CN109410983A - Voice question searching method and system - Google Patents
Voice question searching method and system Download PDFInfo
- Publication number
- CN109410983A CN109410983A CN201811410443.2A CN201811410443A CN109410983A CN 109410983 A CN109410983 A CN 109410983A CN 201811410443 A CN201811410443 A CN 201811410443A CN 109410983 A CN109410983 A CN 109410983A
- Authority
- CN
- China
- Prior art keywords
- facility
- study
- topic
- voice
- voice messaging
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000001133 acceleration Effects 0.000 claims description 69
- 230000004913 activation Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 abstract description 36
- 238000012545 processing Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 8
- 230000005484 gravity Effects 0.000 description 8
- 230000000694 effects Effects 0.000 description 7
- 238000012937 correction Methods 0.000 description 6
- 230000035772 mutation Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000009987 spinning Methods 0.000 description 2
- 230000009747 swallowing Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/54—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for retrieval
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B7/00—Electrically-operated teaching apparatus or devices working with questions and answers
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
- H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/225—Feedback of the input speech
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Signal Processing (AREA)
- Computational Linguistics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the technical field of data processing, and discloses a voice question searching method and a system, wherein the method comprises the following steps: the learning equipment starts a voice question searching function according to a wake-up signal sent by a user; the learning equipment judges whether the learning equipment is in a moving state; when the learning equipment judges that the learning equipment is in a moving state, the learning equipment switches the current sound pickup mode into a directional sound pickup mode; the learning equipment collects voice information sent by the user through the directional pickup mode; and the learning equipment executes corresponding question searching operation according to the voice information. According to the method and the device, when the learning equipment is judged to be in the moving state, the current pickup mode is switched to the directional pickup mode, and the voice information sent by the user is collected through the directional pickup mode, so that the problem of low voice recognition accuracy rate caused by the fact that the user moves to a noisy environment can be solved, the voice problem searching efficiency is improved, and the use experience of the user is improved.
Description
Technical field
The invention belongs to technical field of data processing, in particular to a kind of voice searches topic method and system.
Background technique
With the rapid development of intelligent terminal and network technology, people are more and more habitually completed using intelligent terminal various
Demand such as carries out mobile learning using intelligent terminal.With use intelligent terminal carry out mobile learning mode increasingly by people
Paid attention to, mobile learning realizes a kind of important way and hand of learning society by becoming as a kind of new mode of learning
Section.
When user, which holds facility for study, to move, the collected voice signal of facility for study is to own in ambient enviroment
The voice signal in direction, when user is moved in noisy environment, the recognition accuracy for being easy to cause voice to search topic function is low, shadow
It rings voice and searches topic efficiency, so that the experience sense of user is poor, be unfavorable for the popularization and use of product.
Summary of the invention
The object of the present invention is to provide a kind of voices to search topic method and system, moves judging that user and facility for study are in
When dynamic state, the voice signal that facility for study is issued by oriented acquisition user is caused to reduce due to being moved to noisy environment
Voice signal the low situation of recognition accuracy, improve the efficiency that voice searches topic.
Technical solution provided by the invention is as follows:
On the one hand, a kind of voice is provided and searches topic method, comprising:
The wake-up signal starting voice that facility for study is issued according to user searches topic function；
The facility for study judges whether itself is in moving condition；
When the facility for study determines that itself is in moving condition, the facility for study is by current pickup pattern switching
To orient pickup mode；
The voice messaging that the facility for study is issued by user described in the orientation pickup type collection；
The facility for study executes according to the voice messaging and searches topic operation accordingly.
It is further preferred that the facility for study judges whether itself specifically includes in moving condition:
The facility for study judges whether itself is in handheld state；
When the facility for study is in handheld state, the facility for study carries out position positioning to itself, obtains position
Positioning result；
The facility for study judges whether itself is in moving condition according to the position positioning result.
It is further preferred that the facility for study judges whether itself specifically includes in moving condition:
The facility for study judges whether itself is in handheld state；
When the facility for study is in handheld state, the facility for study obtains the acceleration information of itself；
The facility for study judges whether itself is in moving condition according to the acceleration information.
It is specifically included it is further preferred that the facility for study executes to search to inscribe to operate accordingly according to the voice messaging:
The facility for study parses the voice messaging, obtains the topic information that the voice messaging includes；
The facility for study executes according to the topic information and searches topic operation accordingly.
It is further preferred that the facility for study parses the voice messaging, the topic that the voice messaging is included is obtained
Mesh information specifically includes:
The voice messaging is identified as text information by the facility for study；
The facility for study carries out semantic parsing to the text information, obtains semantic parsing result；
The facility for study obtains the topic information that the voice messaging includes according to the semantic parsing result.
On the other hand, a kind of voice is also provided and searches topic system, comprising:
Topic function activation module is searched, the wake-up signal starting voice for issuing according to user searches topic function；
Moving condition judgment module, for judging whether itself is in moving condition；
Pickup mode switch module, for when the facility for study determines that itself is in moving condition, the study to be set
Standby by current pickup pattern switching is orientation pickup mode；
Voice messaging acquisition module, the voice messaging for being issued by user described in the orientation pickup type collection；
Topic execution module is searched, searches topic operation accordingly for executing according to the voice messaging.
It is further preferred that the moving condition judgment module includes:
Handheld state judging unit, for judging whether itself is in handheld state；
Positioning unit obtains position positioning knot for carrying out position positioning to itself when itself is in handheld state
Fruit；
Moving condition judging unit, for judging whether itself is in moving condition according to the position positioning result.
It is further preferred that the moving condition judgment module includes:
Handheld state judging unit, for judging whether itself is in handheld state；
Acceleration obtainment unit, for obtaining the acceleration information of itself when itself is in handheld state；
Moving condition judging unit, for judging whether itself is in moving condition according to the acceleration information.
It is further preferred that the topic execution module of searching includes:
Topic information acquiring unit obtains the topic information that the voice messaging includes for parsing the voice messaging；
Topic execution unit is searched, searches topic operation accordingly for executing according to the topic information.
It is further preferred that the topic information acquiring unit includes:
Subelement is identified, for the voice messaging to be identified as text information；
Semantic parsing subunit obtains semantic parsing result for carrying out semantic parsing to the text information；
Topic information obtains subelement, for obtaining the topic that the voice messaging includes according to the semantic parsing result
Mesh information.
Compared with prior art, a kind of voice provided by the invention is searched topic method and system and is had the advantages that
1, current pickup pattern switching is that orientation is picked up when determining that facility for study itself is in moving condition by the present invention
Sound mode, the voice messaging issued by orientation pickup type collection user, can reduce and make when user is moved to noisy environment
At the low problem of speech recognition accuracy, improve voice and search topic efficiency, improve the usage experience of user.
2, in a preferred embodiment, first judge whether facility for study is in handheld state, further according to the position of facility for study
It sets or the acceleration of facility for study judges whether facility for study is in moving condition, whether confirm by judging twice study
The accuracy rate of judgement can be improved in moving condition.
Detailed description of the invention
Below by clearly understandable mode, preferred embodiment is described with reference to the drawings, to a kind of voice search topic method and
Above-mentioned characteristic, technical characteristic, advantage and its implementation of system are further described.
Fig. 1 is the flow diagram for the first embodiment that a kind of voice of the present invention searches topic method；
Fig. 2 is the flow diagram for the second embodiment that a kind of voice of the present invention searches topic method；
Fig. 3 is the flow diagram for the 3rd embodiment that a kind of voice of the present invention searches topic method；
Fig. 4 is the flow diagram one for the fourth embodiment that a kind of voice of the present invention searches topic method；
Fig. 5 is the flow diagram two for the fourth embodiment that a kind of voice of the present invention searches topic method；
Fig. 6 is the structural schematic block diagram of one embodiment that a kind of voice of the present invention searches topic system；
Fig. 7 is the structural schematic block diagram for another embodiment that a kind of voice of the present invention searches topic system；
Fig. 8 is the structural schematic block diagram for the further embodiment that a kind of voice of the present invention searches topic system.
Drawing reference numeral explanation
100, topic function activation module is searched；200, moving condition judgment module；
210, handheld state judging unit；220, positioning unit；
230, moving condition judging unit；240, acceleration obtainment unit；
300, pickup mode switch module；400, voice messaging acquisition module；
500, topic execution module is searched；510, topic information acquiring unit；
511, subelement is identified；512, semantic parsing subunit；
513, topic information obtains subelement；520, topic execution unit is searched.
Specific embodiment
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, Detailed description of the invention will be compareed below
A specific embodiment of the invention.It should be evident that drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing, and obtain other embodiments.
To make simplified form, part related to the present invention is only schematically shown in each figure, they are not represented
Its practical structures as product.In addition, there is identical structure or function in some figures so that simplified form is easy to understand
Component only symbolically depicts one of those, or has only marked one of those.Herein, "one" is not only indicated
" only this ", can also indicate the situation of " more than one ".
The first embodiment provided according to the present invention, as shown in Figure 1, a kind of voice searches topic method, comprising:
The wake-up signal starting voice that S100 facility for study is issued according to user searches topic function；
Specifically, the starting method that voice searches topic function can be set on facility for study in advance, such as can be set as leading to
It crosses and opens a certain application program to start voice and search topic function, or start language by double-clicking a certain icon on facility for study
Sound searches topic function, or searches topic function by the special sound information that user inputs to start voice.Wherein, special sound information
It can include the related voice information of keyword " searching topic " for " searching topic ", " search topic ", " topic function is searched in unlatching " etc..
Facility for study described in S200 judges whether itself is in moving condition；
Specifically, after the wake-up signal starting voice issued according to user searches topic function, first judge whether itself is in shifting
Dynamic state, judges that itself whether being in moving condition can be judged by the acceleration of facility for study or position etc..
For S300 when the facility for study determines that itself is in moving condition, the facility for study is by current pickup mode
It is switched to orientation pickup mode；
Specifically, when judging that facility for study is in moving condition according to acceleration or position etc., facility for study will be certainly
The current pickup pattern switching of body is orientation pickup mode.The voice messaging in a direction can only be acquired by orienting pickup mode, and
Ignore the voice messaging in other directions, to inhibit the noise in other directions, improves the pickup effect of facility for study, and then improve language
Sound searches the accuracy rate of topic.
The voice messaging that facility for study described in S400 is issued by user described in the orientation pickup type collection；
Specifically, after opening orientation pickup mode, the direction that the pickup direction for orienting pickup mode can be mobile for user, i.e.,
The only voice messaging that acquisition user issues, without acquiring other sound in ambient enviroment, to inhibit the noise in other directions,
Improve the pickup effect of facility for study.
Facility for study described in S500 executes according to the voice messaging and searches topic operation accordingly.
Specifically, after by orientation pickup type collection to the voice messaging of user, according to the pass for including in voice messaging
Topic operation is searched in the execution such as keyword accordingly, to obtain the topic to match with voice messaging and answer explanation etc..
Current pickup pattern switching is orientation pickup when determining that facility for study itself is in moving condition by the present invention
Mode, the voice messaging issued by orientation pickup type collection user, can reduce and cause when user is moved to noisy environment
The low problem of speech recognition accuracy, improve voice and search topic efficiency, improve the usage experience of user.
The second embodiment provided according to the present invention, as shown in Fig. 2, a kind of voice searches topic method, comprising:
The wake-up signal starting voice that S100 facility for study is issued according to user searches topic function；
Facility for study described in S210 judges whether itself is in handheld state；
Specifically, judge facility for study whether be in handheld state can by facility for study be arranged gravity sensor, so
The component of acceleration for obtaining the gravity sensor acquisition on facility for study afterwards, then according to the component of acceleration on facility for study
Mutation, obtains the motion profile of facility for study, finally judges whether facility for study is in hand according to the motion profile of facility for study
Hold state.
For example, some plane of facility for study in the relative coordinate system of its own (in such as one minute) in a short time
On produce 38 degree of deflection, then facility for study knows facility for study according to the variation of the component of acceleration within the short time
38 degree have been deflected in some plane.
When knowing that facility for study deflects on some direction from stationary state according to the component of acceleration on facility for study
(deflecting in some plane) then proves that facility for study enters handheld state.
For S220 when the facility for study is in handheld state, the facility for study carries out position positioning to itself, obtains
Position positioning result；
Facility for study described in S230 judges whether itself is in moving condition according to the position positioning result；
Specifically, when determining that facility for study is in handheld state, then position positioning is carried out to facility for study, such as passes through position
It sets sensor to position facility for study, position sensor can experience the position of measured object and be converted into usable output signal
Sensor can get the location information of facility for study by the way that position sensor is arranged on facility for study, to set to study
It is standby to carry out position positioning.Then judge whether facility for study itself is in moving condition by positioning result.
For example, acquiring the location information of facility for study after determining that facility for study is in handheld state, facility for study is acquired
Location information when, can at interval of certain time acquire a facility for study location information, such as be adopted every 10s or 20s or 30s
The location information for collecting a facility for study collects facility for study after the location information of each sampling time point, adopts when each
When the location information difference at sample time point, that is, illustrate that user is in movement, can judge that facility for study is in movement at this time
State.
Or the signal strength by obtaining Wifi positions facility for study, then judges to learn according to positioning result
Whether equipment is in moving condition.
Judge whether facility for study is in moving condition except through the signal strength of position sensor or interior wifi
Outside, also it can judge whether facility for study is in moving condition by obtaining indoor picture.
For example, shooting the picture of an indoor floor at interval of certain time (such as 15s, 20s), then shooting is obtained
Picture handled, obtain the image data of picture, image data obtained according to processing, picture is compared and analyzed, when
When the floor image being continuously shot there are multiple is not identical, then prove that facility for study is in moving condition.
For S300 when the facility for study determines that itself is in moving condition, the facility for study is by current pickup mode
It is switched to orientation pickup mode；
The voice messaging that facility for study described in S400 is issued by user described in the orientation pickup type collection；
Facility for study described in S500 executes according to the voice messaging and searches topic operation accordingly.
In the present embodiment, first judge whether facility for study is in handheld state, position or further according to facility for study
The acceleration of equipment is practised to judge whether facility for study is in moving condition, confirms study whether in shifting by judging twice
The accuracy rate of judgement can be improved in dynamic state.
The 3rd embodiment provided according to the present invention, as shown in figure 3, a kind of voice searches topic method, comprising:
The wake-up signal starting voice that S100 facility for study is issued according to user searches topic function；
Facility for study described in S210 judges whether itself is in handheld state；
Specifically, judge facility for study whether be in handheld state can by facility for study be arranged gravity sensor, so
The component of acceleration for obtaining the gravity sensor acquisition on facility for study afterwards, then according to the component of acceleration on facility for study
Mutation, obtains the motion profile of facility for study, finally judges whether facility for study is in hand according to the motion profile of facility for study
Hold state.
For example, some plane of facility for study in the relative coordinate system of its own (in such as one minute) in a short time
On produce 38 degree of deflection, then facility for study knows facility for study according to the variation of the component of acceleration within the short time
38 degree have been deflected in some plane.
When knowing that facility for study deflects on some direction from stationary state according to the component of acceleration on facility for study
(deflecting in some plane) then proves that facility for study enters handheld state.
For S240 when the facility for study is in handheld state, the facility for study obtains the acceleration information of itself；
Specifically, when facility for study is in handheld state, can also by the way that acceleration transducer is arranged on facility for study,
The acceleration information of facility for study is obtained by acceleration transducer.Acceleration transducer can for 3-axis acceleration sensor or
Double-shaft acceleration sensor.
Facility for study described in S250 judges whether itself is in moving condition according to the acceleration information；
After obtaining the acceleration information of facility for study, signature analysis is carried out to acceleration information, to generate acceleration information
Feature vector.The feature vector of acceleration information can be used for the variation characteristic of the acceleration of representative learning equipment.
After getting the acceleration information of facility for study, by acceleration information input, trained moving condition is identified in advance
In model, it can judge whether facility for study is in moving condition.
The input of the training sample of moving condition identification model is the acceleration information of facility for study, is exported to accelerate degree
According to the moving condition of corresponding facility for study, moving condition identification model is trained by a large amount of training sample
Moving condition identification model after being trained.
For S300 when the facility for study determines that itself is in moving condition, the facility for study is by current pickup mode
It is switched to orientation pickup mode；
The voice messaging that facility for study described in S400 is issued by user described in the orientation pickup type collection；
Facility for study described in S500 executes according to the voice messaging and searches topic operation accordingly.
In the present embodiment, first judge whether facility for study is in handheld state, position or further according to facility for study
The acceleration of equipment is practised to judge whether facility for study is in moving condition, confirms study whether in shifting by judging twice
The accuracy rate of judgement can be improved in dynamic state.
The fourth embodiment provided according to the present invention, as shown in figure 4, a kind of voice searches topic method, comprising:
The wake-up signal starting voice that S100 facility for study is issued according to user searches topic function；
Facility for study described in S200 judges whether itself is in moving condition；
Specifically, by the way that gravity sensor is arranged in facility for study, the gravity sensor then obtained on facility for study is adopted
The component of acceleration of collection obtains the motion profile of facility for study, most then according to the mutation of the component of acceleration on facility for study
Judge whether facility for study is in handheld state according to the motion profile of facility for study afterwards.
For S300 when the facility for study determines that itself is in moving condition, the facility for study is by current pickup mode
It is switched to orientation pickup mode；
Specifically, when judging that facility for study is in moving condition according to acceleration or position etc., facility for study will be certainly
The current pickup pattern switching of body is orientation pickup mode.The voice messaging in a direction can only be acquired by orienting pickup mode, and
Ignore the voice messaging in other directions, to inhibit the noise in other directions, improves the pickup effect of facility for study, and then improve language
Sound searches the accuracy rate of topic.
The voice messaging that facility for study described in S400 is issued by user described in the orientation pickup type collection；
Facility for study described in S510 parses the voice messaging, obtains the topic information that the voice messaging includes；
Specifically, after by orienting the voice messaging that pickup type collection is issued to user, voice messaging is parsed,
The key message in voice messaging is extracted, to obtain the topic information for including in voice messaging.
Facility for study described in S520 executes according to the topic information and searches topic operation accordingly.
Specifically, after obtaining the topic information for including in voice messaging, according to topic information in the exam pool pre-established
The topic to match with topic information is searched as target topic, and gets the explanation and answer of the target topic, is then pushed away
It recommends to user.
Preferably, as shown in figure 5, facility for study described in S510 parses the voice messaging, the voice messaging packet is obtained
The topic information contained specifically includes:
The voice messaging is identified as text information by facility for study described in S511；
Specifically, after facility for study is by orientation pickup type collection to the voice messaging of user, which is known
Not Wei text information, when identifying the voice messaging, can first to the voice messaging carry out voice correction.
People when speaking, region and habit etc. due to, often will appear situations such as gulping down sound and dragging sound, gulp down sound be by
It is quickly combined into syllables in certain syllables and generates the phenomenon that swallowing, dragging sound is the phenomenon that spinning out sound when speaking.When occurring gulping down sound in voice
And when dragging sound, the discrimination of voice will affect.Therefore, the present embodiment can establish pronunciation first according to the pronunciation characteristic of different geographical
Dictionary includes gulping down sound word and gulping down the corresponding standard words of sound word and drag sound word standard words corresponding with sound word is dragged etc. in the dictionary that pronounces.
It, can be to gulping down sound and drag sound in voice messaging after collecting the voice messaging of user by establishing dictionary
Word is corrected.When user quickly combines certain syllables into syllables, it can generate and swallow phenomenon, that is, cause to lose sound, cause ability to speak not
Clearly, speech recognition is influenced.
To after gulping down sound and sound situation is dragged to be corrected in voice messaging, the voice messaging after correction is identified, is obtained
To corresponding text information.
In the present embodiment, first voice messaging is corrected, then the voice messaging after correction is identified, can be improved
The recognition accuracy of voice messaging, to improve the accuracy rate for searching topic.
Facility for study described in S512 carries out semantic parsing to the text information, obtains semantic parsing result；
Specifically, voice messaging is identified after obtaining text information, semantic parsing is carried out to text information, obtains language
Adopted parsing result.Semanteme parsing can be obtained according to semantics recognition model, and semantics recognition model can be trained by neural network model
It arrives.
Facility for study described in S513 obtains the topic information that the voice messaging includes according to the semantic parsing result.
Specifically, after the semantic parsing result for obtaining voice messaging, the keyword in semantic parsing result is first obtained, then
Keyword is edited by grammer, the topic information for including in voice messaging can be obtained.
After obtaining the topic information for including in voice messaging, by by the topic in the keyword and exam pool in topic information
It is matched, when a certain topic in exam pool and the Keywords matching in the topic information, then the entitled voice messaging pair
The target topic answered obtains the answer process of the target topic and answer and recommends user.
The 5th embodiment provided according to the present invention, as shown in fig. 6, a kind of voice searches topic system, comprising:
Topic function activation module 100 is searched, the wake-up signal starting voice for issuing according to user searches topic function；
Specifically, the starting method that voice searches topic function can be set on facility for study in advance, such as can be set as leading to
It crosses and opens a certain application program to start voice and search topic function, or start language by double-clicking a certain icon on facility for study
Sound searches topic function, or searches topic function by the special sound information that user inputs to start voice.Wherein, special sound information
It can include the related voice information of keyword " searching topic " for " searching topic ", " search topic ", " topic function is searched in unlatching " etc..
Moving condition judgment module 200, for judging whether itself is in moving condition；
Specifically, after the wake-up signal starting voice issued according to user searches topic function, first judge whether itself is in shifting
Dynamic state, judges that itself whether being in moving condition can be judged by the acceleration of facility for study or position etc..
Pickup mode switch module 300, for when the facility for study determines that itself is in moving condition, the study
Current pickup pattern switching is orientation pickup mode by equipment；
Specifically, when judging that facility for study is in moving condition according to acceleration or position etc., facility for study will be certainly
The current pickup pattern switching of body is orientation pickup mode.The voice messaging in a direction can only be acquired by orienting pickup mode, and
Ignore the voice messaging in other directions, to inhibit the noise in other directions, improves the pickup effect of facility for study, and then improve language
Sound searches the accuracy rate of topic.
Voice messaging acquisition module 400, the voice for being issued by user described in the orientation pickup type collection are believed
Breath；
Specifically, after opening orientation pickup mode, the direction that the pickup direction for orienting pickup mode can be mobile for user, i.e.,
The only voice messaging that acquisition user issues, without acquiring other sound in ambient enviroment, to inhibit the noise in other directions,
Improve the pickup effect of facility for study.
Topic execution module 500 is searched, searches topic operation accordingly for executing according to the voice messaging.
Facility for study described in S500 executes according to the voice messaging and searches topic operation accordingly.
Specifically, after by orientation pickup type collection to the voice messaging of user, according to the pass for including in voice messaging
Topic operation is searched in the execution such as keyword accordingly, to obtain the topic to match with voice messaging and answer explanation etc..
It is fixed by current pickup pattern switching when determining that facility for study itself is in moving condition in the present embodiment
To pickup mode, the voice messaging issued by orientation pickup type collection user can be reduced when user is moved to noisy environment
The low problem of speech recognition accuracy caused by and improves voice and searches topic efficiency, improves the usage experience of user.
The sixth embodiment provided according to the present invention, as shown in fig. 7, a kind of voice searches topic system, comprising:
Topic function activation module 100 is searched, the wake-up signal starting voice for issuing according to user searches topic function；
Specifically, the starting method that voice searches topic function can be set on facility for study in advance, such as can be set as leading to
It crosses and opens a certain application program to start voice and search topic function, or start language by double-clicking a certain icon on facility for study
Sound searches topic function, or searches topic function by the special sound information that user inputs to start voice.Wherein, special sound information
It can include the related voice information of keyword " searching topic " for " searching topic ", " search topic ", " topic function is searched in unlatching " etc..
It searches topic function activation module 100 and topic function is searched by voice wake-up signal starting voice, or is a certain by clicking
Icon starting voice searches topic function.
Moving condition judgment module 200, for judging whether itself is in moving condition；
Specifically, after the wake-up signal starting voice issued according to user searches topic function, moving condition judgment module 200 is first
Judge whether itself is in moving condition, judging itself whether to be in moving condition can acceleration by facility for study or position
Etc. being judged.
Pickup mode switch module 300, for when the facility for study determines that itself is in moving condition, the study
Current pickup pattern switching is orientation pickup mode by equipment；
Specifically, when judging that facility for study is in moving condition according to acceleration or position etc., facility for study passes through
Itself current pickup pattern switching is orientation pickup mode by pickup mode switch module 300.Orientation pickup mode can be adopted only
Collect the voice messaging in a direction, and ignore the voice messaging in other directions, to inhibit the noise in other directions, to improve study
The pickup effect of equipment, and then the voice for improving facility for study searches the accuracy rate of topic.
Voice messaging acquisition module 400, the voice for being issued by user described in the orientation pickup type collection are believed
Breath；
Specifically, after opening orientation pickup mode, voice messaging acquisition module 400 passes through orientation pickup type collection user
The voice messaging of sending.The pickup direction for orienting pickup mode can be the direction of user's movement, i.e., the language that only acquisition user issues
Message breath improves the pickup of facility for study without acquiring other sound in ambient enviroment to inhibit the noise in other directions
Effect.
Topic execution module 500 is searched, searches topic operation accordingly for executing according to the voice messaging.
Specifically, after by orientation pickup type collection to the voice messaging of user, topic execution module 500 is searched according to language
Topic operation is searched in the execution such as the keyword for including in message breath accordingly, is said with the topic and answer that obtain matching with voice messaging
Solution etc..
It is fixed by current pickup pattern switching when determining that facility for study itself is in moving condition in the present embodiment
To pickup mode, the voice messaging issued by orientation pickup type collection user can be reduced when user is moved to noisy environment
The low problem of speech recognition accuracy caused by and improves voice and searches topic efficiency, improves the usage experience of user.
Preferably, the moving condition judgment module 200 includes:
Handheld state judging unit 210, for judging whether itself is in handheld state；
Specifically, judge facility for study whether be in handheld state can by facility for study be arranged gravity sensor, so
The collected component of acceleration of gravity sensor on facility for study is obtained afterwards, then according to the component of acceleration on facility for study
Mutation, obtain the motion profile of facility for study, finally judge whether facility for study is according to the motion profile of facility for study
Handheld state.
For example, some plane of facility for study in the relative coordinate system of its own (in such as one minute) in a short time
On produce 38 degree of deflection, then facility for study knows facility for study according to the variation of the component of acceleration within the short time
38 degree have been deflected in some plane.
When knowing that facility for study deflects on some direction from stationary state according to the component of acceleration on facility for study
(deflecting in some plane) then proves that facility for study enters handheld state.
Positioning unit 220 obtains position positioning for carrying out position positioning to itself when itself is in handheld state
As a result；
Moving condition judging unit 230, for judging whether itself is in moving condition according to the position positioning result.
Specifically, when determining that facility for study is in handheld state, then position positioning is carried out to facility for study, such as passes through position
It sets sensor to position facility for study, position sensor can experience the position of measured object and be converted into usable output signal
Sensor can get the location information of facility for study by the way that position sensor is arranged on facility for study, to set to study
Standby to carry out position positioning, moving condition judging unit 230 judges itself whether be in mobile according to the positioning result of facility for study
State.
For example, acquiring the location information of facility for study after determining that facility for study is in handheld state, facility for study is acquired
Location information when, can at interval of certain time acquire a facility for study location information, such as be adopted every 10s or 20s or 30s
The location information for collecting a facility for study collects facility for study after the location information of each sampling time point, adopts when each
When the location information difference at sample time point, that is, illustrate that user is in movement, can judge that facility for study is in movement at this time
State.
Or the signal strength by obtaining Wifi positions facility for study, then judges to learn according to positioning result
Whether equipment is in moving condition.
Judge whether facility for study is in moving condition except through the signal strength of position sensor or interior wifi
Outside, also it can judge whether facility for study is in moving condition by obtaining indoor picture.
For example, shooting the picture of an indoor floor at interval of certain time (such as 15s, 20s), then shooting is obtained
Picture handled, obtain the image data of picture, image data obtained according to processing, picture is compared and analyzed, when
When the floor image being continuously shot there are multiple is not identical, then prove that facility for study is in moving condition.
In the present embodiment, first judge whether facility for study is in handheld state, sentence further according to the position of facility for study
Whether disconnected facility for study is in moving condition, confirms whether study is in moving condition and judgement can be improved by judging twice
Accuracy rate.
Preferably, the topic execution module 500 of searching includes:
Topic information acquiring unit 510 obtains the topic letter that the voice messaging includes for parsing the voice messaging
Breath；
Specifically, after by orienting the voice messaging that pickup type collection is issued to user, voice messaging is parsed,
The key message in voice messaging is extracted, to obtain the topic information for including in voice messaging.
Topic execution unit 520 is searched, searches topic operation accordingly for executing according to the topic information.
Specifically, after obtaining the topic information for including in voice messaging, according to topic information in the exam pool pre-established
The topic to match with topic information is searched as target topic, and gets the explanation and answer of the target topic, is then pushed away
It recommends to user.
Preferably, the topic information acquiring unit 510 includes:
Subelement 511 is identified, for the voice messaging to be identified as text information；
Specifically, after facility for study is by orientation pickup type collection to the voice messaging of user, which is known
Not Wei text information, when identifying the voice messaging, can first to the voice messaging carry out voice correction.
People when speaking, region and habit etc. due to, often will appear situations such as gulping down sound and dragging sound, gulp down sound be by
It is quickly combined into syllables in certain syllables and generates the phenomenon that swallowing, dragging sound is the phenomenon that spinning out sound when speaking.When occurring gulping down sound in voice
And when dragging sound, the discrimination of voice will affect.Therefore, the present embodiment can establish pronunciation first according to the pronunciation characteristic of different geographical
Dictionary includes gulping down sound word and gulping down the corresponding standard words of sound word and drag sound word standard words corresponding with sound word is dragged etc. in the dictionary that pronounces.
It, can be to gulping down sound and drag sound in voice messaging after collecting the voice messaging of user by establishing dictionary
Word is corrected.When user quickly combines certain syllables into syllables, it can generate and swallow phenomenon, that is, cause to lose sound, cause ability to speak not
Clearly, speech recognition is influenced.
To after gulping down sound and sound situation is dragged to be corrected in voice messaging, the voice messaging after correction is identified, is obtained
To corresponding text information.
In the present embodiment, first voice messaging is corrected, then the voice messaging after correction is identified, can be improved
The recognition accuracy of voice messaging, to improve the accuracy rate for searching topic.
Semantic parsing subunit 512 obtains semantic parsing result for carrying out semantic parsing to the text information；
Specifically, voice messaging is identified after obtaining text information, semantic parsing is carried out to text information, obtains language
Adopted parsing result.Semanteme parsing can be obtained according to semantics recognition model, and semantics recognition model can be trained by neural network model
It arrives.
Topic information obtains subelement 513, for according to the semantic parsing result, obtaining the voice messaging to include
Topic information.
Specifically, after the semantic parsing result for obtaining voice messaging, topic information obtains subelement 513 and first obtains semanteme
Then keyword in parsing result is edited keyword by grammer, the topic letter for including in voice messaging can be obtained
Breath.
After obtaining the topic information for including in voice messaging, by by the topic in the keyword and exam pool in topic information
It is matched, when a certain topic in exam pool and the Keywords matching in the topic information, then the entitled voice messaging pair
The target topic answered obtains the answer process of the target topic and answer and recommends user.
The 7th embodiment provided according to the present invention, as shown in figure 8, a kind of voice searches topic system, the present embodiment with it is above-mentioned
The difference of sixth embodiment is:
The moving condition judgment module 200 includes:
Handheld state judging unit 210, for judging whether itself is in handheld state；
Acceleration obtainment unit 240, for obtaining the acceleration information of itself when itself is in handheld state；
Specifically, when facility for study is in handheld state, can also by the way that acceleration transducer is arranged on facility for study,
The acceleration information of facility for study is obtained by acceleration transducer.Acceleration transducer can for 3-axis acceleration sensor or
Double-shaft acceleration sensor.
Moving condition judging unit 230, for judging whether itself is in moving condition according to the acceleration information.
After obtaining the acceleration information of facility for study, signature analysis is carried out to acceleration information, to generate acceleration information
Feature vector.The feature vector of acceleration information can be used for the variation characteristic of the acceleration of representative learning equipment.
After getting the acceleration information of facility for study, by acceleration information input, trained moving condition is identified in advance
In model, it can judge whether facility for study is in moving condition.
The input of the training sample of moving condition identification model is the acceleration information of facility for study, is exported to accelerate degree
According to the moving condition of corresponding facility for study, moving condition identification model is trained by a large amount of training sample
Moving condition identification model after being trained.
In the present embodiment, first judge whether facility for study is in handheld state, come further according to the acceleration of facility for study
Judge whether facility for study is in moving condition, confirms whether study is in moving condition and judgement can be improved by judging twice
Accuracy rate.
It should be noted that above-described embodiment can be freely combined as needed.The above is only of the invention preferred
Embodiment, it is noted that for those skilled in the art, in the premise for not departing from the principle of the invention
Under, several improvements and modifications can also be made, these modifications and embellishments should also be considered as the scope of protection of the present invention.
Claims (10)
1. a kind of voice searches topic method characterized by comprising
The wake-up signal starting voice that facility for study is issued according to user searches topic function；
The facility for study judges whether itself is in moving condition；
When the facility for study determines that itself is in moving condition, current pickup pattern switching is fixed by the facility for study
To pickup mode；
The voice messaging that the facility for study is issued by user described in the orientation pickup type collection；
The facility for study executes according to the voice messaging and searches topic operation accordingly.
2. a kind of voice according to claim 1 searches topic method, which is characterized in that whether the facility for study judges itself
It is specifically included in moving condition:
The facility for study judges whether itself is in handheld state；
When the facility for study is in handheld state, the facility for study carries out position positioning to itself, obtains position positioning
As a result；
The facility for study judges whether itself is in moving condition according to the position positioning result.
3. a kind of voice according to claim 1 searches topic method, which is characterized in that whether the facility for study judges itself
It is specifically included in moving condition:
The facility for study judges whether itself is in handheld state；
When the facility for study is in handheld state, the facility for study obtains the acceleration information of itself；
The facility for study judges whether itself is in moving condition according to the acceleration information.
4. a kind of voice according to claim 1-3 searches topic method, which is characterized in that the facility for study according to
The voice messaging is executed to search to inscribe to operate accordingly and be specifically included:
The facility for study parses the voice messaging, obtains the topic information that the voice messaging includes；
The facility for study executes according to the topic information and searches topic operation accordingly.
5. a kind of voice according to claim 4 searches topic method, which is characterized in that the facility for study parses the voice
Information obtains the topic information that the voice messaging is included and specifically includes:
The voice messaging is identified as text information by the facility for study；
The facility for study carries out semantic parsing to the text information, obtains semantic parsing result；
The facility for study obtains the topic information that the voice messaging includes according to the semantic parsing result.
6. a kind of voice searches topic system characterized by comprising
Topic function activation module is searched, the wake-up signal starting voice for issuing according to user searches topic function；
Moving condition judgment module, for judging whether itself is in moving condition；
Pickup mode switch module, for when the facility for study determines that itself is in moving condition, the facility for study will
Current pickup pattern switching is orientation pickup mode；
Voice messaging acquisition module, the voice messaging for being issued by user described in the orientation pickup type collection；
Topic execution module is searched, searches topic operation accordingly for executing according to the voice messaging.
7. a kind of voice according to claim 6 searches topic system, which is characterized in that the moving condition judgment module packet
It includes:
Handheld state judging unit, for judging whether itself is in handheld state；
Positioning unit, for carrying out position positioning to itself, obtaining position positioning result when itself is in handheld state；
Moving condition judging unit, for judging whether itself is in moving condition according to the position positioning result.
8. a kind of voice according to claim 6 searches topic system, which is characterized in that the moving condition judgment module packet
It includes:
Handheld state judging unit, for judging whether itself is in handheld state；
Acceleration obtainment unit, for obtaining the acceleration information of itself when itself is in handheld state；
Moving condition judging unit, for judging whether itself is in moving condition according to the acceleration information.
9. searching topic system according to a kind of described in any item voices of claim 6-8, which is characterized in that described to search topic execution module
Include:
Topic information acquiring unit obtains the topic information that the voice messaging includes for parsing the voice messaging；
Topic execution unit is searched, searches topic operation accordingly for executing according to the topic information.
10. a kind of voice according to claim 6 searches topic system, which is characterized in that the topic information acquiring unit packet
It includes:
Subelement is identified, for the voice messaging to be identified as text information；
Semantic parsing subunit obtains semantic parsing result for carrying out semantic parsing to the text information；
Topic information obtains subelement, for obtaining the topic letter that the voice messaging includes according to the semantic parsing result
Breath.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811410443.2A CN109410983A (en) | 2018-11-23 | 2018-11-23 | Voice question searching method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811410443.2A CN109410983A (en) | 2018-11-23 | 2018-11-23 | Voice question searching method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109410983A true CN109410983A (en) | 2019-03-01 |
Family
ID=65455422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811410443.2A Pending CN109410983A (en) | 2018-11-23 | 2018-11-23 | Voice question searching method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109410983A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853650A (en) * | 2019-11-28 | 2020-02-28 | 京东方科技集团股份有限公司 | Image information acquisition method and image information acquisition device |
| CN112216283A (en) * | 2020-09-24 | 2021-01-12 | 建信金融科技有限责任公司 | Voice recognition method, device, equipment and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090164219A1 (en) * | 2007-12-19 | 2009-06-25 | Enbiomedic | Accelerometer-Based Control of Wearable Devices |
| WO2015106401A1 (en) * | 2014-01-15 | 2015-07-23 | 宇龙计算机通信科技(深圳)有限公司 | Speech processing method and speech processing apparatus |
| CN105302596A (en) * | 2015-09-29 | 2016-02-03 | 努比亚技术有限公司 | Mobile terminal application opening device and method |
| CN105379307A (en) * | 2013-06-27 | 2016-03-02 | 语音处理解决方案有限公司 | Handheld mobile recording device with microphone characteristic selection means |
| CN106095072A (en) * | 2016-05-27 | 2016-11-09 | 努比亚技术有限公司 | A kind of control method, smart machine and server |
| CN107452397A (en) * | 2017-09-13 | 2017-12-08 | 安徽硕威智能科技有限公司 | Speech sound enhancement device and method based on microphone array |
| CN107680593A (en) * | 2017-10-13 | 2018-02-09 | 歌尔股份有限公司 | The sound enhancement method and device of a kind of smart machine |
| CN107800832A (en) * | 2016-09-07 | 2018-03-13 | 中兴通讯股份有限公司 | A kind of audio-frequency processing method, device and mobile terminal |
| CN108735226A (en) * | 2018-07-09 | 2018-11-02 | 科沃斯商用机器人有限公司 | Voice acquisition method, device and equipment |
| CN108735216A (en) * | 2018-06-12 | 2018-11-02 | 广东小天才科技有限公司 | Voice question searching method based on semantic recognition and family education equipment |
| CN108831474A (en) * | 2018-05-04 | 2018-11-16 | 广东美的制冷设备有限公司 | Speech recognition apparatus and its voice signal catching method, device and storage medium |
-
2018
- 2018-11-23 CN CN201811410443.2A patent/CN109410983A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090164219A1 (en) * | 2007-12-19 | 2009-06-25 | Enbiomedic | Accelerometer-Based Control of Wearable Devices |
| CN105379307A (en) * | 2013-06-27 | 2016-03-02 | 语音处理解决方案有限公司 | Handheld mobile recording device with microphone characteristic selection means |
| WO2015106401A1 (en) * | 2014-01-15 | 2015-07-23 | 宇龙计算机通信科技(深圳)有限公司 | Speech processing method and speech processing apparatus |
| CN105302596A (en) * | 2015-09-29 | 2016-02-03 | 努比亚技术有限公司 | Mobile terminal application opening device and method |
| CN106095072A (en) * | 2016-05-27 | 2016-11-09 | 努比亚技术有限公司 | A kind of control method, smart machine and server |
| CN107800832A (en) * | 2016-09-07 | 2018-03-13 | 中兴通讯股份有限公司 | A kind of audio-frequency processing method, device and mobile terminal |
| CN107452397A (en) * | 2017-09-13 | 2017-12-08 | 安徽硕威智能科技有限公司 | Speech sound enhancement device and method based on microphone array |
| CN107680593A (en) * | 2017-10-13 | 2018-02-09 | 歌尔股份有限公司 | The sound enhancement method and device of a kind of smart machine |
| CN108831474A (en) * | 2018-05-04 | 2018-11-16 | 广东美的制冷设备有限公司 | Speech recognition apparatus and its voice signal catching method, device and storage medium |
| CN108735216A (en) * | 2018-06-12 | 2018-11-02 | 广东小天才科技有限公司 | Voice question searching method based on semantic recognition and family education equipment |
| CN108735226A (en) * | 2018-07-09 | 2018-11-02 | 科沃斯商用机器人有限公司 | Voice acquisition method, device and equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110853650A (en) * | 2019-11-28 | 2020-02-28 | 京东方科技集团股份有限公司 | Image information acquisition method and image information acquisition device |
| CN112216283A (en) * | 2020-09-24 | 2021-01-12 | 建信金融科技有限责任公司 | Voice recognition method, device, equipment and storage medium |
| CN112216283B (en) * | 2020-09-24 | 2024-02-23 | 建信金融科技有限责任公司 | Voice recognition method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12087069B2 (en) | Artificial intelligence-based action recognition method and related apparatus | |
| CN110444191B (en) | Rhythm level labeling method, model training method and device | |
| US20210127003A1 (en) | Interactive voice-control method and apparatus, device and medium | |
| CN107665708B (en) | Intelligent voice interaction method and system | |
| CN108509619B (en) | Voice interaction method and device | |
| CN108000526B (en) | Dialogue interaction method and system for intelligent robot | |
| CN107240398B (en) | Intelligent voice interaction method and device | |
| CN108182937B (en) | Keyword recognition method, device, equipment and storage medium | |
| CN106297789B (en) | Personalized interaction method and system for intelligent robot | |
| WO2020083110A1 (en) | Speech recognition and speech recognition model training method and apparatus | |
| CN110503942A (en) | A kind of voice driven animation method and device based on artificial intelligence | |
| Gao et al. | Sign language recognition based on HMM/ANN/DP | |
| WO2016150001A1 (en) | Speech recognition method, device and computer storage medium | |
| CN106406806A (en) | A control method and device for intelligent apparatuses | |
| JPWO2016136062A1 (en) | Information processing apparatus, information processing method, and program | |
| CN109933198B (en) | Semantic recognition method and device | |
| WO2021115268A1 (en) | Method and apparatus for determining running direction of metro train, and terminal and storage medium | |
| CN112017645B (en) | Voice recognition method and device | |
| CN110570840A (en) | Intelligent device awakening method and device based on artificial intelligence | |
| CN109101663A (en) | A kind of robot conversational system Internet-based | |
| CN108318042A (en) | Navigation mode-switching method, device, terminal and storage medium | |
| CN109410983A (en) | Voice question searching method and system | |
| CN106356054A (en) | Method and system for collecting information of agricultural products based on voice recognition | |
| CN110534109A (en) | Audio recognition method, device, electronic equipment and storage medium | |
| CN110808050B (en) | Speech recognition method and intelligent device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190301 |
|
| RJ01 | Rejection of invention patent application after publication |