CN104240700B - A kind of global voice interactive method and system towards vehicle-mounted terminal equipment - Google Patents
A kind of global voice interactive method and system towards vehicle-mounted terminal equipment Download PDFInfo
- Publication number
- CN104240700B CN104240700B CN201410425978.2A CN201410425978A CN104240700B CN 104240700 B CN104240700 B CN 104240700B CN 201410425978 A CN201410425978 A CN 201410425978A CN 104240700 B CN104240700 B CN 104240700B
- Authority
- CN
- China
- Prior art keywords
- module
- result
- test
- content direct
- optimal result
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a kind of global voice interactive method towards vehicle-mounted terminal equipment, includes the following steps：(1) natural-sounding of user is acquired；(2) natural-sounding is converted into text, and semantic analysis is carried out to text；(3) if semantic analysis result is through order, order is directly executed；If it is not, then going to step (4)；(4) it is scanned for according to semantic analysis result, and selects one of them as optimal result from search result；(5) parameter testing is carried out to optimal result, obtains matching result；(6) matching result is transmitted to corresponding application component or function module, executes order.Using the present invention, the whole operation process of driver only needs one-touch and a natural language that input can be completed, and corresponding feedback is also reported in the form of speech, and whole process coordinates without trick, greatly improves drive safety and operation ease.
Description
Technical field
The present invention relates to a kind of man-machine interaction method more particularly to a kind of global interactive voices towards vehicle-mounted terminal equipment
Method, while being related to the global voice interactive system for realizing the above method, belong to vehicle-mounted human-computer interaction technique field.
Background technology
During car steering, driver need to pass through button, knob and touch screen etc. if necessary to use navigation feature
Pattern completion navigation starting, place input, selection and confirmation operation.Similarly, if necessary to use music or radio reception
Machine function, driver equally need to realize that opening, broadcasting and the switching of music or broadcast are selected by button, knob and touch screen
The operations such as select.During aforesaid operations, driver needs trick cooperation to complete aforesaid operations.And driver is above-mentioned in progress
When operation, hand has left steering wheel, while the sight of eyes has left the front of vehicle, and then pays close attention to interior operating position.
It can cause greatly to drive risk in this way.
To solve the above-mentioned problems, people have carried out various technology explorations.Such as notification number be 202092650U
Chinese utility model patent in, disclose a kind of vehicle mounted multimedia of the Voice Navigation with button.It is public in the patent document
Following operating procedure is opened：A. speech recognition controlled button is pressed, voice conversion module receives phonetic order；B. speech recognition
Module voice command recognition；C. instruct conversion module by the control instruction and ground of phonetic order and the navigation module after identification
Location instruction is associated with one by one；D. the control instruction of navigation module and address instruction are handed down to navigation module by master controller；E. it navigates
Module starts to navigate.The utility model can make the eyes of driver not have to stare at screen, realize safe driving, safety navigation
Function, and it is easy for installation.But user needs the training for carrying out certain time to be made to user to be familiar with various control commands
At great inconvenience.
In the Chinese patent application of Publication No. 103295572A, a kind of audio recognition method is disclosed, including as follows
Step：Whether A, system is detected needs that voice signal is identified, if it is desired, then turns to step B, otherwise, continues
Detection；B, voice audio signals are obtained, denoising and amplification are carried out to audio signal；C, audio signal is identified, and to knowing
Other result is read out and judges into line command word class, if it is display class order word, turns to step D, otherwise turns to step
E；D, result is carried out to show and voice broadcast and terminate this secondary control end of identification；E, then judge into line command word rank, if
It is father's control command, turns to step F, step G is then turned to if it is leaf control command；F, voice broadcast prompt and Speech Record are carried out
Enter, is transferred to step B；G, performing leaf control instruction is transferred to step D.This method is for overcoming navigation chip processing capacity finite sum
The jejune disadvantage of speech recognition can significantly improve the efficiency and effect of speech recognition.But this method is not directed to semanteme
The technical solutions such as through are ordered in analysis, and there are certain defects on practicability.
Invention content
In view of the deficiencies of the prior art, primary technical problem to be solved by this invention is to provide one kind towards vehicle-mounted end
The global voice interactive method of end equipment.
Another technical problem to be solved by this invention is to provide for a kind of global voice for realizing the above method
Interactive system.
For achieving the above object, the present invention uses following technical solutions：
A kind of global voice interactive method towards vehicle-mounted terminal equipment, includes the following steps：
(1) natural-sounding of user is acquired；
(2) natural-sounding is converted into text, and semantic analysis is carried out to the text；
(3) if the semantic analysis result is through order, order is directly executed；If it is not, then going to step (4)；
(4) it is scanned for according to the semantic analysis result, and selects one of them as optimal knot from search result
Fruit；
(5) parameter testing is carried out to the optimal result, obtains matching result；
(6) matching result is transmitted to corresponding application component or function module, executes order.
Wherein more preferably, the step (2) further comprises the steps：
(21) it when receiving the character stream of text, is identified one by one from the character stream and extracts Text Command
String, and in both the buffers by Text Command string deposit；
(22) a pending Text Command string is taken out from the buffer, and the pending text is ordered
String is enabled to carry out morphological analysis, to obtain the order vocabulary；
(23) syntactic analysis is carried out to each order vocabulary in the order vocabulary, to obtain syntactic analysis tree；
(24) the syntactic analysis tree is traversed, the corresponding semantic attribute of each node is found, to obtain the text text
The syntactic structure body of word.
Wherein more preferably, in the step (5), parameter testing includes motion test, classification test and data test.
Wherein more preferably, after the parameter testing, return a relevant parameter, the relevant parameter be used for identify with
The corresponding application component of the optimal result or function module.
A kind of global voice interactive system towards vehicle-mounted terminal equipment, for realizing above-mentioned global interactive voice side
Method, including：Voice acquisition module, sound identification module, semantic meaning analysis module and content direct module, wherein
The voice acquisition module is connected with the sound identification module, described semantic meaning analysis module one end and institute's predicate
Sound identification module is connected, and the other end is connected with the content direct module.
Wherein more preferably, the content direct module carries out parameter testing, including motion test, classification test and data are surveyed
Examination.
In the motion test, the content direct module includes an action lists, which includes multiple
Action, and cannot be sky；The action of content direct module matching optimal result, and it is all dynamic in the action lists
Compare one by one, if the two action is identical, test passes through；If action lists are sky, test and do not pass through；If optimal
As a result not required movement, and the action lists of content direct module are not sky, then and test passes through.
In classification test, if each type of optimal result and the content direct mould some class in the block
Type matches, and matching process is identical as motion test, then the optimal result is tested by classification.
In the data test, each search result specifies an exclusive identification code and a data type,
The exclusive identification code is partly each different attribute；When in optimal result exclusive identification code and the content direct module
In when being compared, the part only and in the content direct module having been defined for is compared.
Wherein more preferably, after the content direct module terminates parameter testing, a relevant parameter, the corresponding ginseng are returned
Number is used for identifying application component corresponding with the optimal result or function module；The relevant parameter is transmitted to correspondence
Application component or function module, execute order.
Compared with prior art, the invention has the advantages that：
1. simple operation：All operations of the driver in driving procedure, it is only necessary to which user presses that " one is pressed on the steering wheel
Corresponding operation can be completed in key " combination " a word " (natural language).
2. supporting full natural language：It only needs driver to say according to conventional natural grammer what to need to do behaviour can be completed
Make, the restriction order any without learning and memory.
Description of the drawings
Fig. 1 is the step schematic diagram of global voice interactive method provided by the present invention；
Fig. 2 is the work flow diagram of global voice interactive method provided by the present invention；
Fig. 3 is the structure diagram of global voice interactive system provided by the present invention.
Specific implementation mode
The technology contents of the present invention are described in further detail in the following with reference to the drawings and specific embodiments.
Traditional vehicle device interworking flow is：The first step, user using button or screen touch screen click navigation (or its
He applies)；Second step, user click the input frame in application；Third walks, and user inputs target using modes such as phonetic, strokes
Place name；4th step clicks search button；5th step, user select destination or departure place；6th step, Yong Huxuan
The path of navigation is selected, navigation starts.During entire, user needs to touch or click screen 10 times or more, it is contemplated that behaviour
It is 2 minutes or so to make the time.Since the operations such as the necessary trick cooperation button of whole operation process, click, knob, stroke could be complete
At information input, great driving dangerousness will be led to, so user is required to parking in roadside collection when using traditional navigation
Middle attention is operated.
For disadvantages mentioned above, the invention discloses a kind of global voice interactive method towards vehicle-mounted terminal equipment, Fig. 1 institutes
Show be this overall situation voice interactive method work step, specifically include：
(1) voice of user is acquired；
(2) voice is converted into text, and semantic analysis is carried out to the text；
(3) if the semantic analysis result is through order, order is directly executed；If it is not, then going to step (4)；
(4) it is scanned for according to the semantic analysis result, and selects one of them as optimal knot from search result
Fruit；
(5) parameter testing is carried out to the optimal result, obtains matching result；
(6) matching result is transmitted to corresponding application program, order is executed by application program.
Fig. 2 is the workflow of this overall situation voice interactive method.User clicks voice key at any time or in interface
(the voice control key on dedicated voice button or steering wheel in system), the thing that user wishes to do is said with natural language
Feelings.Global voice interactive system by carrying out speech recognition after sound collection, noise reduction filtering, then to the obtained text of identification into
Row semantic analysis judges the intention of user according to semantic analysis result, identifies corresponding order, and then calls corresponding application
The interface of program or system function module obtains corresponding information, finally realizes customer objective.
Wherein, the dissection process process of text includes the following steps：
1. when receiving the character stream of text, above-mentioned Text Command string is identified and extracted one by one from character stream,
And it is deposited in both the buffers；
2. taking out a pending Text Command string from buffer, according to the method for Text Command string rule, treat
The Text Command string of processing carries out morphological analysis, to obtain order vocabulary；
3. each order vocabulary in pair order vocabulary carries out syntactic analysis, to obtain syntactic analysis tree；
4. traversing syntactic analysis tree, the corresponding semantic attribute of each node is found, to obtain User Defined structure.
Wherein, character stream includes other Protocol Control Informations.Following processing is executed in step 2：To pending text
Text strings are scanned character by character；When encountering the character less than identification, determines that morphological analysis fails, be transferred to text
Parse failure handling；And when successfully identifying each character, order vocabulary is obtained.
Following processing is executed in step 3：It is migrated come driving condition according to order vocabulary；And work as order vocabulary
Sky, and when syntax analyzer is in final state, determines syntactic analysis success, obtain syntactic analysis tree, otherwise, syntactic analysis
Failure is transferred to text parsing failure handling.
Following processing is executed in step 4：Preorder traversal is carried out to syntactic analysis tree；When being accessed the one of syntactic analysis tree
When a node, query semantics attribute list finds corresponding semantic attribute；If not finding corresponding semantic attribute, then it represents that
Text word be not present or the node at described in keyword be not present, be transferred to text parsing failure handling；Such as
Fruit finds corresponding semantic attribute, then according to the indication structure User Defined structure of semantic attribute.
Wherein, when finding corresponding semantic attribute, following operation is executed：Check each essential sub- category of the semantic attribute
Property whether all exist, in the absence of whether have default value available；Judged whether according to the storage class of the attribute and data type
Apply for that new there are spaces, and the much memory spaces of application；Position offset of the attribute in parent attribute memory space；With
And according to these instruction above-mentioned User Defined structures of information structuring.
Semantic attribute table is that custom data structure is Hash table, and hash key is attribute-name.Semanteme belongs to
Property describes the semantic feature of a command keyword of a text, and semantic attribute includes at least：Attribute-name, storage class
Type, data type and storage location, the information for being used to indicate the keyword that how handle text indicate.
Judged according to semantic analysis and analysis result, if user say be not effective order if it is anti-to user
Present relevant information；If what user said is effective order, continue to judge whether the semantic analysis with analysis result is direct
Order.If it is direct command, then global voice interactive system directly executes related command；If not direct command, then will
The semantic analysis and analysis result are sent to corresponding application component either in function module and by the application component or function
The searching interface of module retrieves semantic analysis result as keyword, search result can have it is multiple, one or not
It was found that.When not finding search result, user is prompted not find search result；When only there are one search result when, put it into
Content direct module；When there is multiple search results, allow user select one as a result, being set as default behavior and selecting automatically
It selects one and is used as optimal result, and the optimal result is put into content direct module.
Content direct module carries out parameter matching, including motion test, classification test and data to receiving optimal result
Test.Each specific test process is introduced in detail below.
First, content direct module carries out motion test to optimal result, and content direct module includes an action lists,
The action lists include multiple actions, and cannot be sky, otherwise all actions will be organized to export.Content direct module matches
The action of optimal result, optimal result can only specify an action, with the everything in action lists one by one compared with, if two
Person's action is identical, then test passes through；List is empty for content direct module action, then motion test does not pass through；Optimal result does not refer to
Fixed action, and the action lists of content direct module are not sky, then and test passes through.
Again, content direct module carries out classification test to optimal result, if each type and content of optimal result
Through mould some type matching in the block, matching process is identical as motion test, then the optimal result is tested by classification.
Finally, content direct module carries out data test to optimal result, each search result specifies one only
One identification code and a data type.Each of exclusive identification code part is different attribute, including：scheme、host、
port、path.These attributes are all optional, but not mutual indepedent.When in optimal result exclusive identification code and content it is straight
When being compared up to mould corresponding content in the block, the part only and in content direct module having been defined for is compared.For example,
If scheme defined in content direct module, exclusive identification code include all search result objects of the scheme all
By test, for path, asterisk wildcard can be used to match.The data of optimal result also specify data type, usually
In the case of, " * " asterisk wildcard can be used as subtype.
If optimal result, by the matching treatment of content direct module, content direct module can feed back a corresponding ginseng
Number data, the supplemental characteristic are used for determining the application component or function module belonging to above-mentioned optimal result, content direct mould
The supplemental characteristic is submitted to affiliated application component to block or function module executes user command, and it is aobvious to start related function module
Show and broadcasting content.If without effective content, to user feedback relevant information.
The present invention also provides a kind of global voice interactive system towards vehicle-mounted terminal equipment, for realizing above-mentioned complete
Office's voice interactive method.Fig. 3 shows the overall structure of this overall situation voice interactive system.The overall situation voice interactive system includes：
Voice acquisition module, sound identification module, semantic meaning analysis module and content direct module.Voice module acquires user after being activated
Voice after acquisition preserve while sending sound identification module to；Sound identification module analyzes user speech
Processing, is converted into the form of text and sends semantic meaning analysis module to；Semantic meaning analysis module carries out grammer to text
Analysis obtains semantic analysis result；Judge whether semantic analysis result is through order, if it is passes directly to answer accordingly
It is executed with component, if not through order, then the semantic analysis result is transmitted to content direct module, content direct module
Parameter testing is carried out, the application component or function module to match with semantic analysis result is found, executes user command, navigation
Start.
The present invention solves user every using traditional vehicle device by steering wheel control button and global voice interactive system
When function, due to needing trick to coordinate, cause user's sight leave road ahead, user hand departure direction disk, and cause
Potential driving safety problem.Using the present invention, the whole operation process of driver only needs one-touch and a natural language
Input can be completed, feedback is reported in the form of speech, and whole process coordinates without trick, and original cumbersome time-consuming operation is made to become
It must be very natural simplicity, the used time was maintained within 5~10 seconds, greatly improved drive safety and operation ease.
The global voice interactive method to provided by the present invention towards vehicle-mounted terminal equipment and system have carried out in detail above
Thin explanation.To those skilled in the art, it is done under the premise of without departing substantially from true spirit any
Obvious change will all constitute to infringement of patent right of the present invention, will undertake corresponding legal liabilities.
Claims (5)
1. a kind of global voice interactive method towards vehicle-mounted terminal equipment, it is characterised in that include the following steps：
(1) natural-sounding of user is acquired；
(2) natural-sounding is converted into text, and semantic analysis is carried out to the text；
(3) if the semantic analysis result is through order, order is directly executed；If it is not, then going to step (4)；
(4) it is scanned for according to the semantic analysis result, and selects one of them as optimal result from search result；
(5) parameter testing is carried out to the optimal result, obtains matching result；Parameter testing includes motion test, classification test
And data test；In motion test, content direct module includes an action lists, which includes multiple actions,
And cannot be sky；Content direct module matches the action of optimal result, compares one by one with the everything in the action lists
Compared with if the two action is identical, test passes through；If action lists are sky, test and do not pass through；If optimal result does not refer to
Fixed action, and the action lists of content direct module are not sky, then and test passes through；In classification test, if optimal result
Each type and content direct mould some type matching in the block, matching process is identical as motion test, then this is optimal
As a result it is tested by classification；In data test, each search result specifies an exclusive identification code and a data
Type, exclusive identification code are partly each different attributes；When in the exclusive identification code and content direct module in optimal result
When being compared, the part only and in content direct module having been defined for is compared；
(6) matching result is transmitted to corresponding application component or function module, executes order.
2. overall situation voice interactive method as described in claim 1, which is characterized in that
The step (2) further comprises the steps：
(21) when receiving the character stream of text, identify and extract Text Command string one by one from the character stream, and
In both the buffers by Text Command string deposit；
(22) a pending Text Command string is taken out from the buffer, to the pending Text Command string
Morphological analysis is carried out, to obtain the order vocabulary；
(23) syntactic analysis is carried out to each order vocabulary in the order vocabulary, to obtain syntactic analysis tree；
(24) the syntactic analysis tree is traversed, the corresponding semantic attribute of each node is found, to obtain the text
Syntactic structure body.
3. overall situation voice interactive method as described in claim 1, it is characterised in that：
After the parameter testing, a relevant parameter is returned to, the relevant parameter is used for identifying and the optimal result phase
Corresponding application component or function module.
4. a kind of global voice interactive system towards vehicle-mounted terminal equipment, for realizing any one of claims 1 to 3 institute
The global voice interactive method stated, it is characterised in that including：Voice acquisition module, sound identification module, semantic meaning analysis module and
Content direct module, wherein
The voice acquisition module is connected with the sound identification module, and described semantic meaning analysis module one end is known with the voice
Other module is connected, and the other end is connected with the content direct module；
The content direct module carries out parameter testing, including motion test, classification test and data test；It is surveyed in the action
In examination, the content direct module includes an action lists, which includes multiple actions, and cannot be sky；Institute
State content direct module matching optimal result action, with the everything in the action lists one by one compared with, if the two
Act it is identical, then test pass through；If action lists are sky, test and do not pass through；If optimal result not required movement, and
The action lists of content direct module are not sky, then test passes through；In classification test, if each class of optimal result
Type and the content direct mould some type matching in the block, matching process is identical as motion test, then the optimal result
It is tested by classification；In the data test, each search result specifies an exclusive identification code and a data
Type, the exclusive identification code are partly each different attributes；When in optimal result exclusive identification code and the content it is straight
Up to when being compared in module, the part only and in the content direct module having been defined for is compared.
5. overall situation voice interactive system as claimed in claim 4, it is characterised in that：
After the content direct module terminates parameter testing, a relevant parameter is returned to, the relevant parameter is used for identifying and institute
State the corresponding application component of optimal result or function module；By the relevant parameter be transmitted to corresponding application component or
Function module executes order.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410425978.2A CN104240700B (en) | 2014-08-26 | 2014-08-26 | A kind of global voice interactive method and system towards vehicle-mounted terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410425978.2A CN104240700B (en) | 2014-08-26 | 2014-08-26 | A kind of global voice interactive method and system towards vehicle-mounted terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104240700A CN104240700A (en) | 2014-12-24 |
| CN104240700B true CN104240700B (en) | 2018-09-07 |
Family
ID=52228657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410425978.2A Expired - Fee Related CN104240700B (en) | 2014-08-26 | 2014-08-26 | A kind of global voice interactive method and system towards vehicle-mounted terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104240700B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104866612B (en) * | 2015-06-06 | 2016-04-27 | 河南职业技术学院 | A kind of searching method obtaining music file from internet |
| CN106372091A (en) * | 2015-07-23 | 2017-02-01 | 中兴通讯股份有限公司 | Method and device for providing solution on the basis of information identification |
| KR101736109B1 (en) * | 2015-08-20 | 2017-05-16 | 현대자동차주식회사 | Speech recognition apparatus, vehicle having the same, and method for controlling thereof |
| CN105206266B (en) * | 2015-09-01 | 2018-09-11 | 重庆长安汽车股份有限公司 | Vehicle-mounted voice control system and method based on user view conjecture |
| CN106205604B (en) * | 2016-07-05 | 2020-07-07 | 惠州市德赛西威汽车电子股份有限公司 | Application-side voice recognition evaluation system and method |
| CN106297782A (en) * | 2016-07-28 | 2017-01-04 | 北京智能管家科技有限公司 | A kind of man-machine interaction method and system |
| CN108121722A (en) * | 2016-11-28 | 2018-06-05 | 渡鸦科技（北京）有限责任公司 | The construction method and device of knowledge base |
| CN106604208A (en) * | 2016-12-06 | 2017-04-26 | 深圳市京华信息技术有限公司 | Method and system for automatically starting application program and conducting voice interaction |
| CN108428450B (en) * | 2017-02-15 | 2021-10-22 | 斑马智行网络(香港)有限公司 | Operation instruction processing method and device |
| CN107146610B (en) * | 2017-04-10 | 2021-06-15 | 易视星空科技无锡有限公司 | Method and device for determining user intention |
| CN107273227A (en) * | 2017-06-30 | 2017-10-20 | 百度在线网络技术（北京）有限公司 | Using communication means, device, equipment and storage medium |
| CN108776490A (en) * | 2018-04-23 | 2018-11-09 | 西安电子科技大学 | A kind of aircraft script control method based on semantic analysis |
| CN109284505A (en) * | 2018-11-07 | 2019-01-29 | 江苏中润普达信息技术有限公司 | It is a kind of for vehicle-mounted natural language semantic analysis |
| CN112435660A (en) * | 2019-08-08 | 2021-03-02 | 上海博泰悦臻电子设备制造有限公司 | Vehicle control method and system and vehicle |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7398209B2 (en) * | 2002-06-03 | 2008-07-08 | Voicebox Technologies, Inc. | Systems and methods for responding to natural language speech utterance |
| US9318108B2 (en) * | 2010-01-18 | 2016-04-19 | Apple Inc. | Intelligent automated assistant |
| CN101178705A (en) * | 2007-12-13 | 2008-05-14 | 中国电信股份有限公司 | Free-running speech comprehend method and man-machine interactive intelligent system |
| CN102013254A (en) * | 2010-11-17 | 2011-04-13 | 广东中大讯通信息有限公司 | Man-machine interactive system and method for digital television voice recognition |
| CN102629246B (en) * | 2012-02-10 | 2017-06-27 | 百纳（武汉）信息技术有限公司 | Recognize the server and browser voice command identification method of browser voice command |
| CN102750366B (en) * | 2012-06-18 | 2015-05-27 | 海信集团有限公司 | Video search system and method based on natural interactive import and video search server |
| CN102968409B (en) * | 2012-11-23 | 2015-09-09 | 海信集团有限公司 | Intelligent human-machine interaction semantic analysis and interactive system |
| CN103700369B (en) * | 2013-11-26 | 2016-08-31 | 科大讯飞股份有限公司 | Phonetic navigation method and system |
| CN103730116B (en) * | 2014-01-07 | 2016-08-17 | 苏州思必驰信息科技有限公司 | Intelligent watch realizes the system and method that intelligent home device controls |
-
2014
- 2014-08-26 CN CN201410425978.2A patent/CN104240700B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN104240700A (en) | 2014-12-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104240700B (en) | A kind of global voice interactive method and system towards vehicle-mounted terminal equipment | |
| CN107657950B (en) | Automobile voice control method, system and device based on cloud and multi-command words | |
| US10922322B2 (en) | Systems and methods for speech-based searching of content repositories | |
| WO2022001013A1 (en) | Voice interaction method, vehicle, server, system, and storage medium | |
| US7725322B2 (en) | Spoken dialogue interface apparatus and method | |
| CN106202476B (en) | A kind of interactive method and device of knowledge based collection of illustrative plates | |
| US7917368B2 (en) | Method for interacting with users of speech recognition systems | |
| US9454958B2 (en) | Exploiting heterogeneous data in deep neural network-based speech recognition systems | |
| US8788267B2 (en) | Multi-purpose contextual control | |
| DE112020004504T5 (en) | Account connection with device | |
| DE112016004863T5 (en) | Parameter collection and automatic dialog generation in dialog systems | |
| CN106941000A (en) | Voice interactive method and device based on artificial intelligence | |
| CN106985137A (en) | Multi-modal exchange method and system for intelligent robot | |
| WO2022057152A1 (en) | Voice interaction method, server, and computer-readable storage medium | |
| TW200847004A (en) | Speech-centric multimodal user interface design in mobile technology | |
| CN108388553B (en) | Method for eliminating ambiguity in conversation, electronic equipment and kitchen-oriented conversation system | |
| CN101383150B (en) | Control method of speech soft switch and its application in geographic information system | |
| US20200005051A1 (en) | Visual Perception Method, Apparatus, Device, and Medium Based on an Autonomous Vehicle | |
| US11069351B1 (en) | Vehicle voice user interface | |
| US20110060588A1 (en) | Method and System for Automatic Speech Recognition with Multiple Contexts | |
| CN103246648B (en) | Phonetic entry control method and device | |
| CN115457959A (en) | Voice interaction method, server and computer readable storage medium | |
| CN108062214A (en) | The methods of exhibiting and device of a kind of search interface | |
| CN105808688A (en) | Complementation retrieval method and device based on artificial intelligence | |
| CN115457960B (en) | Voice interaction method, server and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180907 Termination date: 20210826 |