results screening plant |
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 |
CN108536668A (en) * | 2018-02-26 | 2018-09-14 | 科大讯飞股份有限公司 | Wake up word appraisal procedure and device, storage medium, electronic equipment |
CN108564943A (en) * | 2018-04-27 | 2018-09-21 | 京东方科技集团股份有限公司 | voice interactive method and system |
CN108986813A (en) * | 2018-08-31 | 2018-12-11 | 出门问问信息科技有限公司 | Wake up update method, device and the electronic equipment of word |
Non-Patent Citations (2)
Title |
---|
A. ZEHETNER, M. HAGM¨ULLER, AND F. PERNKOPF: "WAKE-UP-WORD SPOTTING FOR MOBILE SYSTEMS", 《2014 22ND EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO)》 * |
左祥,巴振宇等: "基于神经网络和身份认证矢量的自定义唤醒词检测", 《第十三届全国人机语音通讯学术会议》 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN110838289A (en) * | 2019-11-14 | 2020-02-25 | 腾讯科技(深圳)有限公司 | Awakening word detection method, device, equipment and medium based on artificial intelligence |
CN110838289B (en) * | 2019-11-14 | 2023-08-11 | 腾讯科技(深圳)有限公司 | Wake-up word detection method, device, equipment and medium based on artificial intelligence |
EP3832643A1 (en) * | 2019-12-05 | 2021-06-09 | SoundHound, Inc. | Dynamic wakewords for speech-enabled devices |
US11295741B2 (en) | 2019-12-05 | 2022-04-05 | Soundhound, Inc. | Dynamic wakewords for speech-enabled devices |
US11948571B2 (en) | 2019-12-05 | 2024-04-02 | Soundhound Ai Ip, Llc | Wakeword selection |
CN111292726A (en) * | 2020-03-10 | 2020-06-16 | 科通工业技术(深圳)有限公司 | Method and system for changing awakening words offline |
CN111128138A (en) * | 2020-03-30 | 2020-05-08 | 深圳市友杰智新科技有限公司 | Voice wake-up method and device, computer equipment and storage medium |
CN112009493A (en) * | 2020-09-03 | 2020-12-01 | 三一专用汽车有限责任公司 | Awakening method of vehicle-mounted control system, vehicle-mounted control system and vehicle |
CN112164395A (en) * | 2020-09-18 | 2021-01-01 | 北京百度网讯科技有限公司 | Vehicle-mounted voice starting method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109767763B (en) | 2021-01-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109767763A (en) | It is customized wake up word determination method and for determine it is customized wake up word device | |
US10943606B2 (en) | Context-based detection of end-point of utterance | |
US10217463B2 (en) | Hybridized client-server speech recognition | |
US10540970B2 (en) | Architectures and topologies for vehicle-based, voice-controlled devices | |
CN106201424B (en) | A kind of information interacting method, device and electronic equipment | |
US9542956B1 (en) | Systems and methods for responding to human spoken audio | |
JP6507316B2 (en) | Speech re-recognition using an external data source | |
US11935521B2 (en) | Real-time feedback for efficient dialog processing | |
CN111090728B (en) | Dialogue state tracking method and device and computing equipment | |
US9431005B2 (en) | System and method for supplemental speech recognition by identified idle resources | |
CN109119067B (en) | Speech synthesis method and device | |
CN109637548A (en) | Voice interactive method and device based on Application on Voiceprint Recognition | |
US10460719B1 (en) | User feedback for speech interactions | |
CN103635963A (en) | Cross-lingual initialization of language models | |
CN104123938A (en) | Voice control system, electronic device and voice control method | |
JP2019133127A (en) | Voice recognition method, apparatus and server | |
US10629199B1 (en) | Architectures and topologies for vehicle-based, voice-controlled devices | |
CN111081254B (en) | Voice recognition method and device | |
US11295732B2 (en) | Dynamic interpolation for hybrid language models | |
CN110099295A (en) | Voice control method for television set, device, equipment and storage medium | |
US11996081B2 (en) | Visual responses to user inputs | |
CN105446123A (en) | Voice intelligent alarm clock | |
CN107919127A (en) | Method of speech processing, device and electronic equipment | |
JP6306447B2 (en) | Terminal, program, and system for reproducing response sentence using a plurality of different dialogue control units simultaneously | |
US11462208B2 (en) | Implementing a correction model to reduce propagation of automatic speech recognition errors |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CP01 | Change in the name or title of a patent holder | ||
CP01 | Change in the name or title of a patent holder |
Address after: 215123 14 Tengfei Innovation Park, 388 Xinping street, Suzhou Industrial Park, Suzhou, Jiangsu. Patentee after: Sipic Technology Co.,Ltd. Address before: 215123 14 Tengfei Innovation Park, 388 Xinping street, Suzhou Industrial Park, Suzhou, Jiangsu. Patentee before: AI SPEECH Ltd. |