CN103871409B

CN103871409B - A kind of method of speech recognition, the method for information processing and electronic equipment

Info

Publication number: CN103871409B
Application number: CN201210548625.2A
Authority: CN
Inventors: 戴海生; 王天树; 王茜莺; 李向阳
Original assignee: Lenovo Beijing Ltd
Current assignee: Lenovo Beijing Ltd
Priority date: 2012-12-17
Filing date: 2012-12-17
Publication date: 2018-01-23
Anticipated expiration: 2032-12-17
Also published as: CN103871409A

Abstract

The invention discloses a kind of method of speech recognition, the method for information processing and electronic equipment, the method for the speech recognition, applied at least two different sound identification modules in an electronic equipment, are included in the electronic equipment, methods described includes：Obtain a voice messaging to be identified；The present mode for obtaining the electronic equipment is the first use pattern；Based on corresponding relation between use pattern and sound identification module, it is determined that from first use pattern corresponding to sound identification module be described at least two different sound identification modules in the first sound identification module；The voice messaging to be identified is identified by first sound identification module, obtains a recognition result.

Description

A kind of method of speech recognition, the method for information processing and electronic equipment

Technical field

The present invention relates to multimedia technology field, more particularly to a kind of method of speech recognition, the method for information processing and Electronic equipment.

Background technology

With the continuous development of electronic equipment, its corresponding control mode is also more and more, such as：Voice command, mouse Control, touch screen control etc..Wherein, Voice command mode receives increasing user due to easy to operate and rapid Welcome.

In the prior art, electronic equipment would generally include a speech recognition engine, wherein, in the speech recognition engine Include the various speech recognition modelings that are identified of voice messaging to user's input, by the speech recognition engine to Family input voice messaging be identified, just can obtain corresponding to phonetic order or other voice messagings.

In the prior art, it will usually have the following two kinds speech recognition engine：

The first, hardware based speech recognition engine, it is capable of speech letter that the identification engine, which does not need system work, Breath, but due to being confined to hardware technology, so its dictionary is limited, the voice messaging that can be identified is less；

Second, the speech recognition engine based on software, word amount is enriched in the identification engine dictionary, can identify substantial amounts of Voice messaging, but need to start the system of electronic equipment first, it can be identified.

Present inventor at least has found exist in the prior art during the embodiment of the present application technical scheme is realized Following technical problem：

Due in the prior art, if carrying out speech recognition using the first scheme, because dictionary is limited, so identification It is not accurate enough；And if be identified using second scheme, due to need start electronic equipment system, so energy consumption compared with Height, thus there is the technical problem that energy consumption can not be reduced while recognition accuracy is ensured in the prior art.

The content of the invention

The embodiment of the present invention provides a kind of method and electronic equipment of speech recognition, for solving existing for prior art not The technique effect of energy consumption can be reduced while recognition accuracy is ensured.

On the one hand, the application provides following technical scheme by an embodiment：

A kind of audio recognition method, it is different applied in an electronic equipment, including at least two in the electronic equipment Sound identification module, methods described includes：

Obtain the first voice messaging to be identified；

The present mode for obtaining the electronic equipment is the first use pattern；

Based on corresponding relation between use pattern and sound identification module, it is determined that voice corresponding with first use pattern Identification module is the first sound identification module in described at least two different sound identification modules；

The described first voice messaging to be identified is identified by first sound identification module, obtains an identification knot Fruit.

Optionally, before the present mode for obtaining the electronic equipment is the first use pattern, methods described is also Including：

When the electronic equipment is in the second use pattern, detection obtains the second voice messaging to be identified；

By from second use pattern corresponding to the second language in described at least two different sound identification modules Described second voice messaging to be identified is identified sound identification module, is entered with determining whether there is the control electronic equipment First control instruction of first use pattern；

When first control instruction be present, the electronic equipment is controlled to enter first use pattern.

Optionally, determine whether there is control the electronic equipment to enter the first use pattern the first control instruction it Afterwards, methods described also includes：

When first control instruction being present, being recorded in and producing the first moment of first control instruction to switching to The first voice messaging that user inputs between second moment of first use pattern.

Optionally, the first moment for producing first control instruction is being recorded in switching to first use pattern The second moment between user input the first voice messaging after, methods described also includes：

First voice messaging is stored in the internal memory of the electronic equipment.

Optionally, the first voice messaging to be identified of the acquisition, is specifically included：

First voice messaging is read from the internal memory；

Obtain in first voice messaging and not yet carry out the first of speech recognition by second sound identification module Sub- voice messaging, the first sub- voice messaging are the described first voice messaging to be identified.

Optionally, after the control electronic equipment enters first use pattern, methods described also includes：

Judge whether to open the second control instruction of first sound identification module；

After it second control instruction be present, step is performed：Based on corresponding between use pattern and sound identification module Relation, it is determined that sound identification module corresponding with first use pattern is the first sound identification module.

Optionally, determine that sound identification module corresponding with first use pattern is the first speech recognition mould described After block, methods described also includes：

Close second sound identification module.

Optionally, first sound identification module is：Software sound identification module or hardware sound identification module；It is described Second sound identification module is：Hardware sound identification module；

First use pattern is：Use pattern corresponding to open state；Second use pattern is to shut down/treat Machine pattern.

Optionally, it is described to be based on corresponding relation between use pattern and sound identification module, it is determined that using mould with described first Sound identification module corresponding to formula is the first sound identification module, is specially：

When first use pattern is the use pattern that power consumption is more than the first predetermined threshold value, based on use pattern and language Corresponding relation between sound identification module, it is determined that sound identification module corresponding with first use pattern is software identification module；

When first use pattern is the use pattern that power consumption is less than first predetermined threshold value, it is determined that with described the Sound identification module corresponding to one use pattern is hardware identification module.

On the other hand, the application provides following technical scheme by another embodiment：

A kind of method of information processing, applied in an electronic equipment, methods described includes：

When the electronic equipment is in the second use pattern, detection obtains the second voice messaging to be identified, wherein, it is described Second use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

By believing with the second sound identification module corresponding to second use pattern the described second voice to be identified Breath identification, it is determined whether the first control instruction for controlling the electronic equipment to enter the first use pattern be present；

When first control instruction be present, the electronic equipment is controlled to enter first use pattern, wherein, institute State the first use pattern and be in use pattern corresponding during normal operating conditions for the electronic equipment.

Determine the first sound identification module corresponding to first use pattern；

Obtain the first voice messaging to be identified；

The described first voice messaging to be identified is identified based on first sound identification module, to obtain at least one Recognition result.

Optionally, refer in first control for determining whether there is the control electronic equipment into the first use pattern After order, methods described also includes：

Optionally, the first voice messaging to be identified of the acquisition, it is specially：

Obtain in first voice messaging and not yet carried out the first of speech recognition by second sound identification module Sub- voice messaging, the first sub- voice messaging are the described first voice messaging to be identified.

A kind of electronic equipment, include at least two different sound identification modules, the electronics in the electronic equipment Equipment also includes：

First obtains chip, for obtaining the first voice messaging to be identified；

Second obtains chip, and the present mode for obtaining the electronic equipment is the first use pattern；

First determines chip, for based on corresponding relation between use pattern and sound identification module, it is determined that with described first Sound identification module corresponding to use pattern is the first speech recognition mould in described at least two different sound identification modules Block；

3rd obtains chip, for being known by first sound identification module to the voice messaging to be identified Not, a recognition result is obtained.

Optionally, the electronic equipment also includes：

First detection chip, for before being the first use pattern in the present mode for obtaining the electronic equipment, working as institute When stating electronic equipment and being in the second use pattern, by from second use pattern corresponding to described at least two different languages The second sound identification module in sound identification module, which detects whether to exist, controls the electronic equipment to use mould into described first First control instruction of formula；

First control chip, for when first control instruction be present, controlling the electronic equipment to enter described the One use pattern.

Optionally, the electronic equipment also includes：

First memorization COMS clip, for determining whether there is the control electronic equipment into the first of the first use pattern After control instruction, when first control instruction being present, being recorded in and producing the first moment of first control instruction extremely Switch to the first voice messaging that user inputs between the second moment of first use pattern.

Optionally, the electronic equipment also includes：

Storage chip, for making to switching to described first being recorded in the first moment for producing first control instruction After the first voice messaging of user's input between the second moment of pattern, first voice messaging is stored in the electricity The internal memory of sub- equipment.

Optionally, described first chip is obtained, specifically included：

Sub- chip is read, for reading first voice messaging from the internal memory；

Sub- chip is obtained, language is not yet carried out by second sound identification module for obtaining in first voice messaging The first sub- voice messaging that sound identified, the first sub- voice messaging are the described first voice messaging to be identified.

Optionally, the electronic equipment also includes：

Judge chip：For after controlling the electronic equipment to enter first use pattern, judging whether Open the second control instruction of first sound identification module；And

Optionally, the electronic equipment also includes：

Chip is closed, for it is determined that sound identification module corresponding with first use pattern is the first speech recognition After module, second sound identification module is closed.

First use pattern is：Use pattern corresponding to open state；Second use pattern is shutdown/standby mould Formula.

Optionally, described first chip is determined, specifically included：

First true stator chip, for being the use pattern that power consumption is more than the first predetermined threshold value when first use pattern When, based on corresponding relation between use pattern and sound identification module, it is determined that speech recognition corresponding with first use pattern Module is software identification module；

Second true stator chip, for being the use that power consumption is less than first predetermined threshold value when first use pattern During pattern, it is determined that sound identification module corresponding with first use pattern is hardware identification module.

On the other hand, the application provides the following technical scheme in Europe by another embodiment：

A kind of electronic equipment, methods described include：

A kind of electronic equipment, the electronic equipment include：

Second detection chip, for when the electronic equipment is in the second use pattern, it is to be identified that detection obtains second Voice messaging, wherein, second use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

Second determine chip, for by with the second sound identification module corresponding to second use pattern to described Second voice messaging identification to be identified, it is determined whether the first control that the control electronic equipment enters the first use pattern be present Instruction；

Second control chip, for when first control instruction be present, controlling the electronic equipment to enter described the One use pattern, wherein, first use pattern is that the electronic equipment is in use corresponding during normal operating conditions Pattern.

Optionally, the electronic equipment also includes：

3rd determines chip, after entering first use pattern in the control electronic equipment, it is determined that described The first sound identification module corresponding to first use pattern；

Chip is obtained, for obtaining the first voice messaging to be identified；

4th obtains chip, for being carried out based on first sound identification module to the described first voice messaging to be identified Identification, to obtain an at least recognition result.

Optionally, the electronic equipment also includes：

Second memorization COMS clip, for determining whether there is the control electronic equipment into the first of the first use pattern After control instruction, when first control instruction being present, being recorded in and producing the first moment of first control instruction extremely Switch to the first voice messaging that user inputs between the second moment of first use pattern.

Optionally, the acquisition chip, is specifically used for：

The one or more technical schemes provided in the embodiment of the present application, have at least the following technical effects or advantages：

(1) at least two different speech recognition moulds are set due in the embodiment of the present application, employing electron equipment Block, so as to the different use patterns based on electronic equipment, the technical scheme of different sound identification modules can be used, so, In the case where electronic equipment is in the less use pattern of control instruction, the relatively low speech recognition system of energy consumption can be used, is only existed Electronic equipment is under the more use pattern of control instruction, just using the higher speech recognition system of energy consumption, so solve The technical problem of recognition accuracy can not be improved while energy consumption is reduced in the prior art, has reached and has ensured recognition accuracy While reduce energy consumption technique effect.

(2) due in the embodiment of the present application, when electronic equipment is in the second use pattern, such as：Shutdown/standby mode When, can by the second sound identification module, such as：Control instruction is identified for hardware identification module, with described in control Electronic equipment enters first use pattern, and voice knowledge can be achieved because hardware identification module need not open electronic equipment Not, so having reached the technique effect for saving energy consumption.

(3) due in the embodiment of the present application, when controlling the electronic equipment to enter first use pattern, recording The first voice messaging that user is inputted, the first speech recognition system so is used in electronic equipment, such as：Software speech recognition During module, first voice messaging can be called by the software sound identification module and then identification, and software voice is known For other module is compared to hardware sound identification module, its vocabulary is more, so reached the technology for improving recognition accuracy Effect.

(4) due in the embodiment of the present application, after control electronics enter first use pattern, first having to The second control instruction of the first sound identification module of unlatching is judged whether, first speech recognition is just opened in the presence of Module, so reach the technique effect for saving energy consumption.

(5) due in the embodiment of the present application, while first sound identification module is opened, closing described second Sound identification module, so reached the technique effect of further reduction energy consumption.

Brief description of the drawings

Fig. 1 is the flow chart of the method for speech recognition in the embodiment of the present application one；

Fig. 2 is the flow chart for controlling the electronic equipment to enter the first use pattern in the method for the embodiment of the present application one；

Fig. 3 is to judge whether to open the flow chart of first sound identification module in the method for the embodiment of the present application one；

Fig. 4 is the flow chart of the method for information processing in the embodiment of the present application four；

Fig. 5 is the process chart entered in the method for the embodiment of the present application four after the first use pattern；

Fig. 6 is the structure chart of electronic equipment in the embodiment of the present application five；

Fig. 7 is the structure chart of electronic equipment in the embodiment of the present application six.

Embodiment

The embodiment of the present invention provides a kind of method and electronic equipment of speech recognition, for solving existing for prior art not The technical problem of energy consumption can be reduced while recognition accuracy is ensured.

When a voice messaging to be identified is identified, it is first determined the present mode of electronic equipment, such as：For power consumption Use pattern, power consumption more than the first predetermined threshold value are less than use pattern of the first predetermined threshold value etc.；It is then based on current mould The difference of formula, different sound identification modules is determined, such as：When first use pattern is that power consumption is more than the first predetermined threshold value Use pattern when, it is determined that sound identification module corresponding with first use pattern is software identification module, in this feelings Under condition, illustrating that electronic equipment is in normal operating conditions, the phonetic order of its user is more, so entered using software identification module Row identification, to improve recognition accuracy；When first use pattern is that power consumption uses mould less than first predetermined threshold value During formula, it is determined that sound identification module corresponding with first use pattern is hardware identification module, in this case, electronics Equipment is likely to be at standby mode/park mode etc., and in this case, electronic equipment, which needs only to perform some wake-ups, to be referred to Order, so, vocabulary corresponding to it simultaneously need not be too many, so it is identified using hardware identification module, to save energy Consumption.

Due to when voice being identified using such scheme, the less use mould of control instruction is in electronic equipment Under formula, the relatively low speech recognition system of energy consumption can be used, is only in the more use pattern of control instruction in electronic equipment Under, just using the higher speech recognition system of energy consumption, so, having reached reduces energy consumption in the case where ensureing recognition accuracy Technique effect.

Embodiment one

The embodiment of the present application one provides a kind of method of speech recognition, and using an electronic equipment, the electronic equipment is for example For：Notebook computer, mobile phone, tablet personal computer etc..

The electronic equipment includes at least two different speech recognition systems, wherein, described at least two is different Speech recognition system is, for example,：Hardware sound identification module, software sound identification module；Or at least two difference Speech recognition system be：Small vocabulary speech recognition system, major term table speech recognition system etc., for described at least two not Which kind of speech recognition system is same speech recognition system be specially, and the embodiment of the present application is not restricted.

Fig. 1 is refer to, the method for the speech recognition comprises the following steps：

Step S101：Obtain the first voice messaging to be identified；

Step S102：The present mode for obtaining the electronic equipment is the first use pattern；

Step S103：Based on corresponding relation between use pattern and sound identification module, it is determined that with first use pattern Corresponding sound identification module is the first sound identification module in described at least two different sound identification modules；

Step S104：The voice messaging to be identified is identified by first sound identification module, obtains one Recognition result.

In specific implementation process, voice messaging or step S102 to be identified are obtained based on step S101 and obtain the electronics Before the present mode of equipment is the first use pattern, such as Fig. 2, shown methods described also comprises the following steps：

Step S201：When the electronic equipment is in the second use pattern, detection obtains the second voice messaging to be identified；

Step S202：By from second use pattern corresponding in described at least two different sound identification modules The second sound identification module the described second voice messaging to be identified is identified, to determine whether there is the control electronics Equipment enters the first control instruction of first use pattern；

Step S203：When first control instruction be present, the electronic equipment is controlled to use mould into described first Formula.

In specific implementation process, the second use pattern in step S201, it is, for example,：Shutdown mode, standby mode, In this case, second sound identification module is, for example,：Hardware identification module, wherein, closed when the electronic equipment is in When machine pattern or standby mode, the microphone in the hardware identification module and electronic equipment is all still in working condition, The voice messaging that is inputted of user is gathered by the microphone can and is stored in the content of electronic equipment.

Under normal conditions, the different phonetic identification module of electronic equipment can share microphone, that is to say, that no matter electricity Which kind of sound identification module is sub- equipment use, and can use same microphone collection voice messaging, and share microphone Caching, that is to say, that when electronic equipment uses the second sound identification module, the storage location of its voice messaging gathered With when the electronic equipment uses the first sound identification module, the storage location of its voice messaging gathered is identical.

After second voice messaging to be identified is obtained based on step S201, it is possible to based on step S202, also It is that the described second voice messaging to be identified is identified by second sound identification module, and then it is simple to obtain some Control instruction, such as：Start-up command, wake up instruction etc..

So after the microphone of the electronic equipment detects voice messaging, just pass through the hardware identification engine The voice messaging is identified, to determine whether there is the voice messaging related to starting shooting or waking up, correlation be present Voice messaging, just produce a start-up command or wake up instruction, to control the electronic equipment to enter the first use pattern, such as： Use pattern corresponding to open state, wherein, such as：Normal use pattern, power saving use pattern etc..

As seen from the above description, due in the embodiment of the present application, when electronic equipment is in the second use pattern, such as： During shutdown/standby mode, can by the second sound identification module, such as：Hardware identification module is known to voice messaging Not, to control the electronic equipment to enter first use pattern, because hardware identification module need not open electronic equipment Speech recognition can be achieved, so reached the technique effect for saving energy consumption.

First control of the control electronic equipment into the first use pattern is being determined whether there is based on step S202 After instruction, methods described also includes：

In specific implementation process, when user controls the electronic equipment from second use pattern by voice messaging Switch to after first use pattern, it will usually it is desirable that electronic equipment performs certain operations, such as:Open text editing Device, open music player, open web browser etc..

So user may can also produce some other voice messagings after by voice messaging control start, such as： " opening text editor ", " opening music player ", " unlatching web browser " etc., wherein, due to second voice Identifying system, such as：Hardware sound identification module, comprising vocabulary it is less, may only contain some control start, The vocabulary of wake-up, and for vocabulary caused by above-mentioned user, it can not identify, so the hardware identification module is detecting To during the start-up command/wake-up instruction, it is possible to record the first voice messaging that user is inputted.

In specific implementation process, after first voice messaging is obtained, first voice messaging can be deposited The internal memory of the electronic equipment is stored in, it is of course also possible to be stored in other positions, which kind of holder be stored in for described, this Application embodiment is not restricted.

As seen from the above description, due in the embodiment of the present application, controlling the electronic equipment to make into first existing During with the first control instruction of pattern, the first voice messaging that record user is inputted so uses the first language in electronic equipment Sound identifying system, such as：During software sound identification module, first language can be called by the software sound identification module Message ceases and then identified, and for software sound identification module is compared to hardware sound identification module, its vocabulary is more, so The technique effect for improving recognition accuracy is reached.

Wherein, in step S101, a variety of situations may be divided into by obtaining the first voice messaging to be identified, be set forth below therein Two kinds are introduced, and certainly, in specific implementation process, are not limited to following two situations.

The first, specifically comprises the following steps：

First voice messaging is read from the internal memory；

In specific implementation process, mould is used due to switching to first from the second use pattern in the control electronic equipment During formula, the voice messaging of recorded user, and be stored in internal memory, so described first can be directly read in internal memory Voice messaging, as the voice messaging to be identified.Wherein, when first voice messaging is stored in into internal memory, Ke Yiji The time end point for the voice messaging that first voice messaging is identified by the second sound identification module is recorded, at this Voice messaging after time point then the first sub- voice messaging not identified by second sound identification module.

Second, directly pass through the voice messaging of microphone user in real.

In specific implementation process, simultaneously non-user is used to control the electronic equipment possible first voice messaging The voice messaging of system, or first voice messaging is not recorded, so real-time voice is directly caught by microphone Information is controlled to the electronic equipment.

Wherein, after the electronic equipment enters first use pattern, the electronic equipment can produce one and carry Show information, for prompting to be used to not currently exist the voice messaging for being controlled electronic equipment, and determine the need for reality Apply the voice messaging of record user.

For example, produce following prompt message:

" voice messaging being controlled to electronic equipment is there is currently no, corresponding voice messaging please be input with to electricity Sub- equipment is controlled.”

Certainly, in specific implementation process, the prompt message is not limited to above-mentioned prompt message, believes for the prompting Cease and be not restricted for which kind of prompt message, the embodiment of the present application.

Wherein, in step S102, various ways can be used to determine the first current use pattern of the electronic equipment, Two ways therein is set forth below to be introduced, certainly, in specific implementation process, is not limited to following two situations.

The first, determines the energy consumption of the electronic equipment, such as：If the energy consumption of the electronic equipment is less than a default threshold Value, it is determined that the use pattern of the electronic equipment is power saving use pattern；And if the energy consumption of the electronic equipment is more than institute State predetermined threshold value, it is determined that the use pattern of the electronic equipment is normal use pattern etc..

Second, determine to be in the application program of running status on the electronic equipment, such as：If the application program For simple games, image processing program, it is determined that control it required managing terminology less, thus can determine It is small vocabulary use pattern；And when the application program is complicated image processing program, text editor, it is determined that right Vocabulary required for it is controlled is more, thus determines that it is big vocabulary use pattern etc..

, can be in the correspondence that application program and speech recognition system are set in the electronic equipment in specific implementation process Relation table, the speech recognition system corresponding to application program is determined based on mapping table；The either electronic equipment is direct The vocabulary used in the application program is detected, is then based on different vocabularies, is determined corresponding to different application programs Different phonetic identifying system.

In specific implementation process, based on step S102 control the electronic equipment enter first use pattern it Afterwards, as shown in figure 3, methods described also comprises the following steps：

Step S301：Judge whether to open the second control instruction of first sound identification module；

Step S302：After it second control instruction be present, step is performed：S103.

In specific implementation process, above-mentioned second control instruction can be obtained using any-mode, such as：When described When second speech recognition system is in opening, the voice messaging of user is caught by second speech recognition system and divided Analysis, when it is determined that including the result of voice analysis for controlling first sound identification module in voice messaging, just perform step Rapid S103；Or second control instruction can be triggered by way of responding user and clicking on a certain specific button；And or Person can trigger described second control instruction etc. by responding slide of the user on touch screen.For being embodied During, which kind of mode is second control instruction is triggered using, the embodiment of the present application is not restricted.

As seen from the above description, due in the embodiment of the present application, mould being used into described first in control electronics After formula, first have to judge whether the second control instruction for opening the first sound identification module, institute is just opened in the presence of The first sound identification module is stated, so reach the technique effect for saving energy consumption.

Wherein, it is described to be based on corresponding relation between use pattern and sound identification module in step S103, it is determined that with described the Sound identification module corresponding to one use pattern is the first sound identification module, is specifically included：

In specific implementation process, when the electronic equipment enters first use pattern, such as：Open state institute is right After the use pattern answered, wherein, based on the different use patterns corresponding to open state, different voices may be corresponded to and known Other system, it is assumed that there are following mapping table：

Wherein, because the electronic equipment uses power saving use pattern, it can be seen that it is desirable that power consumption is saved, so using word The less sound identification module of remittance amount, such as：The software identification module of hardware sound identification module either small vocabulary, thus save Save energy consumption；And when the electronic equipment uses normal mode of operation, it is desirable to be able to identification it is more accurate, so using vocabulary More sound identification module, such as：The software sound identification module relative with hardware sound identification module and small vocabulary software Identification module relative to big vocabulary software identification module.

In addition, in specific implementation process, can also the difference based on application program used in the electronic equipment, come Determine its corresponding first speech recognition system.

Such as：When application program used in the electronic equipment is relatively simple application program, that is, the electricity When the use pattern of sub- equipment is small vocabulary use pattern, its corresponding control instruction is less and relatively simple, such as：The One application program is Photo Browser, and its corresponding control instruction is less, such as：" upper one ", " next ", " diminution ", " put Greatly " etc., in this case, using hardware sound identification module or the sound identification module of small vocabulary.

And when first application program is complex application program, such as：At text input application program, image Application program etc. is managed, because its corresponding control instruction is more, that is, the use pattern of the electronic equipment is big vocabulary During use pattern, using the more sound identification module of vocabulary, such as：The software voice relative with hardware sound identification module Identification module, with small vocabulary software identification module relative to big vocabulary software identification module etc., to reach more accurate Identification.

In specific implementation process, the step S101 and step S102, step S103, priority is not suitable for its execution sequence Point of sequence, can be first carried out with step S101, can also step S102 and step S103 first carry out, or both while perform, it is right This embodiment of the present application is not restricted.

In specific implementation process, based on step S103, it is determined that speech recognition corresponding with first use pattern Module is after the first sound identification module, methods described also includes：

Close second sound identification module.

In specific implementation process, if determining that first speech recognition system is known for software voice in step S103 If other module, because the precision of its identification is higher, second sound identification module i.e. hardware voice are not needed The auxiliary of identification module, it can also complete accurately to identify, so in this case, close the second speech recognition mould Block.

As seen from the above description, due in the embodiment of the present application, while first sound identification module is opened, Second sound identification module is closed, so reached the technique effect of further reduction energy consumption.

, can when the voice messaging to be identified being identified by first sound identification module in step S104 To obtain a variety of recognition results, the recognition result is, for example, the control instruction to application program, such as：Open PS image procossings Software, open music player；In another example be the operation to file, such as：It is named as on opening C disks file, opening desktop The WORD documents of " application documents " etc.；In another example some settings to electronic equipment, such as：Close volume control button, more Desktop wallpaper etc. is changed, is which kind of result for the recognition result, the embodiment of the present application is not restricted.

Embodiment two

In order that those skilled in the art can connect the side of the speech recognition described in the embodiment of the present application one The specific implementation process of method, in the present embodiment, station is introduced to the specific implementation of the method for the information processing in user side detailed annotation Process.

In the present embodiment, using the electronic equipment as notebook computer, first sound identification module is：Hardware language Sound identification module or software sound identification module, second sound identification module be hardware sound identification module exemplified by be situated between Continue.

At the T1 moment, the notebook computer is in off-mode, wherein, the microphone and hardware of the notebook computer Speech recognition system is in opening.

At the T2 moment, user A wishes to open the notebook computer and carries out some image procossings, so, user A is produced Following voice messaging " start "；

The processor detection of notebook computer is identified to it, started shooting with after the above-mentioned voice messaging of user Control instruction, so control the notebook computer to boot up state；The processor of the notebook computer starts to remember simultaneously The voice messaging of the user A is recorded, " it is elegant to open U.S. figure wherein recorded following first voice messaging, and opens and is deposited on desktop Store up the jpg files of entitled ' the bright moon during Qin ' ", and first voice messaging is stored in internal memory.

At the T3 moment, the notebook computer enters opening, also, due to the user using the notebook computer And its mode of operation is not provided with, so it is normal use pattern, the notebook computer is it is determined that the first use pattern is just After normal use pattern, the software speech recognition system is switched on, simultaneously closes off the hardware speech recognition system.

Meanwhile first voice messaging is identified the notebook computer, obtains following phonetic control command：

" open U.S. figure elegant elegant ", " bright moon .jpg " when opening the Qin

So the U.S. elegant program of figure of the processor control of the notebook computer is in open mode, elegant in the U.S. figure of control Elegant program is in after open mode, the bright moon .jpg when opening the Qin during U.S. schemes elegant program.

Embodiment three

The present embodiment will be using the electronic equipment as tablet personal computer, and first speech recognition system is：Small vocabulary software It is introduced exemplified by identifying system or big vocabulary software identifying system.

At the T4 moment, user B plays the game of Tetris using the tablet personal computer, wherein, Tetris only wraps Include four control instructions such as "left", "right" " acceleration " " rotation "；

The processor of the tablet personal computer is it is determined that the application program run on the tablet personal computer is Russia side After block, it is small vocabulary use pattern to determine the tablet personal computer, thus opens the small vocabulary software identification of the tablet personal computer System, the control instruction of user to be identified.

At the T5 moment, user B closes Russia Bricks Game, and opens WORD and start to input document, the tablet personal computer After detecting that user opens WORD documents, determine that its control instruction is more, so big vocabulary use pattern is determined that it is, thus User B voice messaging is identified the big vocabulary software identifying system for opening the tablet personal computer.

Example IV

Based on same inventive concept, the embodiment of the present application four provides a kind of method of information processing, is set applied to an electronics In standby, the electronic equipment is, for example,：Tablet personal computer, mobile phone, notebook computer etc..

Fig. 4 is refer to, the method for described information processing comprises the following steps：

Step S401：When the electronic equipment is in the second use pattern, detection obtains the second voice messaging to be identified, Wherein, second use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

Step S402：By waiting to know to described second with the second sound identification module corresponding to second use pattern Other voice messaging identification, it is determined whether the first control instruction for controlling the electronic equipment to enter the first use pattern be present；

Step S403：When first control instruction be present, the electronic equipment is controlled to use mould into described first Formula, wherein, first use pattern is that the electronic equipment is in use pattern corresponding during normal operating conditions.

Wherein, in step S401, second use pattern can the electronic equipment be in corresponding during power saving shape Any use pattern, such as：Shutdown mode, stand-by/sleep mode etc., the embodiment of the present application is not restricted.

In this case, the microphone of the electronic equipment is still within working condition, can record user institute in real time Caused voice messaging.

Wherein, in step S402, second sound identification module is, for example,：Hardware identification module, when the electronics is set During in shutdown mode, standby mode or standby mode, the hardware identification module is still within working condition, passes through institute Stating hardware identification module, just the second voice messaging acquired in microphone can be identified, and then obtain some simple controls System instruction, such as：Start-up command, wake up instruction etc..

Wherein, in step S403, first use pattern is right when being in normal operating conditions for the electronic equipment Any use pattern answered.

In specific implementation process, based on controlled in step S403 the electronic equipment enter first use pattern it Afterwards, Fig. 5 is refer to, methods described also comprises the following steps：

S501：Determine the first sound identification module corresponding to first use pattern；

S502：Obtain the first voice messaging to be identified；

S503：The described first voice messaging to be identified is identified based on first sound identification module, to obtain An at least recognition result.

Wherein, in step S501, first use pattern due to for normal work when corresponding use pattern, so First sound identification module is usually software identification module, certainly, in specific implementation process, can be cut in the electronic equipment Shift to after first use pattern, the electronic equipment just automatically turns on first sound identification module；Can also be After the operation for receiving user, control instruction corresponding to generation, and then control the electronic equipment to open first voice Identification module, first sound identification module is opened to this under what circumstances, and the embodiment of the present application is not restricted.

In specific implementation process, in step S502, first voice messaging to be identified can be obtained using various ways , it is set forth below two kinds therein and is introduced.

The first, refers in first control for determining whether there is the control electronic equipment into the first use pattern After order, methods described also includes：

In this case, the first voice messaging to be identified of the acquisition, it is specially：

In specific implementation process, after the first voice messaging of microphone record user, electricity can be stored in The internal memory of sub- equipment, can be directly in internal memory for first voice messaging after the first sound identification module is opened.Its In, when first voice messaging is stored in into internal memory, can record by the second sound identification module to first language The time end point for the voice messaging that message breath is identified, the voice messaging after this time point is not then to be described The first sub- voice messaging that second sound identification module is identified.

Second, after it first control instruction be present, first voice messaging is not recorded, so, After starting first sound identification module, pass through the voice messaging of microphone user in real.

In specific implementation process, simultaneously non-user is used to control the electronic equipment possible first voice messaging The voice messaging of system, or first voice messaging is not recorded, used so directly catching real-time capture by microphone The voice messaging at family is as the described first voice messaging to be identified.

Embodiment five

Based on same inventive concept, the embodiment of the present application four provides a kind of electronic equipment, and the electronic equipment is, for example,：Pen Remember this computer, mobile phone, tablet personal computer etc..

Include at least two different sound identification modules in the electronic equipment, it is described in specific implementation process At least two different speech recognition systems are, for example,：Hardware sound identification module, software sound identification module；Or institute Stating at least two different speech recognition systems is：Small vocabulary speech recognition system, major term table speech recognition system etc., for Which kind of speech recognition system is described at least two different speech recognition systems be specially, and the embodiment of the present application is not restricted.

Fig. 6 is refer to, the electronic equipment also includes following structure：

First obtains chip 601, for obtaining a voice messaging to be identified；

Second obtains chip 602, and the present mode for obtaining the electronic equipment is the first use pattern；

Chip 603 is determined, for based on corresponding relation between use pattern and sound identification module, it is determined that making with described first The sound identification module corresponding to pattern is the first sound identification module in described at least two different sound identification modules；

3rd obtains chip 604, for being carried out by first sound identification module to the voice messaging to be identified Identification, obtain a recognition result.

In specific implementation process, the electronic equipment also includes following structure：

Detection chip, for before being the first use pattern in the present mode for obtaining the electronic equipment, when the electricity When sub- equipment is in the second use pattern, by from second use pattern corresponding to described at least two different voices know The second sound identification module in other module, which detects whether to exist, controls the electronic equipment to enter first use pattern First control instruction；

Control chip, make for when first control instruction be present, controlling the electronic equipment to enter described first Use pattern.

In specific implementation process, second use pattern, it is, for example,：Shutdown mode, standby mode etc., this In the case of, second sound identification module is, for example,：Hardware identification module.Wherein, when the electronic equipment is in shutdown mould When formula or standby mode, the microphone in the hardware identification module and electronic equipment is all still in working condition, passes through The hardware identification module can identify some simple voice messagings, and then produce some simple control instructions, so as to To detect the voice messaging of user, to determine whether to control the electronic equipment to enter first use pattern.

As seen from the above description, due in the embodiment of the present application, when electronic equipment is in the second use pattern, such as： During shutdown/standby mode, can by the second sound identification module, such as：Hardware identification module is known to control instruction Not, to control the electronic equipment to enter first use pattern, because hardware identification module need not open electronic equipment Speech recognition can be achieved, so reached the technique effect for saving energy consumption.

In specific implementation process, the electronic equipment also includes：

Memorization COMS clip, for after controlling the control electronic equipment to enter first use pattern, recording The electronic equipment from second use pattern switch to first use pattern when user inputted the first voice letter Breath, first voice messaging is the voice messaging to be identified.

In specific implementation process, the electronic equipment also includes：

Storage chip, for using mould recording the electronic equipment and switch to described first from second use pattern After the first voice messaging that user is inputted during formula, first voice messaging is stored in the internal memory of the electronic equipment.

As seen from the above description, due in the embodiment of the present application, making controlling the electronics to set into described first During with pattern, the first voice messaging that user is inputted is recorded, so uses the first speech recognition system in electronic equipment, than Such as：During software sound identification module, first voice messaging can be called by the software sound identification module and then knowledge Not, and for software sound identification module is compared to hardware sound identification module, its vocabulary is more, knows so having reached raising The technique effect of other accuracy rate.

In specific implementation process, described first obtains chip 601, various ways can be used to obtain first voice Information, it is set forth below two kinds therein and is introduced, certainly, in specific implementation process, is not limited to following two situations.

The first, described first obtains chip 601, is specifically used for：

First voice messaging is read from the internal memory.

Second, described first obtains chip 601, is specifically used for：

Directly pass through the voice messaging of microphone user in real.

In specific implementation process, the electronic equipment also includes：

In specific implementation process, the determination chip 604, specifically include：

In specific implementation process, the electronic equipment also includes：

By the electronic equipment that the present embodiment five is introduced, to implement the speech recognition in the embodiment of the present application one to three Electronic equipment used by method, so the method based on the speech recognition described in the embodiment of the present application one to three, ability Domain those of skill in the art can understand the embodiment of electronic equipment and its various change shape in the embodiment of the present application five Formula, so being no longer discussed in detail for the electronic equipment herein.As long as those skilled in the art implement the embodiment of the present application Electronic equipment used by the method for speech recognition in one to three, belongs to the scope to be protected of the application.

Embodiment six

Based on same inventive concept, the embodiment of the present application six provides a kind of electronic equipment, and the electronic equipment is, for example,：Pen Remember this computer, tablet personal computer, mobile phone etc..

Fig. 7 is refer to, the electronic equipment includes following structure；

Second detection chip 701, for when the electronic equipment is in the second use pattern, detection to obtain second and waits to know Other voice messaging, wherein, second use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

Second determines chip 702, for passing through the second sound identification module pair corresponding to second use pattern The second voice messaging identification to be identified, it is determined whether the control electronic equipment be present and enter the first of the first use pattern Control instruction；

Second control chip 703, for when first control instruction be present, controlling described in the electronic equipment entrance First use pattern, wherein, first use pattern is corresponding when the electronic equipment is in normal operating conditions makes Use pattern.

In specific implementation process, the electronic equipment also includes：

Chip is obtained, for obtaining the first voice messaging to be identified；

In specific implementation process, the acquisition chip can use various ways to obtain the described first voice letter to be identified Breath, is set forth below two kinds therein and is introduced, certainly, in specific implementation process, be not limited to following two situations.

The first, the electronic equipment also includes：

In this case, the acquisition chip, is specifically used for：

Second, the acquisition chip, it is specifically used for：

Described first voice messaging to be identified is used as by the voice messaging of microphone real-time capture user.

By the electronic equipment that the present embodiment six is introduced, to implement the method for the information processing in the embodiment of the present application four Used electronic equipment, so the method based on the information processing described in the embodiment of the present application four, the affiliated skill in this area Art personnel can understand the embodiment of electronic equipment and its various change form in the embodiment of the present application six, so This is no longer discussed in detail for the electronic equipment.As long as those skilled in the art implement in the embodiment of the present application four at information Electronic equipment used by the method for reason, belong to the scope to be protected of the application.

One or more technical schemes that the application provides, have at least the following technical effects or advantages：

(1) at least two different speech recognition moulds are set due in the embodiment of the present application, employing electron equipment Block, so as to the different use patterns based on electronic equipment, the technical scheme of different sound identification modules can be used, so, In the case where electronic equipment is in the less use pattern of control instruction, the relatively low speech recognition system of energy consumption can be used, is only existed Electronic equipment is under the more use pattern of control instruction, just using the higher speech recognition system of energy consumption, so solve The technical problem of recognition accuracy can not be improved while energy consumption is reduced in the prior art, has reached and has ensured recognition accuracy In the case of reduce energy consumption technique effect.

(3) due in the embodiment of the present application, when controlling the electronics to set into first use pattern, recording The first voice messaging that user is inputted, the first speech recognition system so is used in electronic equipment, such as：Software speech recognition During module, first voice messaging can be called by the software sound identification module and then identification, and software voice is known For other module is compared to hardware sound identification module, its vocabulary is more, so reached the technology for improving recognition accuracy Effect.

Although preferred embodiments of the present invention have been described, but those skilled in the art once know basic creation Property concept, then can make other change and modification to these embodiments.So appended claims be intended to be construed to include it is excellent Select embodiment and fall into having altered and changing for the scope of the invention.

Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to comprising including these changes and modification.

Claims

1. a kind of audio recognition method, applied in an electronic equipment, it is characterised in that include at least in the electronic equipment Two different sound identification modules, methods described include：

By from second use pattern corresponding to the second voice in described at least two different sound identification modules know Described second voice messaging to be identified is identified other module, enters first to determine whether there is the control electronic equipment First control instruction of use pattern；

When first control instruction be present, the electronic equipment is controlled to enter first use pattern；

Obtain the first voice messaging to be identified；

Based on corresponding relation between use pattern and sound identification module, it is determined that speech recognition corresponding with first use pattern Module is the first sound identification module in described at least two different sound identification modules；

The described first voice messaging to be identified is identified by first sound identification module, obtains a recognition result.

2. the method as described in claim 1, it is characterised in that enter first determining whether there is the control electronic equipment After first control instruction of use pattern, methods described also includes：

When first control instruction be present, it is recorded in that to produce the first moment of first control instruction described to switching to The first voice messaging that user inputs between second moment of the first use pattern.

3. method as claimed in claim 2, it is characterised in that produce the first of first control instruction in described be recorded in After the first voice messaging that user inputs between moment to the second moment for switching to first use pattern, methods described Also include：

4. method as claimed in claim 3, it is characterised in that the first voice messaging to be identified of the acquisition, specifically include：

First voice messaging is read from the internal memory；

Obtain the first sub- language for not yet being carried out speech recognition in first voice messaging by second sound identification module Message ceases, and the first sub- voice messaging is the described first voice messaging to be identified.

5. the method as described in claim 1, it is characterised in that enter described first in the control electronic equipment and use After pattern, methods described also includes：

After it second control instruction be present, step is performed：Based on corresponding relation between use pattern and sound identification module, It is determined that sound identification module corresponding with first use pattern is the first sound identification module.

6. the method as described in claim 1, it is characterised in that determine voice corresponding with first use pattern described Identification module is after the first sound identification module, methods described also includes：

Close second sound identification module.

7. the method as described in any claims of claim 1-6, it is characterised in that first sound identification module is：Software Sound identification module or hardware sound identification module；Second sound identification module is：Hardware sound identification module；

First use pattern is：Use pattern corresponding to open state；Second use pattern is shutdown/standby mode.

8. the method as described in any claims of claim 1-6, it is characterised in that described to be based on use pattern and speech recognition mould Corresponding relation between block, it is determined that sound identification module corresponding with first use pattern is the first sound identification module, specifically For：

When first use pattern is the use pattern that power consumption is more than the first predetermined threshold value, known based on use pattern and voice Other intermodule corresponding relation, it is determined that sound identification module corresponding with first use pattern is software identification module；

When first use pattern is the use pattern that power consumption is less than first predetermined threshold value, it is determined that making with described first The sound identification module corresponding to pattern is hardware identification module.

A kind of 9. method of information processing, applied in an electronic equipment, it is characterised in that methods described includes：

When the electronic equipment is in the second use pattern, detection obtains the second voice messaging to be identified, wherein, described second Use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

By knowing with the second sound identification module corresponding to second use pattern to the described second voice messaging to be identified Not, it is determined whether the first control instruction for controlling the electronic equipment to enter the first use pattern be present；

When first control instruction be present, the electronic equipment is controlled to enter first use pattern, wherein, described One use pattern is that the electronic equipment is in use pattern corresponding during normal operating conditions.

10. method as claimed in claim 9, it is characterised in that entering described first in the control electronic equipment makes After pattern, methods described also includes：

Obtain the first voice messaging to be identified；

The described first voice messaging to be identified is identified based on first sound identification module, to obtain at least one identification As a result.

11. method as claimed in claim 10, it is characterised in that enter in the control electronic equipment that determines whether there is After the first control instruction for entering the first use pattern, methods described also includes：

12. method as claimed in claim 11, it is characterised in that the first voice messaging to be identified of the acquisition, be specially：

13. a kind of electronic equipment, it is characterised in that include at least two different speech recognition moulds in the electronic equipment Block, the electronic equipment also include：

First obtains chip, for obtaining the first voice messaging to be identified；

First determines chip, for based on corresponding relation between use pattern and sound identification module, it is determined that being used with described first Sound identification module corresponding to pattern is the first sound identification module in described at least two different sound identification modules；

3rd obtains chip, for being known by first sound identification module to the described first voice messaging to be identified Not, a recognition result is obtained；

The electronic equipment also includes：

First detection chip, for before being the first use pattern in the present mode for obtaining the electronic equipment, when the electricity When sub- equipment is in the second use pattern, by from second use pattern corresponding to described at least two different voices know The second sound identification module in other module, which detects whether to exist, controls the electronic equipment to enter first use pattern First control instruction；

First control chip, make for when first control instruction be present, controlling the electronic equipment to enter described first Use pattern.

14. electronic equipment as claimed in claim 13, it is characterised in that the electronic equipment also includes：

First memorization COMS clip, for determining whether there is first control of the control electronic equipment into the first use pattern After instruction, when first control instruction be present, the first moment extremely switching for producing first control instruction is recorded in To the first voice messaging that user inputs between the second moment of first use pattern.

15. electronic equipment as claimed in claim 14, it is characterised in that the electronic equipment also includes：

Storage chip, for using mould to switching to described first being recorded in the first moment for producing first control instruction After the first voice messaging that user inputs between second moment of formula, first voice messaging is stored in the electronics and set Standby internal memory.

16. electronic equipment as claimed in claim 15, it is characterised in that described first obtains chip, specifically includes：

Sub- chip is obtained, voice knowledge is not yet carried out by second sound identification module for obtaining in first voice messaging The the first sub- voice messaging do not crossed, the first sub- voice messaging are the described first voice messaging to be identified.

17. electronic equipment as claimed in claim 13, it is characterised in that the electronic equipment also includes：

Judge chip：For after controlling the electronic equipment to enter first use pattern, judging whether to open Second control instruction of first sound identification module；And

18. electronic equipment as claimed in claim 13, it is characterised in that the electronic equipment also includes：

Chip is closed, for it is determined that sound identification module corresponding with first use pattern is the first sound identification module Afterwards, second sound identification module is closed.

19. the electronic equipment as described in any claims of claim 13-18, it is characterised in that first sound identification module For：Software sound identification module or hardware sound identification module；Second sound identification module is：Hardware speech recognition mould Block；

20. the electronic equipment as described in any claims of claim 13-18, it is characterised in that described first determines chip, specifically Including：

First true stator chip, for when first use pattern be power consumption be more than the first predetermined threshold value use pattern when, Based on corresponding relation between use pattern and sound identification module, it is determined that sound identification module corresponding with first use pattern For software identification module；

Second true stator chip, for being the use pattern that power consumption is less than first predetermined threshold value when first use pattern When, it is determined that sound identification module corresponding with first use pattern is hardware identification module.

21. a kind of electronic equipment, it is characterised in that the electronic equipment includes：

Second detection chip, for when the electronic equipment is in the second use pattern, detection to obtain the second voice to be identified Information, wherein, second use pattern is that the electronic equipment is in use pattern corresponding during power down mode；

Second determine chip, for by with the second sound identification module corresponding to second use pattern to described second Voice messaging identification to be identified, it is determined whether the first control that the control electronic equipment into the first use pattern be present refers to Order；

Second control chip, make for when first control instruction be present, controlling the electronic equipment to enter described first With pattern, wherein, first use pattern is that the electronic equipment is in use pattern corresponding during normal operating conditions.

22. electronic equipment as claimed in claim 21, it is characterised in that the electronic equipment also includes：

3rd determines chip, for after controlling the electronic equipment to enter first use pattern, determining described first The first sound identification module corresponding to use pattern；

Chip is obtained, for obtaining the first voice messaging to be identified；

4th obtains chip, for being known based on first sound identification module to the described first voice messaging to be identified Not, to obtain an at least recognition result.

23. electronic equipment as claimed in claim 22, it is characterised in that the electronic equipment also includes：

Second memorization COMS clip, for determining whether there is first control of the control electronic equipment into the first use pattern After instruction, when first control instruction be present, the first moment extremely switching for producing first control instruction is recorded in To the first voice messaging that user inputs between the second moment of first use pattern.

24. electronic equipment as claimed in claim 23, it is characterised in that the acquisition chip, be specifically used for：