CN110427460A - Method and device for interactive information - Google Patents
Method and device for interactive information Download PDFInfo
- Publication number
- CN110427460A CN110427460A CN201910720051.4A CN201910720051A CN110427460A CN 110427460 A CN110427460 A CN 110427460A CN 201910720051 A CN201910720051 A CN 201910720051A CN 110427460 A CN110427460 A CN 110427460A
- Authority
- CN
- China
- Prior art keywords
- word
- entry
- interrupt
- processed
- interrupts
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Acoustics & Sound (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
Abstract
Embodiment of the disclosure discloses the method and device for interactive information, is related to field of cloud calculation.One specific embodiment of this method gets voice messaging to be processed when including: in response to playing the first voice messaging, obtain the text information to be processed of above-mentioned voice messaging to be processed;Judge above-mentioned text information to be processed and interrupt the matching relationship of condition, wherein the above-mentioned condition that interrupts is used to indicate above-mentioned first voice messaging of stopping broadcasting;It is matched in response to above-mentioned text information to be processed with the condition of interrupting, stops playing above-mentioned first voice messaging.This embodiment improves the flexibilities that information exchange is carried out with user, improve information exchange efficiency.
Description
Technical field
Embodiment of the disclosure is related to technical field of data processing, and in particular to the method and device for interactive information.
Background technique
Conversational program on intelligent robot or smart machine can in several ways (such as can be voice, text,
Gesture etc.) and user's progress information exchange, to be embodied as the functions such as user query information.On intelligent robot or smart machine
Conversational program, can be with the information of real-time reception user when interacting with user.
Summary of the invention
Embodiment of the disclosure proposes the method and device for interactive information.
In a first aspect, embodiment of the disclosure provides a kind of method for interactive information, this method comprises: in response to
Voice messaging to be processed is got when playing the first voice messaging, obtains the text to be processed letter of above-mentioned voice messaging to be processed
Breath;Judge above-mentioned text information to be processed and to interrupt the matching relationship of condition, wherein it is above-mentioned interrupt condition and be used to indicate stopping broadcast
It puts and states the first voice messaging;It is matched in response to above-mentioned text information to be processed with the condition of interrupting, stops playing above-mentioned first language
Message breath.
In some embodiments, the matching relationship of the above-mentioned text information to be processed of above-mentioned judgement and the condition that interrupts, comprising: obtain
The broadcasting attribute of above-mentioned first voice messaging is taken, when above-mentioned broadcasting attribute includes at least one of the following: that playing duration, broadcasting interrupt
Between;Inquire the acquisition time of above-mentioned voice messaging to be processed;In response to above-mentioned acquisition time in above-mentioned playing duration, and it is above-mentioned
Time of the acquisition time in above-mentioned playing duration is later than above-mentioned broadcasting and interrupts the time, judges above-mentioned text information to be processed and beats
The matching of broken strip part.
In some embodiments, the matching relationship of the above-mentioned text information to be processed of above-mentioned judgement and the condition that interrupts, comprising: will
Above-mentioned text information to be processed is divided at least one entry;Word set is interrupted in response to existing to belong at least one above-mentioned entry
The entry of conjunction judges that above-mentioned text information to be processed is matched with the condition of interrupting, wherein the above-mentioned set of words that interrupts includes at least one
Word is interrupted, the above-mentioned word that interrupts is used to indicate above-mentioned first voice messaging of stopping broadcasting.
In some embodiments, the matching relationship of the above-mentioned text information to be processed of above-mentioned judgement and the condition that interrupts, comprising: will
Above-mentioned text information to be processed is divided at least one entry;For the entry at least one above-mentioned entry, which is led
Enter to interrupt word judgment models, obtain corresponding to the entry interrupts judgement information, wherein the above-mentioned word identification model that interrupts is for obtaining
The correspondence entry interrupts judgement information, and above-mentioned interrupt judges that information is interrupted including presence and there is no interrupt;In response to above-mentioned
It interrupts and judges that information interrupts to exist, judge that above-mentioned text information to be processed is matched with the condition of interrupting.
In some embodiments, the above-mentioned word judgment models that interrupt are obtained by following steps training: obtaining multiple sample words
Sample corresponding to each sample entry in item and above-mentioned multiple sample entries interrupts judgement information;By above-mentioned multiple sample words
Each sample entry in item interrupts sample corresponding to each sample entry in above-mentioned multiple sample entries as input
Information is judged as output, and training obtains interrupting word judgment models.
In some embodiments, above-mentioned each sample entry using in above-mentioned multiple sample entries, will be above-mentioned as input
Sample corresponding to each sample entry in multiple sample entries, which interrupts, judges information as output, and training, which obtains interrupting word, to be sentenced
Disconnected model, comprising: execute following training step: each sample entry in above-mentioned multiple sample entries is sequentially input to initial
Word judgment models are interrupted, prediction corresponding to each sample entry in above-mentioned multiple sample entries is obtained and interrupts judgement information,
Prediction corresponding to each sample entry in above-mentioned multiple sample entries is interrupted and is judged corresponding to information and the sample entry
Sample interrupt and judge that information is compared, obtain the above-mentioned predictablity rate for initially interrupting word judgment models, determine above-mentioned pre-
It surveys whether accuracy rate is greater than default accuracy rate threshold value, if more than above-mentioned default accuracy rate threshold value, then initially interrupts word by above-mentioned and sentence
Disconnected model interrupts word judgment models as what training was completed.
In some embodiments, above-mentioned each sample entry using in above-mentioned multiple sample entries, will be above-mentioned as input
Sample corresponding to each sample entry in multiple sample entries, which interrupts, judges information as output, and training, which obtains interrupting word, to be sentenced
Disconnected model, comprising: in response to adjusting the above-mentioned parameter for initially interrupting word judgment models no more than above-mentioned default accuracy rate threshold value,
And continue to execute above-mentioned training step.
In some embodiments, above-mentioned sample entry is obtained by following steps: being obtained at least one and is interrupted word;For upper
It states at least one and interrupts in word and interrupt word, obtain the word of equal value for interrupting word, above-mentioned equivalence word includes at least one of the following: pair
The near synonym of word should be interrupted, corresponding this interrupts the dialect of word;By it is above-mentioned at least one interrupt in word to interrupt word corresponding etc.
Valence word interrupts the sample entry of word labeled as this.
In some embodiments, the above method further include: it is unsatisfactory for interrupting condition in response to above-mentioned text information to be processed,
Continue to play above-mentioned first voice messaging.
In some embodiments, after above-mentioned stopping plays above-mentioned first voice messaging, the above method further include: label is worked as
Preceding interaction mode is state of speaking.
In some embodiments, the above method further include: obtain the second voice letter of corresponding above-mentioned voice messaging to be processed
Breath, and play above-mentioned second voice messaging.
Second aspect, embodiment of the disclosure provide a kind of device for interactive information, which includes: text letter
Acquiring unit is ceased, voice messaging to be processed is got when in response to playing the first voice messaging, being configured to obtain above-mentioned wait locate
Manage the text information to be processed of voice messaging;Matching relationship judging unit, be configured to judge above-mentioned text information to be processed with
Interrupt the matching relationship of condition, wherein the above-mentioned condition that interrupts is used to indicate above-mentioned first voice messaging of stopping broadcasting;First interaction
Control unit matches with the condition of interrupting in response to above-mentioned text information to be processed, is configured to stop to play above-mentioned first voice
Information.
In some embodiments, above-mentioned matching relationship judging unit includes: and plays attribute to obtain subelement, is configured to obtain
The broadcasting attribute of above-mentioned first voice messaging is taken, when above-mentioned broadcasting attribute includes at least one of the following: that playing duration, broadcasting interrupt
Between;Acquisition time obtains subelement, is configured to inquire the acquisition time of above-mentioned voice messaging to be processed;First matching relationship is sentenced
Disconnected subelement, in response to above-mentioned acquisition time in above-mentioned playing duration, and above-mentioned acquisition time is in above-mentioned playing duration
Time is later than above-mentioned broadcasting and interrupts the time, is configured to judge that above-mentioned text information to be processed is matched with the condition of interrupting.
In some embodiments, above-mentioned matching relationship judging unit include: the first entry divide subelement, be configured to by
Above-mentioned text information to be processed is divided at least one entry;Second matching relationship judgment sub-unit, in response to above-mentioned at least one
Exist in a entry and belong to the entry for interrupting set of words, be configured to judge above-mentioned text information to be processed and interrupt condition
Match, wherein the above-mentioned set of words that interrupts includes that at least one interrupts word, and the above-mentioned word that interrupts is used to indicate above-mentioned first language of stopping broadcasting
Message breath.
In some embodiments, above-mentioned matching relationship judging unit include: the second entry divide subelement, be configured to by
Above-mentioned text information to be processed is divided at least one entry;Interrupt and judge acquisition of information subelement, for it is above-mentioned at least one
Entry in entry is configured to import the entry and interrupts word judgment models, and obtain corresponding to the entry interrupts judgement information,
Wherein, above-mentioned to interrupt word identification model and interrupt judgement information for obtain the corresponding entry, above-mentioned interrupt judges that information includes
In the presence of interrupting and there is no interrupt;Third matching relationship judgment sub-unit judges that information interrupts to exist in response to above-mentioned interrupt,
It is configured to judge that above-mentioned text information to be processed is matched with the condition of interrupting.
In some embodiments, above-mentioned apparatus further includes interrupting word judgment models training unit, is configured to training and interrupts
Word judgment models, the above-mentioned word judgment models training unit that interrupts includes: sample acquisition subelement, is configured to obtain multiple samples
Sample corresponding to each sample entry in entry and above-mentioned multiple sample entries interrupts judgement information;Interrupt word judgment models
Training subelement, is configured to using each sample entry in above-mentioned multiple sample entries as input, by above-mentioned multiple samples
Sample corresponding to each sample entry in entry, which interrupts, judges information as output, and training obtains interrupting word judgment models.
In some embodiments, the above-mentioned word judgment models training subelement that interrupts includes: to interrupt word judgment models training mould
Block is configured to sequentially input each sample entry in above-mentioned multiple sample entries to initially word judgment models are interrupted, obtain
Judgement information is interrupted to prediction corresponding to each sample entry in above-mentioned multiple sample entries, by above-mentioned multiple sample entries
In each sample entry corresponding to prediction interrupt and judge that sample corresponding to information and the sample entry interrupts judgement information
It is compared, obtains the above-mentioned predictablity rate for initially interrupting word judgment models, it is pre- to determine whether above-mentioned predictablity rate is greater than
If accuracy rate threshold value, if more than above-mentioned default accuracy rate threshold value, then using the above-mentioned word judgment models that initially interrupt as training completion
Interrupt word judgment models.
In some embodiments, the above-mentioned word judgment models training subelement that interrupts includes: parameter adjustment module, in response to not
Greater than above-mentioned default accuracy rate threshold value, be configured to adjust it is above-mentioned initially interrupt the parameter of word judgment models, and return to above-mentioned beat
Hyphenation judgment models training module.
In some embodiments, above-mentioned apparatus further includes sample entry acquiring unit, is configured to obtain sample entry, on
Stating sample entry acquiring unit includes: to interrupt word to obtain subelement, is configured to obtain at least one and interrupts word;Word of equal value obtains
Subelement, for it is above-mentioned at least one interrupt in word and interrupt word, be configured to obtain the word of equal value for interrupting word, above-mentioned equivalence
Word includes at least one of the following: that corresponding this interrupts the near synonym of word, corresponding this interrupts the dialect of word;It is single that sample entry obtains son
Member is configured at least one above-mentioned corresponding word of equal value of word that interrupts interrupted in word interrupting the sample word of word labeled as this
Item.
In some embodiments, above-mentioned apparatus further include: the second interaction control unit is believed in response to above-mentioned text to be processed
Breath is unsatisfactory for interrupting condition, is configured to continue to play above-mentioned first voice messaging.
In some embodiments, after above-mentioned first interaction control unit, above-mentioned apparatus further include: dry state symbol unit,
Being configured to mark current interaction mode is state of speaking.
In some embodiments, above-mentioned apparatus further include: third interaction control unit, be configured to obtain it is corresponding it is above-mentioned to
The second voice messaging of voice messaging is handled, and plays above-mentioned second voice messaging.
The third aspect, embodiment of the disclosure provide a kind of electronic equipment, comprising: one or more processors;Storage
Device is stored thereon with one or more programs, when said one or multiple programs are executed by said one or multiple processors,
So that said one or multiple processors execute the method for interactive information of above-mentioned first aspect.
Fourth aspect, embodiment of the disclosure provide a kind of computer-readable medium, are stored thereon with computer program,
It is characterized in that, the program realizes the method for interactive information of above-mentioned first aspect when being executed by processor.
The method and device for interactive information that embodiment of the disclosure provides, first when playing the first voice messaging
When getting voice messaging to be processed, the text information to be processed of above-mentioned voice messaging to be processed is obtained;Then, judge to be processed
Text information and the matching relationship for interrupting condition;Finally, when text information to be processed is matched with the condition of interrupting, stop playing the
One voice messaging.Technical scheme improves the flexibility that information exchange is carried out with user, improves information exchange efficiency.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the disclosure is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that one embodiment of the disclosure can be applied to exemplary system architecture figure therein;
Fig. 2 is the flow chart according to one embodiment of the method for interactive information of the disclosure;
Fig. 3 is the schematic diagram according to an application scenarios of the method for interactive information of the disclosure;
Fig. 4 is the flow chart according to another embodiment of the method for interactive information of the disclosure;
Fig. 5 is the structural schematic diagram according to one embodiment of the device for interactive information of the disclosure;
Fig. 6 is adapted for the electronic devices structure schematic diagram for realizing embodiment of the disclosure.
Specific embodiment
The disclosure is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can phase
Mutually combination.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can the method for interactive information using embodiment of the disclosure or the dress for interactive information
The exemplary system architecture 100 set.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can interact with terminal device 101,102,103.Terminal device 101,102,103 can pass through network
104 interact with server 105, to receive or send message etc..Various communications can be installed on terminal device 101,102,103
Client application, for example, voice obtain application, speech recognition application, information search application, voice selecting application, voice play answer
With etc..
Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard
When part, can be the various electronic equipments with voice play function and information processing, including but not limited to intelligent robot,
Smart phone, tablet computer, pocket computer on knee and desktop computer etc..When terminal device 101,102,103 is soft
When part, it may be mounted in above-mentioned cited electronic equipment.Its may be implemented into multiple softwares or software module (such as
Distributed Services are provided), single software or software module also may be implemented into, be not specifically limited herein.
Server 105 can be to provide the server of various services, such as to the letter that terminal device 101,102,103 is sent
Cease the server handled.Server can be handled the information received, and processing result is fed back to terminal and is set
It is standby.
It should be noted that the method provided by embodiment of the disclosure for interactive information is generally by terminal device
101, it 102,103 executes, correspondingly, the device for interactive information is generally positioned in terminal device 101,102,103.
It should be noted that server can be hardware, it is also possible to software.When server is hardware, may be implemented
At the distributed server cluster that multiple servers form, individual server also may be implemented into.It, can when server is software
To be implemented as multiple softwares or software module (such as providing Distributed Services), single software or software also may be implemented into
Module is not specifically limited herein.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for interactive information according to the disclosure is shown
200.This for interactive information method the following steps are included:
Voice messaging to be processed is got when step 201, in response to playing the first voice messaging, obtains above-mentioned language to be processed
The text information to be processed of message breath.
In the present embodiment, for the executing subject of the method for interactive information (such as terminal device shown in FIG. 1 101,
102,103) the first voice messaging can be actively played, (such as user's sending can also be can be receiving other signals
Phonetic order) the first voice messaging is played afterwards.Correspondingly, the first voice messaging can be the voice of preset active question
Information is also possible to the voice messaging according to the corresponding answer made after other signals received.
In the prior art, intelligent robot or smart machine be usually after completing the once interaction with user, further according to
The information that user provides is interacted next time.When the conversational program on intelligent robot or smart machine is interacted with user
When, if user has issued information, the conversational program on intelligent robot or smart machine will still continue to the interaction, not right
The information that user issues processes.Therefore, the information exchange of existing intelligent robot or smart machine and user are more mechanical,
Information exchange is inefficient.
The executing subject of the application is available when playing the first voice messaging to arrive voice messaging to be processed.Later, it holds
Row main body can analyze voice messaging to be processed, obtain the text information to be processed for corresponding to voice messaging to be processed.Its
In, text information to be processed can be voice messaging to be processed is directly converted after obtained text information, be also possible to treat
Processing voice messaging, which is denoised, filtered or converted, (such as can be dialect being converted to mandarin, foreign language is converted to Chinese
Deng) after obtained information.
Step 202, judge above-mentioned text information to be processed and interrupt the matching relationship of condition.
After obtaining text information to be processed, in order to judge the pass with the first voice messaging of broadcasting of text information to be processed
System, executing subject may determine that above-mentioned text information to be processed and interrupt the matching relationship of condition.In this way, use can be obtained accurately
The intention at family improves the identification accuracy to voice messaging to be processed.Improve the efficiency of information exchange.Wherein, above-mentioned to interrupt item
Part can serve to indicate that stopping plays above-mentioned first voice messaging.For example, interrupt condition may is that " stopping ", " etc. ".
In some optional implementations of the present embodiment, the above-mentioned above-mentioned text information to be processed of judgement and condition is interrupted
Matching relationship, may comprise steps of:
The first step obtains the broadcasting attribute of above-mentioned first voice messaging.
The voice messaging that executing subject plays is usually pre-set information.These information generally include respective broadcasting
Attribute.Wherein, above-mentioned broadcasting attribute may include at least one of following: playing duration interrupts type, plays and interrupt time etc..
Wherein, playing duration is time span needed for playing the voice messaging;Interrupting type can be whether voice messaging allows
It is interrupted;It plays and interrupts the time is how long to allow to be interrupted after starting to play the voice messaging.For example, a certain voice letter
It is more important to cease A, the playing duration of voice messaging A is 10 seconds, then corresponding broadcasting, which interrupts the time, can be 6 seconds.That is, broadcasting
Allow to be interrupted after putting the 6th second of voice messaging A.For the different phonetic information that executing subject plays, broadcasting interrupts the time can
With difference.
Second step inquires the acquisition time of above-mentioned voice messaging to be processed.
In order to judge whether text information to be processed needs to be considered, executing subject can inquire above-mentioned voice letter to be processed
The acquisition time of breath.Acquisition time, which can be, obtains the time that executing subject has just detected voice messaging to be processed.
Third step, in response to above-mentioned acquisition time in above-mentioned playing duration, and above-mentioned acquisition time is in above-mentioned broadcasting
Time in length is later than above-mentioned broadcasting and interrupts the time, judges that above-mentioned text information to be processed is matched with the condition of interrupting.
When executing subject detects voice messaging to be processed (that is, language to be processed in the playing duration of the first voice messaging
The acquisition time of message breath is in the playing duration of the first voice messaging), and time evening of the acquisition time in playing duration
The time is interrupted in above-mentioned broadcasting, it may be considered that text information to be processed is matched with the condition of interrupting.For example, the first voice messaging
Playing duration is 10 seconds, and broadcasting, which interrupts the time, can be 5 seconds.The acquisition time of voice messaging to be processed in playing duration when
Between be 7 seconds, that is, acquisition time is in the time zone (i.e. the 5th second to the 10th second) that allows to interrupt.Then executing subject can recognize
It is matched for text information to be processed with the condition of interrupting.It is when time of the acquisition time in playing duration of voice messaging to be processed
3 seconds, that is, acquisition time is in the time zone (i.e. the 1st second to the 4th second) for not allowing to interrupt.Then executing subject can consider
Text information to be processed is mismatched with the condition that interrupts.
In some optional implementations of the present embodiment, the above-mentioned above-mentioned text information to be processed of judgement and condition is interrupted
Matching relationship, may comprise steps of:
Above-mentioned text information to be processed is divided at least one entry by the first step.
Executing subject can also be judged text information to be processed by way of entry and interrupt the matching relationship of condition.
For this purpose, text information to be processed can be divided at least one entry by executing subject.
Second step belongs to the entry for interrupting set of words in response to existing at least one above-mentioned entry, judges above-mentioned wait locate
Reason text information is matched with the condition of interrupting.
When existing at least one entry, when belonging to the entry for interrupting set of words, executing subject can consider text to be processed
Information is matched with the condition of interrupting.Wherein, the above-mentioned set of words that interrupts includes that at least one interrupts word.The above-mentioned word that interrupts can be used for referring to
Show that stopping plays above-mentioned first voice messaging.Interrupt word may is that " etc. ", " stopping " etc..In general, interrupting set of words and interrupting
Word is all pre-set, and executing subject can be used directly.
In some optional implementations of the present embodiment, the above-mentioned above-mentioned text information to be processed of judgement and condition is interrupted
Matching relationship, may comprise steps of:
Above-mentioned text information to be processed is divided at least one entry by the first step.
For the matching relationship for judging text information to be processed Yu interrupting condition, executing subject can believe text to be processed
Breath is divided at least one entry.
Entry importing is interrupted word judgment models for the entry at least one above-mentioned entry by second step, is obtained pair
Should entry interrupt judgement information.
Each entry at least one entry can successively be imported and interrupt word judgment models by executing subject, and then be obtained
Corresponding each entry interrupts judgement information.Wherein, the above-mentioned word identification model that interrupts can be used for obtaining beating for the corresponding entry
It is disconnected to judge information.Above-mentioned interrupt judges that information is interrupted including presence and there is no interrupt.Interrupt judge information may also is that " 1 ",
" 0 ", "Yes", "No" etc., no longer repeat one by one herein.Wherein, " 1 " can indicate to interrupt;" 0 " can indicate not interrupt;"Yes"
It can indicate to interrupt;"No" can indicate not interrupt.
Third step judges that information interrupts to exist in response to above-mentioned interrupt, judges above-mentioned text information to be processed and interrupt
Condition matching.
When at least one entry have at least one entry it is corresponding interrupt judge information for exist interrupt when, execute master
Body can consider that text information to be processed is matched with the condition of interrupting.
In some optional implementations of the present embodiment, above-mentioned to interrupt word judgment models trained by following steps
It arrives:
The first step obtains sample corresponding to each sample entry in multiple sample entries and above-mentioned multiple sample entries
Interrupt judgement information.
Corresponding to each sample entry in the available multiple sample entries of executing subject and above-mentioned multiple sample entries
Sample interrupt judgement information.
Second step, using each sample entry in above-mentioned multiple sample entries as input, by above-mentioned multiple sample entries
In each sample entry corresponding to sample interrupt judge information as output, training obtain interrupting word judgment models.
Later, executing subject can be using each sample entry in multiple sample entries as input, by above-mentioned multiple samples
Sample corresponding to each sample entry in this entry, which interrupts, judges information as output, and training obtains interrupting word and judges mould
Type.
In some optional implementations of the present embodiment, the application can also include the training for interrupting word judgment models
Method, comprising:
The first step obtains sample corresponding to each sample entry in multiple sample entries and above-mentioned multiple sample entries
Interrupt judgement information.
In the present embodiment, the available multiple sample entries of executing subject.Those skilled in the art can be rule of thumb
Corresponding sample is marked to each sample entry of multiple sample entries and interrupts judgement information.
Second step sequentially inputs each sample entry in above-mentioned multiple sample entries to initially interrupting word judges mould
Type obtains prediction corresponding to each sample entry in above-mentioned multiple sample entries and interrupts judgement information.
In the present embodiment, multiple sample entries based on acquisition, executing subject can will be every in multiple sample entries
A sample entry is sequentially input to initially word judgment models are interrupted, to obtain each sample entry institute in multiple sample entries
Corresponding prediction interrupts judgement information.Here, executing subject can be by each sample entry from initially interrupting word judgment models
Input side input successively by initially interrupting the processing of the parameter of each layer in word judgment models, and judges from word is initially interrupted
The outlet side of model exports, and the information of outlet side output is that prediction corresponding to the sample entry interrupts judgement information.Wherein,
Initially interrupting word judgment models can be unbred deep learning model (or other learning models) or does not train completion
Deep learning model, each layer are provided with initiation parameter, and initiation parameter can in the training process for interrupting word judgment models
To be continuously adjusted.
Prediction corresponding to each sample entry in above-mentioned multiple sample entries is interrupted and judges information and be somebody's turn to do by third step
Sample corresponding to sample entry, which interrupts, judges that information is compared, and it is accurate to obtain the above-mentioned prediction for initially interrupting word judgment models
Rate.
In the present embodiment, it is interrupted based on prediction corresponding to each sample entry in obtained multiple sample entries
Judge information, executing subject prediction corresponding to each sample entry in multiple sample entries can be interrupted judge information with
Sample corresponding to the sample entry, which interrupts, judges that information is compared, so that the prediction for initially being interrupted word judgment models is quasi-
True rate.Specifically, if prediction corresponding to a sample entry, which interrupts, judges that sample corresponding to information and the sample entry is beaten
It is disconnected to judge that information is identical, then it is correct initially to interrupt the prediction of word judgment models;Sentence if prediction corresponding to a sample entry interrupts
Disconnected information is interrupted from sample corresponding to the sample entry judges that information is different, then initially interrupts word judgment models prediction error.
Here, executing subject can calculate the ratio for predicting correct number and total sample number, and interrupt word using the ratio as initial
The predictablity rate of judgment models.
4th step, determines whether above-mentioned predictablity rate is greater than default accuracy rate threshold value.
In the present embodiment, the predictablity rates of word judgment models is initially interrupted based on obtained, executing subject can be with
The predictablity rate for initially interrupting word judgment models is compared with default accuracy rate threshold value.If more than default accuracy rate threshold
Value then executes the 5th step;If executing the 6th step no more than default accuracy rate threshold value.
5th step, using it is above-mentioned initially interrupt word judgment models as training completion interrupt word judgment models.
In the present embodiment, the case where the predictablity rate for initially interrupting word judgment models is greater than default accuracy rate threshold value
Under, illustrate that this interrupts the training of word judgment models and completes, at this point, executing subject will can initially interrupt word judgment models as training
That completes interrupts word judgment models.
6th step adjusts the above-mentioned parameter for initially interrupting word judgment models.
In the present embodiment, the feelings of default accuracy rate threshold value are not more than in the predictablity rate for initially interrupting word judgment models
Under condition, executing subject is adjustable initially to interrupt the parameter of word judgment models, and returns to execution second step, until training can
Obtain the corresponding entry interrupt judge information interrupt word judgment models until.
In some optional implementations of the present embodiment, above-mentioned sample entry can be obtained by following steps:
The first step obtains at least one and interrupts word.
Executing subject can obtain in advance at least one and interrupt word.Wherein, interrupting word can be technical staff from actual
It is obtained in voice messaging to be processed for interrupting voice messaging.
Second step, for it is above-mentioned at least one interrupt in word and interrupt word, obtain the word of equal value for interrupting word.
Executing subject can inquire the entry same or similar with word meaning is interrupted, and interrupt word using the entry as this
Word of equal value.Wherein, above-mentioned word of equal value may include at least one of following: corresponding this interrupts the near synonym of word, corresponding this interrupts word
Dialect.
Third step, by it is above-mentioned at least one interrupt the corresponding word of equal value of word that interrupts in word and interrupt the sample of word labeled as this
Entry.
After obtaining word of equal value, at least one can be interrupted each of word and interrupt the corresponding word mark of equal value of word by executing subject
It is denoted as the sample entry for interrupting word.
Step 203, it is matched in response to above-mentioned text information to be processed with the condition of interrupting, stops playing above-mentioned first voice letter
Breath.
When text information to be processed is matched with the condition of interrupting, illustrate that the user for issuing voice messaging to be processed wants to interrupt
The first currently playing voice messaging of executing subject.At this point, executing subject can stop playing above-mentioned first voice messaging.Such as
This, improves executing subject and user carries out the flexibility of information exchange, improves information exchange efficiency.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method for interactive information of the present embodiment
Figure.In the application scenarios of Fig. 3, when terminal device 103 (intelligent robot) plays the first voice messaging to user, use is got
The voice messaging to be processed that family issues may is that " there are also the processing of other things for I, next to say again ";Terminal device 103 is got
After the text information to be processed of " there are also the processing of other things for I, next to say again ", terminal device 103 is from text information to be processed
Get the information such as " next time ", " saying again ";Then, terminal device 103 judges text information to be processed according to " next time ", " saying again "
With the matching for the condition that interrupts, and stop play the first voice messaging.
The method provided by the above embodiment of the disclosure gets voice to be processed when playing the first voice messaging first
When information, the text information to be processed of above-mentioned voice messaging to be processed is obtained;Then, judge text information to be processed and interrupt item
The matching relationship of part;Finally, stopping playing the first voice messaging when text information to be processed is matched with the condition of interrupting.This Shen
Please technical solution improve with user carry out information exchange flexibility, improve information exchange efficiency.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method for interactive information.The use
In the process 400 of the method for interactive information, comprising the following steps:
Voice messaging to be processed is got when step 401, in response to playing the first voice messaging, obtains above-mentioned language to be processed
The text information to be processed of message breath.
The content of step 401 and the content of step 201 are identical, no longer repeat one by one herein.
Step 402, judge above-mentioned text information to be processed and interrupt the matching relationship of condition.
The content of step 402 and the content of step 202 are identical, no longer repeat one by one herein.
Step 403, it is matched in response to above-mentioned text information to be processed with the condition of interrupting, stops playing above-mentioned first voice letter
Breath.
The content of step 403 and the content of step 203 are identical, no longer repeat one by one herein.
In some optional implementations of the present embodiment, the above method further include: in response to above-mentioned text to be processed
Information is unsatisfactory for interrupting condition, continues to play above-mentioned first voice messaging.
If text information to be processed is unsatisfactory for interrupting condition, illustrates that voice messaging to be processed does not include and interrupt the first voice
The information that information plays.Voice messaging i.e. to be processed can be user and other people talk or when making a phone call, are performed
The information that main body is got.At this point, executing subject can continue to play above-mentioned first voice messaging.
Step 404, marking current interaction mode is state of speaking.
When text information to be processed is matched with the condition of interrupting, illustrate that user wants to interrupt the broadcasting of the first voice messaging.
At this point, executing subject can stop playing above-mentioned first voice messaging, and marking currently is shape of speaking with the interaction mode of user
State.Wherein, illustrate that state is carrying out information exchange for characterizing user.
Step 405, the second voice messaging of corresponding above-mentioned voice messaging to be processed is obtained, and plays above-mentioned second voice letter
Breath.
The broadcasting of the first voice messaging of User break, usually user need to handle other affairs, or believe the first voice
Cease situations such as content played is further putd question to.For this purpose, executing subject can believe the voice to be processed that user issues
Breath is analyzed, and the second voice messaging of corresponding voice messaging to be processed is got, and plays the second voice messaging.For example, working as
Voice messaging to be processed are as follows: whens " I there are also the processing of other things, next time to be said again " etc., the second voice messaging may is that " it is good,
Next time, I contacted you again ";When voice messaging to be processed are as follows: " wait down, the XX that you said just now be what the meaning " when, the second voice
Information may is that " XX say be ".It can be seen that second voice messaging can not for different voice messagings to be processed
Together.As it can be seen that the flexibility for carrying out information exchange with user can be improved in technical scheme, information exchange efficiency is improved.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, present disclose provides one kind for interaction letter
One embodiment of the device of breath, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 5, the device 500 for interactive information of the present embodiment may include: text information acquiring unit
501, matching relationship judging unit 502 and the first interaction control unit 503.Wherein, text information acquiring unit 501, in response to
Voice messaging to be processed is got when playing the first voice messaging, is configured to obtain the to be processed of above-mentioned voice messaging to be processed
Text information;Matching relationship judging unit 502 is configured to judge that above-mentioned text information to be processed and the matching for the condition that interrupts close
System, wherein the above-mentioned condition that interrupts is used to indicate above-mentioned first voice messaging of stopping broadcasting;First interaction control unit 503, response
It is matched in above-mentioned text information to be processed with the condition of interrupting, is configured to stop to play above-mentioned first voice messaging.
In some optional implementations of the present embodiment, above-mentioned matching relationship judging unit 502 may include: to play
Attribute obtains subelement (not shown), acquisition time obtains subelement (not shown) and the first matching relationship judgement
Unit (not shown).Wherein, it plays attribute and obtains the broadcasting category that subelement is configured to obtain above-mentioned first voice messaging
Property, above-mentioned broadcasting attribute includes at least one of the following: that playing duration, broadcasting interrupt the time;Acquisition time obtains subelement, is matched
It is set to the acquisition time for inquiring above-mentioned voice messaging to be processed;First matching relationship judgment sub-unit, when in response to above-mentioned acquisition
Between in above-mentioned playing duration, and time of the above-mentioned acquisition time in above-mentioned playing duration is later than above-mentioned broadcasting and interrupts the time,
It is configured to judge that above-mentioned text information to be processed is matched with the condition of interrupting.
In some optional implementations of the present embodiment, above-mentioned matching relationship judging unit 502 may include: first
Entry divides subelement (not shown) and the second matching relationship judgment sub-unit (not shown).Wherein, the first entry
Subelement is divided to be configured to above-mentioned text information to be processed being divided at least one entry;Second matching relationship judgement is single
Member belongs to the entry for interrupting set of words in response to existing at least one above-mentioned entry, is configured to judge above-mentioned text to be processed
Word information is matched with the condition of interrupting, wherein the above-mentioned set of words that interrupts includes that at least one interrupts word, and the above-mentioned word that interrupts is used to indicate
Stop playing above-mentioned first voice messaging.
In some optional implementations of the present embodiment, above-mentioned matching relationship judging unit 502 may include: second
Entry divides subelement (not shown), interrupts and judge that acquisition of information subelement (not shown) and third matching relationship are sentenced
Disconnected subelement (not shown).Wherein, the second entry divides subelement and is configured to divide above-mentioned text information to be processed
For at least one entry;Interrupt and judge acquisition of information subelement, for the entry at least one above-mentioned entry, be configured to by
Entry importing interrupts word judgment models, and obtain corresponding to the entry interrupts judgement information, wherein above-mentioned to interrupt word identification model
Judgement information is interrupted for obtain the corresponding entry, and above-mentioned interrupt judge that information includes that presence interrupts and there is no interrupt;The
Three matching relationship judgment sub-units judge that information interrupts to exist in response to above-mentioned interrupt, are configured to judge above-mentioned to be processed
Text information is matched with the condition of interrupting.
In some optional implementations of the present embodiment, the above-mentioned device 500 for interactive information can also include
Word judgment models training unit (not shown) is interrupted, training is configured to and interrupts word judgment models, it is above-mentioned to interrupt word judgement
Model training unit may include: that sample acquisition subelement (not shown) (is schemed with word judgment models training subelement is interrupted
In be not shown).Wherein, sample acquisition subelement is configured to obtain every in multiple sample entries and above-mentioned multiple sample entries
Sample corresponding to a sample entry interrupts judgement information;Interrupting word judgment models training subelement, be configured to will be above-mentioned multiple
Each sample entry in sample entry is as input, by sample corresponding to each sample entry in above-mentioned multiple sample entries
Originally it interrupts and judges information as output, training obtains interrupting word judgment models.
In some optional implementations of the present embodiment, the above-mentioned word judgment models training subelement that interrupts be can wrap
It includes: interrupting word judgment models training module (not shown), be configured to each sample in above-mentioned multiple sample entries
Entry is sequentially input to initially word judgment models are interrupted, and is obtained corresponding to each sample entry in above-mentioned multiple sample entries
Prediction interrupt judgement information, prediction corresponding to each sample entry in above-mentioned multiple sample entries is interrupted judge information and
Sample corresponding to the sample entry, which interrupts, judges that information is compared, and it is quasi- to obtain the above-mentioned prediction for initially interrupting word judgment models
True rate, determines whether above-mentioned predictablity rate is greater than default accuracy rate threshold value, then will be upper if more than above-mentioned default accuracy rate threshold value
State initially interrupt word judgment models as training completion interrupt word judgment models.
In some optional implementations of the present embodiment, the above-mentioned word judgment models training subelement that interrupts be can wrap
Include: parameter adjustment module (not shown) is configured to adjust above-mentioned first in response to being not more than above-mentioned default accuracy rate threshold value
Begin to interrupt the parameters of word judgment models, and returns and above-mentioned interrupt word judgment models training module.
In some optional implementations of the present embodiment, the above-mentioned device 500 for interactive information can also include
Sample entry acquiring unit (not shown), is configured to obtain sample entry, and above-mentioned sample entry acquiring unit can wrap
It includes: interrupting word and obtain subelement, word acquisition subelement of equal value and sample entry acquisition subelement.Wherein, it interrupts word and obtains son list
Member is configured to obtain at least one and interrupts word;Word of equal value obtains subelement, for it is above-mentioned at least one interrupt interrupting in word
Word, is configured to obtain the word of equal value for interrupting word, and above-mentioned equivalence word includes at least one of the following: that corresponding this interrupts the nearly justice of word
Word, corresponding this interrupt the dialect of word;Sample entry obtain subelement be configured to by it is above-mentioned at least one interrupt interrupting in word
The corresponding word of equal value of word interrupts the sample entry of word labeled as this.
In some optional implementations of the present embodiment, the above-mentioned device 500 for interactive information can also include:
Second interaction control unit (not shown) is unsatisfactory for interrupting condition, be configured in response to above-mentioned text information to be processed
Continue to play above-mentioned first voice messaging.
It is above-mentioned for handing over after above-mentioned first interaction control unit in some optional implementations of the present embodiment
The device 500 of mutual information further include: dry state symbol unit (not shown), being configured to mark current interaction mode is to speak
State.
In some optional implementations of the present embodiment, the above-mentioned device 500 for interactive information can also include:
Third interaction control unit (not shown) is configured to obtain the second voice letter of corresponding above-mentioned voice messaging to be processed
Breath, and play above-mentioned second voice messaging.
The present embodiment additionally provides a kind of electronic equipment, comprising: one or more processors;Memory is stored thereon with
One or more programs, when said one or multiple programs are executed by said one or multiple processors, so that said one
Or multiple processors execute the above-mentioned method for interactive information.
The present embodiment additionally provides a kind of computer-readable medium, is stored thereon with computer program, and the program is processed
Device realizes the above-mentioned method for interactive information when executing.
Below with reference to Fig. 6, it illustrates the electronic equipments for being suitable for being used to realize embodiment of the disclosure (for example, in Fig. 1
Server 105) computer system 600 structural schematic diagram.Electronic equipment shown in Fig. 6 is only an example, should not be right
The function and use scope of embodiment of the disclosure bring any restrictions.
As shown in fig. 6, electronic equipment 600 may include processing unit (such as central processing unit, graphics processor etc.)
601, random access can be loaded into according to the program being stored in read-only memory (ROM) 602 or from storage device 608
Program in memory (RAM) 603 and execute various movements appropriate and processing.In RAM 603, it is also stored with electronic equipment
Various programs and data needed for 600 operations.Processing unit 601, ROM 602 and RAM603 are connected with each other by bus 604.
Input/output (I/O) interface 605 is also connected to bus 604.
In general, following device can connect to I/O interface 605: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 606 of head, microphone, accelerometer, gyroscope etc.;Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 607 of dynamic device etc.;Storage device 608 including such as tape, hard disk etc.;And communication device 609.Communication device
609, which can permit electronic equipment 600, is wirelessly or non-wirelessly communicated with other equipment to exchange data.Although Fig. 6 shows tool
There is the electronic equipment 600 of various devices, it should be understood that being not required for implementing or having all devices shown.It can be with
Alternatively implement or have more or fewer devices.Each box shown in Fig. 6 can represent a device, can also root
According to needing to represent multiple devices.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communication device 609, or from storage device 608
It is mounted, or is mounted from ROM 602.When the computer program is executed by processing unit 601, the implementation of the disclosure is executed
The above-mentioned function of being limited in the method for example.
It is situated between it should be noted that the above-mentioned computer-readable medium of embodiment of the disclosure can be computer-readable signal
Matter or computer readable storage medium either the two any combination.Computer readable storage medium for example can be with
System, device or the device of --- but being not limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or it is any more than
Combination.The more specific example of computer readable storage medium can include but is not limited to: have one or more conducting wires
Electrical connection, portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type are programmable
Read-only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic are deposited
Memory device or above-mentioned any appropriate combination.In embodiment of the disclosure, computer readable storage medium, which can be, appoints
What include or the tangible medium of storage program that the program can be commanded execution system, device or device use or and its
It is used in combination.And in embodiment of the disclosure, computer-readable signal media may include in a base band or as carrier wave
The data-signal that a part is propagated, wherein carrying computer-readable program code.The data-signal of this propagation can be adopted
With diversified forms, including but not limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal is situated between
Matter can also be any computer-readable medium other than computer readable storage medium, which can be with
It sends, propagate or transmits for by the use of instruction execution system, device or device or program in connection.Meter
The program code for including on calculation machine readable medium can transmit with any suitable medium, including but not limited to: electric wire, optical cable,
RF (radio frequency) etc. or above-mentioned any appropriate combination.
Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment;It is also possible to individualism, and not
It is fitted into the electronic equipment.Above-mentioned computer-readable medium carries one or more program, when said one or more
When a program is executed by the electronic equipment, so that the electronic equipment: being got when in response to playing the first voice messaging to be processed
Voice messaging obtains the text information to be processed of above-mentioned voice messaging to be processed;Judge above-mentioned text information to be processed and interrupts
The matching relationship of condition, wherein the above-mentioned condition that interrupts is used to indicate above-mentioned first voice messaging of stopping broadcasting;In response to it is above-mentioned to
Processing text information is matched with the condition of interrupting, and stops playing above-mentioned first voice messaging.
The behaviour for executing embodiment of the disclosure can be write with one or more programming languages or combinations thereof
The computer program code of work, above procedure design language include object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).
Flow chart and block diagram in attached drawing illustrate system, method and the computer of the various embodiments according to the disclosure
The architecture, function and operation in the cards of program product.In this regard, each box in flowchart or block diagram can be with
A part of a module, program segment or code is represented, a part of the module, program segment or code includes one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are practical
On can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it wants
It is noted that the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart, Ke Yiyong
The dedicated hardware based system of defined functions or operations is executed to realize, or can be referred to specialized hardware and computer
The combination of order is realized.
Being described in unit involved in embodiment of the disclosure can be realized by way of software, can also be passed through
The mode of hardware is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor
Including text information acquiring unit, matching relationship judging unit and the first interaction control unit.Wherein, the title of these units exists
The restriction to the unit itself is not constituted in the case of certain, for example, the first interaction control unit is also described as " being used for
When satisfaction interrupts condition, stop the unit for playing the first voice messaging ".
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (24)
1. a kind of method for interactive information, comprising:
Get voice messaging to be processed when in response to playing the first voice messaging, obtain the voice messaging to be processed wait locate
Manage text information;
Judge the text information to be processed and interrupt the matching relationship of condition, wherein the condition that interrupts is used to indicate stopping
Play first voice messaging;
It is matched in response to the text information to be processed with the condition of interrupting, stops playing first voice messaging.
2. according to the method described in claim 1, wherein, the matching of the judgement text information to be processed and the condition that interrupts
Relationship, comprising:
The broadcasting attribute of first voice messaging is obtained, the broadcasting attribute includes at least one of the following: playing duration, plays
Interrupt the time;
Inquire the acquisition time of the voice messaging to be processed;
In response to the acquisition time in the playing duration, and time evening of the acquisition time in the playing duration
The time is interrupted in the broadcasting, judges that the text information to be processed is matched with the condition of interrupting.
3. according to the method described in claim 1, wherein, the matching of the judgement text information to be processed and the condition that interrupts
Relationship, comprising:
The text information to be processed is divided at least one entry;
Interrupt the entry of set of words in response to existing to belong at least one described entry, judge the text information to be processed and
Interrupt condition matching, wherein the set of words that interrupts includes that at least one interrupts word, and the word that interrupts is used to indicate stopping broadcasting
First voice messaging.
4. according to the method described in claim 1, wherein, the matching of the judgement text information to be processed and the condition that interrupts
Relationship, comprising:
The text information to be processed is divided at least one entry;
For the entry at least one described entry, entry importing is interrupted into word judgment models, obtains corresponding to the entry
Interrupt judgement information, wherein it is described to interrupt word identification model and interrupt judgement information for obtain the corresponding entry, it is described to interrupt
Judge that information is interrupted including presence and there is no interrupt;
Judge that information interrupts to exist in response to described interrupt, judges that the text information to be processed is matched with the condition of interrupting.
5. according to the method described in claim 4, wherein, the word judgment models that interrupt are obtained by following steps training:
It obtains sample corresponding to each sample entry in multiple sample entries and the multiple sample entry and interrupts judgement letter
Breath;
Using each sample entry in the multiple sample entry as input, by each sample in the multiple sample entry
Sample corresponding to entry, which interrupts, judges information as output, and training obtains interrupting word judgment models.
6. according to the method described in claim 5, wherein, each sample entry using in the multiple sample entry as
Sample corresponding to each sample entry in the multiple sample entry is interrupted and judges information as output, training by input
It obtains interrupting word judgment models, comprising:
It executes following training step: each sample entry in the multiple sample entry is sequentially input to initially interrupting word and sentence
Disconnected model, obtains prediction corresponding to each sample entry in the multiple sample entry and interrupts judgement information, will be described more
Prediction corresponding to each sample entry in a sample entry, which interrupts, judges that sample corresponding to information and the sample entry is beaten
It is disconnected to judge that information is compared, the predictablity rate for initially interrupting word judgment models is obtained, determines the predictablity rate
Whether it is greater than default accuracy rate threshold value, if more than the default accuracy rate threshold value, then makees the word judgment models that initially interrupt
Word judgment models are interrupted for what training was completed.
7. according to the method described in claim 6, wherein, each sample entry using in the multiple sample entry as
Sample corresponding to each sample entry in the multiple sample entry is interrupted and judges information as output, training by input
It obtains interrupting word judgment models, comprising:
In response to being not more than the default accuracy rate threshold value, adjustment is described initially to interrupt the parameter of word judgment models, and continues to hold
The row training step.
8. according to the method described in claim 6, wherein, the sample entry is obtained by following steps:
It obtains at least one and interrupts word;
For it is described at least one interrupt in word and interrupt word, obtain the word of equal value for interrupting word, the equivalence word includes following
At least one of: corresponding this interrupts the near synonym of word, corresponding this interrupts the dialect of word;
By it is described at least one interrupt the corresponding word of equal value of word that interrupts in word and interrupt the sample entry of word labeled as this.
9. according to the method described in claim 1, wherein, the method also includes:
It is unsatisfactory for interrupting condition in response to the text information to be processed, continues to play first voice messaging.
10. according to the method described in claim 1, wherein, after the stopping plays first voice messaging, the method
Further include:
Marking current interaction mode is state of speaking.
11. according to the method described in claim 10, wherein, the method also includes:
The second voice messaging of the corresponding voice messaging to be processed is obtained, and plays second voice messaging.
12. a kind of device for interactive information, comprising:
Text information acquiring unit getting voice messaging to be processed when in response to playing the first voice messaging, being configured to obtain
Take the text information to be processed of the voice messaging to be processed;
Matching relationship judging unit is configured to judge the text information to be processed and interrupts the matching relationship of condition, wherein
The condition that interrupts is used to indicate stopping broadcasting first voice messaging;
First interaction control unit matches with the condition of interrupting in response to the text information to be processed, is configured to stop to play
First voice messaging.
13. device according to claim 12, wherein the matching relationship judging unit includes:
It plays attribute and obtains subelement, be configured to obtain the broadcasting attribute of first voice messaging, the broadcasting attribute packet
Include at least one of following: playing duration, broadcasting interrupt the time;
Acquisition time obtains subelement, is configured to inquire the acquisition time of the voice messaging to be processed;
First matching relationship judgment sub-unit, in response to the acquisition time in the playing duration, and the acquisition time
Time in the playing duration is later than the broadcasting and interrupts the time, is configured to judge the text information to be processed and beats
The matching of broken strip part.
14. device according to claim 12, wherein the matching relationship judging unit includes:
First entry divides subelement, is configured to the text information to be processed being divided at least one entry;
Second matching relationship judgment sub-unit belongs to the entry for interrupting set of words in response to existing at least one described entry,
It is configured to judge that the text information to be processed is matched with the condition of interrupting, wherein the set of words that interrupts includes at least one
Word is interrupted, the word that interrupts is used to indicate stopping broadcasting first voice messaging.
15. device according to claim 12, wherein the matching relationship judging unit includes:
Second entry divides subelement, is configured to the text information to be processed being divided at least one entry;
It interrupts and judges acquisition of information subelement, for the entry at least one described entry, be configured to import the entry
Interrupt word judgment models, obtain corresponding to the entry interrupts judgement information, wherein it is described interrupt word identification model for obtain pair
Should entry interrupt judgement information, described interrupt judges that information includes existing to interrupt and there is no interrupt;
Third matching relationship judgment sub-unit judges that information interrupts to exist in response to described interrupt, is configured to judge described
Text information to be processed is matched with the condition of interrupting.
16. device according to claim 15, wherein described device further includes interrupting word judgment models training unit, quilt
It is configured to training and interrupts word judgment models, the word judgment models training unit that interrupts includes:
Sample acquisition subelement is configured to obtain each sample entry in multiple sample entries and the multiple sample entry
Corresponding sample interrupts judgement information;
Word judgment models training subelement is interrupted, is configured to using each sample entry in the multiple sample entry as defeated
Enter, sample corresponding to each sample entry in the multiple sample entry is interrupted judge information as output, it is trained
To interrupting word judgment models.
17. device according to claim 16, wherein it is described interrupt word judgment models training subelement include:
Word judgment models training module is interrupted, is configured to sequentially input each sample entry in the multiple sample entry
To initially word judgment models are interrupted, obtains prediction corresponding to each sample entry in the multiple sample entry and interrupt judgement
Prediction corresponding to each sample entry in the multiple sample entry is interrupted and judges information and the sample entry institute by information
Corresponding sample, which interrupts, judges that information is compared, and obtains the predictablity rate for initially interrupting word judgment models, determines institute
It states whether predictablity rate is greater than default accuracy rate threshold value, if more than the default accuracy rate threshold value, is then initially interrupted described
Word judgment models interrupt word judgment models as what training was completed.
18. device according to claim 17, wherein it is described interrupt word judgment models training subelement include:
Parameter adjustment module is configured to adjust and described initially interrupt word and sentence in response to being not more than the default accuracy rate threshold value
The parameter of disconnected model, and word judgment models training module is interrupted described in return.
19. device according to claim 17, wherein described device further includes sample entry acquiring unit, is configured to
Sample entry is obtained, the sample entry acquiring unit includes:
It interrupts word and obtains subelement, be configured to obtain at least one and interrupt word;
Word of equal value obtains subelement, for it is described at least one interrupt in word and interrupt word, be configured to obtain this and interrupt word
Word of equal value, the equivalence word include at least one of the following: that corresponding this interrupts the near synonym of word, corresponding this interrupts the dialect of word;
Sample entry obtains subelement, is configured to mark at least one described corresponding word of equal value of word that interrupts interrupted in word
The sample entry of word is interrupted for this.
20. device according to claim 12, wherein described device further include:
Second interaction control unit is unsatisfactory for interrupting condition in response to the text information to be processed, is configured to continue to play
First voice messaging.
21. device according to claim 12, wherein after first interaction control unit, described device further include:
Dry state symbol unit, being configured to mark current interaction mode is state of speaking.
22. device according to claim 21, wherein described device further include:
Third interaction control unit, is configured to obtain the second voice messaging of the corresponding voice messaging to be processed, and plays
Second voice messaging.
23. a kind of electronic equipment, comprising:
One or more processors;
Memory is stored thereon with one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors
Perform claim requires any method in 1 to 11.
24. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
Method of the Shi Shixian as described in any in claim 1 to 11.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910720051.4A CN110427460A (en) | 2019-08-06 | 2019-08-06 | Method and device for interactive information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910720051.4A CN110427460A (en) | 2019-08-06 | 2019-08-06 | Method and device for interactive information |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110427460A true CN110427460A (en) | 2019-11-08 |
Family
ID=68414329
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910720051.4A Pending CN110427460A (en) | 2019-08-06 | 2019-08-06 | Method and device for interactive information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110427460A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111540349A (en) * | 2020-03-27 | 2020-08-14 | 北京捷通华声科技股份有限公司 | Voice interruption method and device |
CN111599354A (en) * | 2020-06-28 | 2020-08-28 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic device, and computer-readable medium for converting information |
CN111752523A (en) * | 2020-05-13 | 2020-10-09 | 深圳追一科技有限公司 | Human-computer interaction method and device, computer equipment and storage medium |
CN112489642A (en) * | 2020-10-21 | 2021-03-12 | 深圳追一科技有限公司 | Method, apparatus, device and storage medium for controlling voice robot response |
CN112799632A (en) * | 2021-01-26 | 2021-05-14 | 深圳市普渡科技有限公司 | Robot voice playing method and robot |
CN113535908A (en) * | 2020-04-20 | 2021-10-22 | 北京有限元科技有限公司 | Method, device and storage medium for interrupting audio interaction |
-
2019
- 2019-08-06 CN CN201910720051.4A patent/CN110427460A/en active Pending
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111540349A (en) * | 2020-03-27 | 2020-08-14 | 北京捷通华声科技股份有限公司 | Voice interruption method and device |
CN111540349B (en) * | 2020-03-27 | 2023-10-10 | 北京捷通华声科技股份有限公司 | Voice breaking method and device |
CN113535908A (en) * | 2020-04-20 | 2021-10-22 | 北京有限元科技有限公司 | Method, device and storage medium for interrupting audio interaction |
CN113535908B (en) * | 2020-04-20 | 2024-06-21 | 北京中科金有限元技术有限公司 | Method, device and storage medium for interrupting audio interaction |
CN111752523A (en) * | 2020-05-13 | 2020-10-09 | 深圳追一科技有限公司 | Human-computer interaction method and device, computer equipment and storage medium |
CN111599354A (en) * | 2020-06-28 | 2020-08-28 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic device, and computer-readable medium for converting information |
CN112489642A (en) * | 2020-10-21 | 2021-03-12 | 深圳追一科技有限公司 | Method, apparatus, device and storage medium for controlling voice robot response |
CN112489642B (en) * | 2020-10-21 | 2024-05-03 | 深圳追一科技有限公司 | Method, device, equipment and storage medium for controlling voice robot response |
CN112799632A (en) * | 2021-01-26 | 2021-05-14 | 深圳市普渡科技有限公司 | Robot voice playing method and robot |
CN112799632B (en) * | 2021-01-26 | 2023-12-19 | 深圳市普渡科技有限公司 | Robot voice playing method and robot |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110427460A (en) | Method and device for interactive information | |
CN108022586B (en) | Method and apparatus for controlling the page | |
CN109599113A (en) | Method and apparatus for handling information | |
CN109243432A (en) | Speech processing method and the electronic equipment for supporting the speech processing method | |
CN107657017A (en) | Method and apparatus for providing voice service | |
CN110347863B (en) | Speaking recommendation method and device and storage medium | |
CN109545192A (en) | Method and apparatus for generating model | |
CN110140167A (en) | Information processing equipment and information processing method | |
CN107623614A (en) | Method and apparatus for pushed information | |
CN109961792A (en) | The method and apparatus of voice for identification | |
CN107748500A (en) | Method and apparatus for controlling smart machine | |
US11011176B1 (en) | Voice synthesis for virtual agents | |
CN108924218A (en) | Method and apparatus for pushed information | |
CN109739605A (en) | The method and apparatus for generating information | |
CN110009059A (en) | Method and apparatus for generating model | |
CN109829164A (en) | Method and apparatus for generating text | |
CN110308886A (en) | The system and method for voice command service associated with personalized task are provided | |
CN110457476A (en) | Method and apparatus for generating disaggregated model | |
CN109671435A (en) | Method and apparatus for waking up smart machine | |
CN108897517B (en) | Information processing method and electronic equipment | |
CN108877779A (en) | Method and apparatus for detecting voice tail point | |
CN108494975A (en) | Incoming call response method and equipment | |
CN110288683A (en) | Method and apparatus for generating information | |
CN110223694A (en) | Method of speech processing, system and device | |
CN109074809A (en) | Information processing equipment, information processing method and program |
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 |