CN109634692A - Vehicle-mounted conversational system and processing method and system for it - Google Patents
Vehicle-mounted conversational system and processing method and system for it Download PDFInfo
- Publication number
- CN109634692A CN109634692A CN201811239635.1A CN201811239635A CN109634692A CN 109634692 A CN109634692 A CN 109634692A CN 201811239635 A CN201811239635 A CN 201811239635A CN 109634692 A CN109634692 A CN 109634692A
- Authority
- CN
- China
- Prior art keywords
- thread
- processing
- request data
- voice
- voice request
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims description 25
- 230000007704 transition Effects 0.000 claims description 10
- 235000013399 edible fruits Nutrition 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 10
- 230000006399 behavior Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/448—Execution paradigms, e.g. implementations of programming paradigms
- G06F9/4482—Procedural
- G06F9/4484—Executing subprograms
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Abstract
The invention belongs to vehicle-mounted voice fields, specifically provide a kind of vehicle-mounted conversational system and the processing method and system for it.Present invention seek to address that the single thread chain type processing method logic of existing vehicle-mounted conversational system is complicated, the bad problem of scalability.Processing method for vehicle-mounted conversational system of the invention includes the following steps: to identify multiple processing threads according to voice request data；The multiple processing thread parallel is set to execute respective operation；According to the implementing result of the multiple processing thread, the processing result of the voice request data is determined.It is performed simultaneously a plurality of relevant processing thread, the execution speed of entire process flow is improved compared with single thread executes and adds time rollback, can obtain relevant implementing result faster；But also new processing thread can be expanded, improve the scalability of vehicle-mounted conversational system.
Description
Technical field
The invention belongs to vehicle-mounted voice fields, specifically provide a kind of vehicle-mounted conversational system and for its processing method and be
System.
Background technique
The processing mode of existing vehicle-mounted voice system is usually chain structure, i.e., may relate in a speech query information
And when arriving multiple tasks field, system needs first to back up current users session context, then executes the processing logic of field A,
The processing logic of field A is finished and then the processing logic of rollback execution field B, last decision are still led using field A
The implementing result of domain B.
Such as, the query information of the previous sentence of user is " going to Tian An-men ", and more wheels that user can enter navigation field are talked with, system
User can be waited to carry out list selection.If user says " daphne odera " at this time, system thinks that user had both been likely to be carry out list
Selection, it is also possible to be intended to listen song.Therefore, system can first carry out the more wheel dialogues of navigation after backup user session context
Processing logic.After the logical process of navigation field is complete, system can return back to field matching module, and backed up before use
User session context.It is field of media that field matching module, which matches tasks in areas, behind will walk the processing of field of media
Then logic carries out decision again, determine the implementing result finally using navigation field or field of media.This chain type processing knot
Structure needs to back up user session context, supports to retract, and logical comparison is complicated, and scalability is bad.
Correspondingly, this field needs a kind of vehicle-mounted voice system and its processing method to solve the above problems.
Summary of the invention
In a first aspect, the treating method comprises the present invention provides a kind of processing method for vehicle-mounted conversational system
Following steps: multiple processing threads are identified according to voice request data；Execute the multiple processing thread parallel respective
Operation；According to the implementing result of the multiple processing thread, the processing result of the voice request data is determined.
In the optimal technical scheme of above-mentioned processing method, if currently processed thread includes multiple sub threads, " according to institute
State it is multiple processing threads implementing results, determine the processing result of the voice request data " the step of before, the method is also
It include: to identify multiple sub threads according to the voice request data；The multiple sub thread is set to execute respective operation parallel；
According to the implementing result of the multiple sub thread, the implementing result of the currently processed thread is determined.
In the optimal technical scheme of above-mentioned processing method, if the implementing result of the current sub thread includes multiple choosings
, " according to the implementing result of the multiple sub thread, determining the implementing result of the currently processed thread " the step of after,
The method also includes: data are confirmed according to the multiple option and user's voice currently entered, are determined described currently processed
The implementing result of thread.
In the optimal technical scheme of above-mentioned processing method, " multiple processing lines are identified according to the voice request data
The step of journey " includes: the voice keyword for obtaining the voice request data；Based on preset keyword and multiple processing threads
Between corresponding relationship, match corresponding with the voice keyword all processing threads.
In the optimal technical scheme of above-mentioned processing method, " identifying multiple sub threads according to the voice request data "
The step of include: to the voice request data carry out semantic analysis；According to the semantic analysis as a result, determining relevant institute
There is sub thread.
Second aspect, the present invention also provides a kind of processing system for vehicle-mounted conversational system, the processing system packet
The first identification module is included, is used to identify multiple processing threads according to voice request data；First processing module is used to make
The multiple processing thread parallel executes respective operation；First decision-making module is used for according to the multiple processing thread
Implementing result determines the processing result of the voice request data.
It is described if the currently processed thread includes multiple sub threads in the optimal technical scheme of above-mentioned processing system
Processing system further includes the second identification module, is used to identify multiple sub threads according to the voice request data；At second
Module is managed, is used to that the multiple sub thread to be made to execute respective operation parallel；Second decision-making module is used for according to described more
The implementing result of a sub thread determines the implementing result of the currently processed thread.
In the optimal technical scheme of above-mentioned processing system, the processing system further includes state transition module, if described
The implementing result of current sub thread includes multiple options, and the state transition module according to the multiple option with user for working as
The voice of preceding input confirms data, determines the implementing result of the currently processed thread.
In the optimal technical scheme of above-mentioned processing system, first identification module is configured with disaggregated model, is used for
Based on the corresponding relationship between preset keyword and multiple processing threads, all places corresponding with the voice keyword are matched
Lineation journey.
In the optimal technical scheme of above-mentioned processing system, second identification module is configured with semantic understanding model,
For carrying out semantic analysis and according to the semantic analysis as a result, determining relevant sub thread to the voice request data.
The third aspect, the present invention also provides any one of a kind of vehicle-mounted conversational systems, including above-mentioned optimal technical scheme
The processing system for vehicle-mounted conversational system.
Fourth aspect, the present invention also provides a kind of vehicle device, the vehicle device includes voice output module and above-mentioned preferred skill
Vehicle-mounted conversational system described in art scheme, the voice output module are used for the voice number of the vehicle-mounted conversational system output
It is believed that breath is converted into voice output.
5th aspect, the present invention also provides a kind of vehicle, the vehicle includes described in above-mentioned optimal technical scheme
Vehicle device.
6th aspect, the present invention also provides a kind of storage devices, wherein be stored with a plurality of program, described program be suitable for by
Processor load is used for the processing method of vehicle-mounted conversational system to execute described in any one of above-mentioned optimal technical scheme.
7th aspect, the present invention also provides a kind of control device, including processor, storage equipment, the storage equipment
Suitable for storing a plurality of program, described program is suitable for being loaded by the processor to execute any one of above-mentioned optimal technical scheme institute
The processing method for vehicle-mounted conversational system stated.
Compared with the immediate prior art, technical solution of the present invention at least has as follows
The utility model has the advantages that
1, the present invention will recognise that relevant to the voice request data after receiving the voice request data of user
Multiple processing threads, and this relevant multiple processing thread parallel is made to execute respective operation, finally according to this multiple processing
The implementing result of thread determines the processing result of the voice request data.It is performed simultaneously a plurality of relevant processing thread, with list
Thread executes the execution speed added time and retracted compared to entire process flow is improved, and can obtain faster and relevant execute knot
Fruit improves the usage experience of vehicle-mounted conversational system.
2, when processing thread relevant to voice request data of the invention includes a plurality of sub thread, according to the voice request
Data make relevant sub thread execute respective operation parallel, finally determine the processing line according to the implementing result of related sub thread
The implementing result of journey.So that the sub thread of relevant processing thread and relevant processing thread is executed respective operation parallel, not only mentions
The high execution speed of entire process flow, but also can easily extend new processing thread and handle under thread
Sub thread etc..
Scheme 1, a kind of processing method for vehicle-mounted conversational system, which is characterized in that the treating method comprises as follows
Step: multiple processing threads are identified according to voice request data；The multiple processing thread parallel is set to execute respective operation；
According to the implementing result of the multiple processing thread, the processing result of the voice request data is determined.
Scheme 2, processing method according to scheme 1, which is characterized in that if currently processed thread includes multiple sub-lines
Journey, " according to it is the multiple processing thread implementing result, determine the processing result of the voice request data " the step of it
Before, the method also includes: multiple sub threads are identified according to the voice request data；Hold the multiple sub thread parallel
The respective operation of row；According to the implementing result of the multiple sub thread, the implementing result of the currently processed thread is determined.
Scheme 3, the processing method according to scheme 2, which is characterized in that if the implementing result packet of the current sub thread
Multiple options are included, " according to the implementing result of the multiple sub thread, determining the implementing result of the currently processed thread "
After step, the method also includes: data are confirmed according to the multiple option and user's voice currently entered, determine described in
The implementing result of currently processed thread.
Scheme 4, processing method according to scheme 1, which is characterized in that " identified according to the voice request data
The step of multiple processing threads " includes: the voice keyword for obtaining the voice request data；Based on preset keyword and more
Corresponding relationship between a processing thread matches all processing threads corresponding with the voice keyword.
Scheme 5, the processing method according to scheme 2, which is characterized in that " identified according to the voice request data
The step of multiple sub threads " includes: to carry out semantic analysis to the voice request data；According to the semantic analysis as a result,
Determine relevant all sub threads.
Scheme 6, a kind of processing system for vehicle-mounted conversational system, which is characterized in that the processing system includes first
Identification module is used to identify multiple processing threads according to voice request data；First processing module is used to make described more
A processing thread parallel executes respective operation；First decision-making module is used for the execution knot according to the multiple processing thread
Fruit determines the processing result of the voice request data.
Scheme 7, processing system according to scheme 6, which is characterized in that if the currently processed thread includes multiple sons
Thread, the processing system further include the second identification module, are used to identify multiple sub-lines according to the voice request data
Journey；Second processing module is used to that the multiple sub thread to be made to execute respective operation parallel；Second decision-making module, is used for
According to the implementing result of the multiple sub thread, the implementing result of the currently processed thread is determined.
Scheme 8, processing system according to scheme 7, which is characterized in that the processing system further includes state transition mould
Block, if the implementing result of the current sub thread includes multiple options, the state transition module is used for according to the multiple choosing
Item confirms data with user's voice currently entered, determines the implementing result of the currently processed thread.
Scheme 9, processing system according to scheme 6, which is characterized in that first identification module is configured with classification mould
Type is used for based on the corresponding relationship between preset keyword and multiple processing threads, matching and the voice keyword pair
All processing threads answered.
Scheme 10, processing system according to scheme 7, which is characterized in that second identification module is configured with semanteme
Understand model, is used to carry out the voice request data semantic analysis and according to the semantic analysis as a result, determining phase
The sub thread of pass.
Scheme 11, a kind of vehicle-mounted conversational system, which is characterized in that including being used for described in any one of above scheme 6-10
The processing system of vehicle-mounted conversational system.
Scheme 12, a kind of vehicle device, which is characterized in that the vehicle device includes vehicle-mounted described in voice output module and scheme 11
Conversational system, the voice output module are used to convert voice for the speech data information of the vehicle-mounted conversational system output defeated
Out.
Scheme 13, a kind of vehicle, which is characterized in that the vehicle includes vehicle device described in scheme 12.
Scheme 14, a kind of storage device, wherein being stored with a plurality of program, which is characterized in that described program is suitable for by handling
Device load is with the processing method described in any one of the 1-5 that carries into execution a plan for vehicle-mounted conversational system.
Scheme 15, a kind of control device, including processor, storage equipment, the storage equipment are suitable for storing a plurality of program,
It is characterized in that, described program is vehicle-mounted right to be used for described in any one of the 1-5 that carries into execution a plan suitable for being loaded as the processor
The processing method of telephone system.
Detailed description of the invention
The preferred embodiment of the present invention described with reference to the accompanying drawings and in conjunction with vehicle, in attached drawing:
Fig. 1 is the key step figure of the processing method for vehicle-mounted conversational system of an embodiment of the present invention；
Fig. 2 is the primary structure block diagram of the processing system for vehicle-mounted conversational system of an embodiment of the present invention.
Reference signs list:
1, the first identification module；11, task execution module；12, execution module is chatted；13, user's manual execution module；2,
First decision-making module；3, the second identification module；31, field matching module；32, it is intended to matching module；33, behavior execution module；
4, the second decision-making module；5, state transition module；6, result output module.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this
A little embodiments are used only for explaining technical principle of the invention, it is not intended that limit the scope of the invention.
As shown in Figure 1, Fig. 1 is the key step of the processing method for vehicle-mounted conversational system of an embodiment of the present invention
Figure.Referring to Fig.1, in the present embodiment for vehicle-mounted conversational system processing method the following steps are included:
Step S10, multiple processing threads are identified according to voice request data；
Step S20, multiple processing thread parallels is made to execute respective operation；
Step S30, according to the implementing result of multiple processing threads, the processing result of voice request data is determined.
In an embodiment of the present invention, multiple processing threads have been preset.Illustratively, multiple processing threads are task
Execute processing thread, chat executes processing thread and user's manual executes processing thread.By being identified according to voice request data
Multiple processing threads relevant to the voice request data, and relevant multiple processing thread parallels is made to execute respective operation,
The final process result of voice request data is finally determined according to this relevant multiple processing respective implementing result of thread.This
Sample is performed simultaneously a plurality of relevant processing thread, improves entire process flow compared with time rollback is added in single thread execution
Speed is executed, relevant implementing result can be obtained faster, improve the usage experience of vehicle-mounted conversational system.
Step S10 specifically includes the following steps:
Step S101, the voice keyword of voice request data is obtained；
Step S102, based on the corresponding relationship between preset keyword and multiple processing threads, matching is crucial with voice
The corresponding all processing threads of word.
In an embodiment of the present invention, preset has a large amount of voices of corresponding relationship crucial with multiple processing threads
Word.But since the corresponding relationship of voice keyword and relevant processing thread is fuzzy matching, obtaining voice keyword
Later, multiple relevant processing threads may be matched.Such setting can match related to voice keyword simultaneously
All processing threads, improve execution speed, and increasing the multifarious of implementing result while also ensuring execution
As a result reliability.
Step S30 further comprises: selecting one from the implementing result of multiple processing threads by the first priority policy
A implementing result is determined as the processing result of voice request data.Wherein, the first priority strategy can be with are as follows: task execution processing
Thread runs succeeded > user's manual executes processing thread and runs succeeded > chats execution handle thread run succeeded > task execution at
Lineation journey, which executes failure > user's manual and executes processing thread and executes failure > chat execution and handle thread, executes failure.
Specifically, if user is to do some specific things, such as " Tian An-men is gone in navigation ", there is " navigation ", " day
Peace door " etc. is in relation to the specific keyword for executing operation and place, then the conversation sentence of user, which belongs to task execution, handles thread；
If user is arbitrarily to be chatted, such as " I tired ", then conversation sentence, which belongs to chat, executes processing thread；For another example it uses
Family is some functions of wanting enquiring vehicle user's manual, such as " how opening tail-gate ", occur " how ", " tail-gate " etc. about
The keyword of inquiry and vehicle, that conversation sentence belong to user's manual and execute processing thread.If but user session sentence is that " I starves
", then it may be the restaurant for wanting finding nearby, belong to task execution processing thread；Chat may also be belonged to, belong to chat and execute
Thread is handled, at this point, the two processing threads will be matched.The two processing thread parallels are made to execute subsequent operation, such as
Two processing threads of fruit are performed both by success, then selecting the processing thread of task execution processing thread according to the first priority policy
Implementing result as last processing result.
In an embodiment of the present invention, task execution processing thread includes multiple sub threads.According to voice request data
The processing thread (i.e. currently processed thread) identified be task execution processing thread in the case of, before step S30, this reality
The processing method applied in example for vehicle-mounted conversational system is further comprising the steps of:
Step S21, multiple sub threads are identified according to voice request data；
Step S22, multiple sub threads is made to execute respective operation parallel；
Step S23, according to the implementing result of multiple sub threads, the implementing result of currently processed thread is determined.
So that the sub thread of currently processed thread and currently processed thread is executed respective operation parallel, not only increases whole
The execution speed of a process flow, but also can easily extend the sub thread under new processing thread and processing thread
Deng.
Step S21 specifically includes the following steps:
Step S211, semantic analysis is carried out to voice request data；
Step S212, according to semantic analysis as a result, determining relevant all sub threads.
In an embodiment of the present invention, task execution processing thread includes navigation sub thread, media sub thread, phone sub-line
Journey, vehicle control sub thread and weather sub thread.Identifying that processing thread is task execution processing thread according to voice keyword
And then semantic analysis is carried out to voice request data, it determines the specific field and intention of voice request data, that is, determines phase
All sub threads closed.Such setting, can more clear user session intention, keep implementing result more accurate.
Step S23 further comprises: one is selected from the implementing result of multiple sub threads according to the second priority strategy
Implementing result is determined as the implementing result of currently processed thread.Wherein, the second priority strategy can be with are as follows: the execution of current area
As a result > jumps to the implementing result of other field.Wherein " field " can refer to sub thread fields, such as navigation field, media
Field, telephone field, vehicle control field and weather field.
Specifically, for example first conversation sentence of user is " restaurant near searching ", this when is first according to voice
Keyword, which identifies, handles thread for task execution, is determined as sub thread of navigating further according to the result of semantic analysis, sub-line of navigating
Neighbouring multiple restaurants may be found out after Cheng Zhihang and form list, processing system is in " list selection " state at this time,
User is waited to select a result.User then says " daphne odera " that this when carries out semantic analysis and root to voice request data
According to the result of semantic analysis, it is believed that user may be there are two types of intention, the first is to select one in lists to be called " daphne odera "
Dining room, be for second that can determine whether out navigate sub thread and media sub thread two at this time in the song looked into a head and be called " daphne odera "
A sub thread.Starting navigation two sub threads of sub thread and media sub thread respectively come execute it is remaining execute operation, if two
A sub thread is performed both by success, then selecting the implementing result of current area (i.e. navigation sub thread) according to the second priority policy
Final implementing result as currently processed thread.
Further, in an embodiment of the present invention, every sub thread after receiving voice request data can
Field where allotting voice request data；Then according to the meaning of realm information and the result of semantic analysis matching voice request data
Figure；Specific behavior, such as inquiry address, selection address, page turning etc. are finally executed according to intent information.
After receiving voice request data, the user session context of processing system can save the context of dialog procedure
Data (i.e. voice request data, the semantic understanding result to voice request data, voice request data fields, affiliated line
The execution state of journey and the implementing result data of affiliated thread).Since each route journey can generate not in the process of implementation
Same realm information, execution state and implementing result, that is, generate different user session contexts, therefore in order to guarantee user
The consistency of the context of dialogue, user session context generates entanglement when preventing multichannel thread parallel from executing, and needs for per all the way
The thread of starting copies a user session context, and each route journey is made to have the user session context of oneself.
It should be noted that when determining the sub thread of processing thread according to voice request data, if previously sent out
A raw wheel or several wheels are talked with and the conversation tasks are not completed, then step S21 specifically includes the following steps:
Step S211, semantic analysis is carried out to voice request data；
Step S212, according to the result of semantic analysis and historical user's context of dialogue, (i.e. history voice request data are raw
At user session context), determine relevant all sub threads.
In conjunction with semantic analysis and current user session context, can more accurate stator thread really, guarantee same
Field executes the continuity of operation.
In the case of the implementing result of sub thread includes multiple options, after step S23, the processing method further includes
Following steps: data are confirmed according to multiple options and user's voice currently entered, determine the implementing result of currently processed thread.
Specifically, based on the above embodiment, finally the holding as currently processed thread in the implementing result of navigation sub thread
When row result, the implementing result for sub thread of such as navigating is to search name as the dining room of " daphne odera " and generate navigation circuit column
Table.User can select navigation circuit list later, if dialogue can be " selecting nearest navigation circuit ", according to user
Voice confirmation data currently entered and multinomial navigation circuit determine nearest navigation circuit holding as currently processed thread
Row result.
Although it is understood that preset in advance task execution processing thread in the embodiment of the present invention, chatting and executing
It handles thread and user's manual executes processing thread, but this is not that can also carry out its elsewhere to the limitation of processing thread
The expansion of lineation journey and the sub thread under processing thread.
Based on the embodiment of the above-mentioned processing method for vehicle-mounted conversational system, the present invention also provides one kind for vehicle-mounted
The processing system of conversational system.With reference to the accompanying drawing, which is illustrated.
As shown in Fig. 2, Fig. 2 is the primary structure of the processing system for vehicle-mounted conversational system of an embodiment of the present invention
Block diagram.Referring to Fig. 2, in an embodiment of the present invention, processing system includes: the first identification module 1, is used for according to voice request
Data identify multiple processing threads；First processing module (not shown) is used to that multiple processing thread parallels to be made to execute
Respective operation；First decision-making module 2 is used to determine voice request data according to the implementing result of multiple processing threads
Processing result.
Wherein, the first identification module 1 is configured with disaggregated model, is used for based on preset keyword and multiple processing threads
Between corresponding relationship, match corresponding with voice keyword all processing threads.
As an example, first processing module includes task execution module 11, chats execution module 12 and user's manual
Execution module 13.Preset the first priority policy, the first priority policy can be with are as follows: task execution module 11 runs succeeded
> user's manual execution module 13 runs succeeded > chat execution module 12 run succeeded > task execution module 11 executes failure > use
Family handbook execution module 13 executes failure > chat execution module 12 and executes failure.First decision-making module 2 can be preferential according to first
Grade strategy selects processing result of the implementing result as voice request data from the implementing result of multiple processing threads.
In an embodiment of the present invention, processing system further include: the second identification module 3 is used for according to voice request number
According to identifying multiple sub threads；Second processing module is used to that multiple sub threads to be made to execute respective operation parallel；Second decision
Module 4 is used for the implementing result according to multiple sub threads, determines the implementing result of currently processed thread.
Wherein, the second identification module 3 is configured with semantic understanding model, is used to carry out semantic analysis to voice request data
And according to semantic analysis as a result, determining relevant sub thread.
When determining the sub thread of processing thread according to voice request data, if previously having existed user session or more
When text, i.e. front have occurred and that a wheel or a few wheels are talked with, semantic understanding model can be also used for carrying out language to voice request data
Justice is analyzed and according to the result of semantic analysis and historical user's context of dialogue, determines relevant sub thread.
In an embodiment of the present invention, Second processing module include navigation sub thread, media sub thread, phone sub thread,
Vehicle control sub thread and weather sub thread, and every sub thread includes field matching module 31, is intended to matching module 32 and row
For execution module 33.Wherein field matching module 31 is used to match field where voice request data, and by realm information
It passes to and is intended to matching module 32；It is intended to matching module 32 to manage for receiving realm information, and according to realm information and semanteme
The result that solution model carries out semantic understanding matches the intention of voice request data, and intent information is passed to behavior and executes mould
Block 33；Behavior execution module 33 executes specific behavior for receiving intent information, and according to intent information, for example inquires
POI selects POI, page turning etc., and the result of execution is passed to the second decision-making module 4.
As an example, the second priority policy is preset, the second priority policy can be with are as follows: current area is held
Row result > jumps to the implementing result of other field.Second decision-making module 4 can be according to the second priority policy from multiple sons
The implementing result of thread selects implementing result of the implementing result as currently processed thread.
In an embodiment of the present invention, processing system further includes state transition module 5 and result output module 6.In sub-line
In the case of the implementing result of journey includes multiple options, state transition module 5 according to multiple options with user for currently inputting
Voice confirm data, determine the implementing result of currently processed thread.As a result output module 6 is used for above and below according to user session
The result generation speech data information that jumps of text, implementing result and state transition is exported to the first decision-making module 2.
Based on the above-mentioned processing system embodiment for vehicle-mounted conversational system, the present invention also provides a kind of vehicle-mounted dialogues to be
System, including the processing system for vehicle-mounted conversational system.
The processing system is configured in vehicle-mounted conversational system, can accelerate the same of the speed of service of vehicle-mounted conversational system
When, guarantee the accuracy of the output result of vehicle-mounted conversational system；And the expansibility of vehicle-mounted conversational system is also added, it improves
The usage experience of vehicle-mounted conversational system.
Based on above-mentioned vehicle-mounted conversational system embodiment, the present invention also provides a kind of vehicle device, which includes voice output
Module and above-mentioned vehicle-mounted conversational system, the speech data information that voice output module is used to export vehicle-mounted conversational system convert
For voice output.
Based on above-mentioned vehicle device embodiment, the present invention also provides a kind of vehicle, which includes above-mentioned vehicle device.
Based on above method embodiment, the present invention also provides a kind of storage devices, wherein it is stored with a plurality of program, the journey
Sequence may be adapted to be loaded by processor to execute the above-mentioned processing method for vehicle-mounted conversational system.
Based on above method embodiment, the present invention also provides a kind of control devices, including processor, storage equipment；It deposits
It stores up equipment and is suitable for storing a plurality of program, which may be adapted to be loaded by processor above-mentioned for vehicle-mounted conversational system to execute
Processing method.
Person of ordinary skill in the field can be understood that for convenience of description and succinctly, the present invention fills
Set, the specific work process and related description of system embodiment, can refer to preceding method embodiment corresponding process, and with
Above method embodiment beneficial effect having the same, details are not described herein.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can be executed with hardware, processor
The combination of software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only memory
(ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field
In any other form of storage medium well known to interior.
Those skilled in the art should be able to recognize that, side described in conjunction with the examples disclosed in the embodiments of the present disclosure
Method step, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate electronic hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is executed actually with electronic hardware or software mode, specific application and design constraint depending on technical solution.
Those skilled in the art can use different methods to achieve the described function each specific application, but this reality
Now it should not be considered as beyond the scope of the present invention.
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, rather than be used to describe or indicate specific sequence or precedence.It should be understood that this
The data that sample uses can be interchanged in appropriate circumstances, so that the embodiment of the present invention described herein can be in addition at this
In illustrate or description those of other than sequence implement.
So far, it has been combined preferred embodiment shown in the drawings and describes technical solution of the present invention, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this
Under the premise of the principle of invention, those skilled in the art can make equivalent change or replacement to the relevant technologies feature, these
Technical solution after change or replacement will fall within the scope of protection of the present invention.
Claims (10)
1. a kind of processing method for vehicle-mounted conversational system, which is characterized in that the treating method comprises following steps:
Multiple processing threads are identified according to voice request data；
The multiple processing thread parallel is set to execute respective operation；
According to the implementing result of the multiple processing thread, the processing result of the voice request data is determined.
2. processing method according to claim 1, which is characterized in that if currently processed thread includes multiple sub threads,
Before the step of " according to the implementing result of the multiple processing thread, determining the processing result of the voice request data ", institute
State method further include:
Multiple sub threads are identified according to the voice request data；
The multiple sub thread is set to execute respective operation parallel；
According to the implementing result of the multiple sub thread, the implementing result of the currently processed thread is determined.
3. processing method according to claim 2, which is characterized in that if the implementing result of the current sub thread includes more
A option, " according to the implementing result of the multiple sub thread, determining the implementing result of the currently processed thread " the step of
Later, the method also includes:
Data are confirmed according to the multiple option and user's voice currently entered, determine the execution knot of the currently processed thread
Fruit.
4. processing method according to claim 1, which is characterized in that " identified according to the voice request data multiple
Processing thread " the step of include:
Obtain the voice keyword of the voice request data；
Based on the corresponding relationship between preset keyword and multiple processing threads, institute corresponding with the voice keyword is matched
There is processing thread.
5. processing method according to claim 2, which is characterized in that " identified according to the voice request data multiple
The step of sub thread " includes:
Semantic analysis is carried out to the voice request data；
According to the semantic analysis as a result, determining relevant all sub threads.
6. a kind of processing system for vehicle-mounted conversational system, which is characterized in that the processing system includes
First identification module is used to identify multiple processing threads according to voice request data；
First processing module is used to that the multiple processing thread parallel to be made to execute respective operation；
First decision-making module is used to determine the voice request data according to the implementing result of the multiple processing thread
Processing result.
7. processing system according to claim 6, which is characterized in that if the currently processed thread includes multiple sub-lines
Journey, the processing system further include
Second identification module is used to identify multiple sub threads according to the voice request data；
Second processing module is used to that the multiple sub thread to be made to execute respective operation parallel；
Second decision-making module is used for the implementing result according to the multiple sub thread, determines holding for the currently processed thread
Row result.
8. processing system according to claim 7, which is characterized in that the processing system further includes state transition module,
If the implementing result of the current sub thread includes multiple options, the state transition module is used for according to the multiple choosing
Item confirms data with user's voice currently entered, determines the implementing result of the currently processed thread.
9. processing system according to claim 6, which is characterized in that first identification module is configured with disaggregated model,
It is used to match corresponding with the voice keyword based on the corresponding relationship between preset keyword and multiple processing threads
All processing threads.
10. processing system according to claim 7, which is characterized in that second identification module is configured with semantic understanding
Model is used to carry out the voice request data semantic analysis and according to the semantic analysis as a result, determination is relevant
Sub thread.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811239635.1A CN109634692A (en) | 2018-10-23 | 2018-10-23 | Vehicle-mounted conversational system and processing method and system for it |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811239635.1A CN109634692A (en) | 2018-10-23 | 2018-10-23 | Vehicle-mounted conversational system and processing method and system for it |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109634692A true CN109634692A (en) | 2019-04-16 |
Family
ID=66066627
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811239635.1A Pending CN109634692A (en) | 2018-10-23 | 2018-10-23 | Vehicle-mounted conversational system and processing method and system for it |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109634692A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110335621A (en) * | 2019-05-28 | 2019-10-15 | 深圳追一科技有限公司 | Method, system and the relevant device of audio processing |
| CN111145751A (en) * | 2019-12-31 | 2020-05-12 | 百度在线网络技术（北京）有限公司 | Audio signal processing method and device and electronic equipment |
| CN111312254A (en) * | 2020-03-26 | 2020-06-19 | 镁佳(北京)科技有限公司 | Voice conversation method and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020133346A1 (en) * | 2001-03-16 | 2002-09-19 | International Business Machines Corporation | Method for processing initially recognized speech in a speech recognition session |
| US20100030560A1 (en) * | 2006-03-23 | 2010-02-04 | Nec Corporation | Speech recognition system, speech recognition method, and speech recognition program |
| US20110161077A1 (en) * | 2009-12-31 | 2011-06-30 | Bielby Gregory J | Method and system for processing multiple speech recognition results from a single utterance |
| US20150199965A1 (en) * | 2014-01-16 | 2015-07-16 | CloudCar Inc. | System and method for recognition and automatic correction of voice commands |
| US20170011742A1 (en) * | 2014-03-31 | 2017-01-12 | Mitsubishi Electric Corporation | Device and method for understanding user intent |
| CN107644642A (en) * | 2017-09-20 | 2018-01-30 | 广东欧珀移动通信有限公司 | Method for recognizing semantics, device, storage medium and electronic equipment |
| CN107665706A (en) * | 2016-07-29 | 2018-02-06 | 科大讯飞股份有限公司 | Rapid Speech exchange method and system |
| CN108288468A (en) * | 2017-06-29 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Audio recognition method and device |
-
2018
- 2018-10-23 CN CN201811239635.1A patent/CN109634692A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020133346A1 (en) * | 2001-03-16 | 2002-09-19 | International Business Machines Corporation | Method for processing initially recognized speech in a speech recognition session |
| US20100030560A1 (en) * | 2006-03-23 | 2010-02-04 | Nec Corporation | Speech recognition system, speech recognition method, and speech recognition program |
| US20110161077A1 (en) * | 2009-12-31 | 2011-06-30 | Bielby Gregory J | Method and system for processing multiple speech recognition results from a single utterance |
| US20150199965A1 (en) * | 2014-01-16 | 2015-07-16 | CloudCar Inc. | System and method for recognition and automatic correction of voice commands |
| US20170011742A1 (en) * | 2014-03-31 | 2017-01-12 | Mitsubishi Electric Corporation | Device and method for understanding user intent |
| CN107665706A (en) * | 2016-07-29 | 2018-02-06 | 科大讯飞股份有限公司 | Rapid Speech exchange method and system |
| CN108288468A (en) * | 2017-06-29 | 2018-07-17 | 腾讯科技（深圳）有限公司 | Audio recognition method and device |
| CN107644642A (en) * | 2017-09-20 | 2018-01-30 | 广东欧珀移动通信有限公司 | Method for recognizing semantics, device, storage medium and electronic equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110335621A (en) * | 2019-05-28 | 2019-10-15 | 深圳追一科技有限公司 | Method, system and the relevant device of audio processing |
| CN111145751A (en) * | 2019-12-31 | 2020-05-12 | 百度在线网络技术（北京）有限公司 | Audio signal processing method and device and electronic equipment |
| CN111312254A (en) * | 2020-03-26 | 2020-06-19 | 镁佳(北京)科技有限公司 | Voice conversation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109634692A (en) | Vehicle-mounted conversational system and processing method and system for it | |
| US11392647B2 (en) | Intent-based question suggestion engine to advance a transaction conducted via a chatbot | |
| US9865264B2 (en) | Selective speech recognition for chat and digital personal assistant systems | |
| CN110659360A (en) | Man-machine conversation method, device and system | |
| EP4224468A2 (en) | Task initiation using long-tail voice commands | |
| US9277004B2 (en) | Prediction of network path quality among peer networking devices | |
| CN108255934A (en) | A kind of sound control method and device | |
| JP2003501754A (en) | Expression and reasoning of goal-oriented conversation | |
| CN113421561B (en) | Voice control method, voice control device, server, and storage medium | |
| CN111813900B (en) | Multi-round dialogue processing method and device, electronic equipment and storage medium | |
| CN111831795B (en) | Multi-round dialogue processing method and device, electronic equipment and storage medium | |
| RU2711104C2 (en) | Method and computer device for determining intention associated with request to create intent-depending response | |
| CN111460120A (en) | Conversation management method, device, equipment and storage medium | |
| Wang et al. | An effective iterated greedy algorithm for online route planning problem | |
| CN112100339A (en) | User intention recognition method and device for intelligent voice robot and electronic equipment | |
| CN106255950B (en) | User interface system, user interface control device and user interface control method | |
| CN105869631B (en) | The method and apparatus of voice prediction | |
| US11797769B1 (en) | Artificial intelligence system using hybrid technique for task-oriented dialog management | |
| US11501754B2 (en) | Specifying trip destinations from spoken dialogs | |
| CN113139044A (en) | Question-answering multi-turn dialogue method supporting multi-intention switching for question-answering system | |
| Freeman et al. | Online proactive escalation in multi-modal automated assistants | |
| CN117370520A (en) | Method, device, equipment and medium for processing split dialogue | |
| CN109841207A (en) | A kind of exchange method and robot, server and storage medium | |
| CN105955698B (en) | Voice control method and device | |
| CN111089603B (en) | Navigation information prompting method based on social application communication content and vehicle |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200902 Address after: Susong Road West and Shenzhen Road North, Hefei Economic and Technological Development Zone, Anhui Province Applicant after: Weilai (Anhui) Holding Co.,Ltd. Address before: 30 Floor of Yihe Building, No. 1 Kangle Plaza, Central, Hong Kong, China Applicant before: NIO NEXTEV Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190416 |