CN106338711A - Voice directing method and system based on intelligent equipment - Google Patents
Voice directing method and system based on intelligent equipment Download PDFInfo
- Publication number
- CN106338711A CN106338711A CN201610760099.4A CN201610760099A CN106338711A CN 106338711 A CN106338711 A CN 106338711A CN 201610760099 A CN201610760099 A CN 201610760099A CN 106338711 A CN106338711 A CN 106338711A
- Authority
- CN
- China
- Prior art keywords
- smart machine
- voice signal
- image
- speech
- sounnd source
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/80—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using ultrasonic, sonic or infrasonic waves
- G01S3/8003—Diversity systems specially adapted for direction finding
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Circuit For Audible Band Transducer (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a voice directing method and system based on intelligent equipment. The method comprises that after that the intelligent equipment is started, a voice signal is obtained in real time; when the voice signal is detected by the intelligent equipment, a present foreground image of the intelligent equipment is obtained, and a candidate direction of a sound source is obtained according to the present foreground image; and according to the candidate direction of a sound source, the intelligent equipment carries out calculation via a positioning algorithm, and then determines a sound source direction of the voice signal according to a calculation result. According to the invention, the candidate direction of the voice signal obtained in advance is calculated by combining image collection, the candidate direction is checked via the algorithm, the optimal sound source direction is obtained, the computing complexity is reduced, and the positioning efficiency of the voice signal is improved.
Description
Technical field
The present invention relates to voice processing technology field, more particularly, to a kind of speech-oriented method based on smart machine and be
System.
Background technology
In smart machine, particularly on domestic intelligent equipment, interactive voice is a kind of important interactive mode.At present
The process of microphone array has had a lot of methods, such as linearly constrained minimum variance, maximum-likelihood criterion and maximum letter
Make an uproar ratio criterion, the method such as Multiple Signal Classification (music), ESPRIT Signal parameter estimation (esprit), related algorithm
Typically complex.
Existing microphone array signals processing method, is exactly typically to adopt doa method, doa method is exactly direction
Of arrival, ripple reaches method estimates it is that the arrival bearing being received using one's own side's radar and being derived from objective emission machine is estimated.doa
Method is estimated to be based primarily upon pure acoustical treatment, and one side computing is larger it is especially desirable to travel direction is searched for, and also there is Search Error
Possibility.Easily error is positioned and computation complexity is high in therefore existing voice localization method.
Therefore, prior art has yet to be improved and developed.
Content of the invention
In view of the deficiencies in the prior art, present invention aim at provide a kind of speech-oriented method based on smart machine and
System easily malfunctions and the high technical problem of computation complexity it is intended to solve positioning in existing voice localization method.
Technical scheme is as follows:
A kind of speech-oriented method based on smart machine, wherein, method includes:
After a, smart machine are opened, obtain voice signal in real time;
B, when smart machine detects voice signal, obtain smart machine current foreground image, according to current foreground image
Obtain the candidate direction of Sounnd source direction;
After c, smart machine are calculated by location algorithm according to the candidate direction of Sounnd source direction, voice letter is positioned according to result of calculation
Number Sounnd source direction.
The described speech-oriented method based on smart machine, wherein, described a specifically includes:
After a1, smart machine are opened, gather voice letter with the speech transducer of arrangement form at certain intervals by several
Number.
The described speech-oriented method based on smart machine, wherein, includes before described a:
The foreground image that s, smart machine obtain in advance without foreground object as background image and stores.
The described speech-oriented method based on smart machine, wherein, described b specifically includes:
B11, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains smart machine
Current foreground image;
B12, the background image obtaining current foreground image and prestoring are calculated, and obtain Sounnd source direction according to result of calculation
Candidate direction.
The described speech-oriented method based on smart machine, wherein, described b specifically includes:
B21, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains smart machine
Current foreground image;
B22, obtain the position of the foreground object in current foreground image by image recognition algorithm, according to the position of foreground object
Obtain the candidate direction of Sounnd source direction.
The described speech-oriented method based on smart machine, wherein, described c specifically includes:
After c1, smart machine obtain the candidate direction of Sounnd source direction, checked by direction of arrival algorithm;
C2, the Sounnd source direction according to checking computation results positioning voice signal.
A kind of speech-oriented system based on smart machine, wherein, system includes:
Voice signal acquisition module, after opening for smart machine, obtains voice signal in real time;
Candidate direction computing module, for when smart machine detects voice signal, obtaining the current foreground picture of smart machine
Picture, obtains the candidate direction of Sounnd source direction according to current foreground image;
Locating module, after being calculated by location algorithm according to the candidate direction of Sounnd source direction for smart machine, is tied according to calculating
Fruit positions the Sounnd source direction of voice signal.
The described speech-oriented system based on smart machine, wherein,
Described system also includes:
Background image acquisition module, the foreground image obtaining in advance without foreground object for intelligence as background image and stores;
Described candidate direction computing module specifically includes:
First foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;
First candidate direction computing unit, the background image for obtaining current foreground image He prestore is calculated, root
Obtain the candidate direction of Sounnd source direction according to result of calculation.
The described speech-oriented system based on smart machine, wherein,
Described voice signal acquisition module is additionally operable to after smart machine opens, by several at certain intervals and arrangement form
Speech transducer collection voice signal;
Described candidate direction computing module specifically includes:
Second foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;
Second candidate direction computing unit, for obtaining the position of the foreground object in current foreground image by image recognition algorithm
Put, the candidate direction of the position acquisition Sounnd source direction according to foreground object.
The described speech-oriented system based on smart machine, wherein, described locating module specifically includes:
Computing unit, is obtained after the candidate direction of Sounnd source direction for smart machine, is checked by direction of arrival algorithm;
Positioning unit, for positioning the Sounnd source direction of voice signal according to checking computation results.
The invention provides a kind of speech-oriented method and system based on smart machine, the present invention is with reference to image acquisition meter
Calculate the candidate direction obtaining voice signal in advance, by algorithm, candidate direction is checked, obtain optimal Sounnd source direction, subtract
Lack the complexity calculating, improve the location efficiency of voice signal.
Brief description
Fig. 1 is a kind of flow chart of the preferred embodiment of speech-oriented method based on smart machine of the present invention.
Fig. 2 a is a kind of Background of the concrete application embodiment of speech-oriented method based on smart machine of the present invention
Piece schematic diagram.
Fig. 2 b be the present invention a kind of concrete application embodiment of speech-oriented method based on smart machine current before
Scape picture schematic diagram.
Fig. 2 c is a kind of difference meter of the concrete application embodiment of speech-oriented method based on smart machine of the present invention
Picture schematic diagram after calculation.
Fig. 3 is a kind of sound source candidate of the concrete application embodiment of speech-oriented method based on smart machine of the present invention
Angular range schematic diagram.
Fig. 4 is a kind of smart machine of the concrete application embodiment of speech-oriented method based on smart machine of the present invention
Schematic diagram.
Fig. 5 is a kind of principle of work and power frame of the preferred embodiment of speech-oriented system based on smart machine of the present invention
Figure.
Specific embodiment
For making the purpose of the present invention, technical scheme and effect clearer, clear and definite, below to the present invention further specifically
Bright.It should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
The invention provides a kind of flow chart of the preferred embodiment of the speech-oriented method based on smart machine, such as Fig. 1
Shown, wherein, method includes:
After step s100, smart machine are opened, obtain voice signal in real time.
When being embodied as, smart machine includes but is not limited to intelligent television, the intelligent terminal such as Intelligent flat.As shown in figure 4,
It is a kind of embodiment of intelligent television for smart machine in the present invention.Smart machine includes sound collection array, images first-class dress
Put.Wherein voice collecting array is used for collecting voice, typically has multiple sensors (mike), for gathering multi-path voice.
Further carry out in example, step s100 specifically includes:
After step s101, smart machine are opened, gathered with the speech transducer of arrangement form at certain intervals by several
Voice signal.
When being embodied as, the voice collecting array in smart machine collects voice, typically has multiple sensor (Mikes
Wind), at certain intervals and spread pattern, can obtain that same source of students sends has different to this multiple sensor
Sound, for example, have different and postpone.Voice collecting array as shown in Figure 4 is located at the top of intelligent television, has 8 Mikes
Wind, they are in uniform linear arrangement.(in other examples, can have other arrangement modes).
Step s200, when smart machine detects voice signal, obtain smart machine current foreground image, according to work as
Front foreground image obtains the candidate direction of Sounnd source direction.
When being embodied as, when smart machine detects voice signal, with the image capture device crossing setting on smart machine
Or video capture device gathers current foreground image.And obtain foreground image every a scheduled time.Smart machine institute
The scene being located at is usually static, and the renewal speed of image does not need quickly, for example, to only need to 3 frames/second.According to current prospect
The result of image determines the candidate direction of Sounnd source direction.Foreground image be smart machine display device towards scene.By pre-
First calculate candidate direction, decrease the amount of calculation of substantial amounts of search, especially under the scene needing to calculate Sounnd source direction in real time,
The optimal direction of acquisition of more precise and high efficiency.
Image capture device or video capture device collection image information thus calculating background and prospect, as shown in figure 4,
Photographic head is located at the top center of intelligent TV set.
In further embodiment, also include before step s100:
The foreground image that step s10, smart machine obtain in advance without foreground object as background image and stores.
When being embodied as, for the user of smart machine, in image change part be typically all " people ", therefore this
It is exactly the position of the sound source of possible candidate.Therefore background image is it is considered that be the image background of nobody.With intelligent television it is
Example, shoot when not playing program can pictures in television set energising but at set intervals, for example every 10 second beats take the photograph one kind
Picture, is averaged according to each time period or illumination condition in one day;Under general household situations, except a small amount of user lives
Dynamic, it is essentially all static background, as shown in Figure 2 a.Static scene or referred to as background.
Candidate direction is calculated and can be calculated with the difference of background image or by image recognition algorithm using foreground image.
Further, when candidate direction calculates the difference that can adopt foreground image and background image, step s200 is specifically wrapped
Include:
Step s211, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains intelligence
The current foreground image of energy equipment;
Step s212, the background image obtaining current foreground image and prestoring are calculated, according to result of calculation acquisition sound
The candidate direction in source direction.
When being embodied as, there is user in the middle of scene, as shown in Figure 2 b, detect prospect using existing detection algorithm,
It is exactly emerging user, that is, possible sound source direction.Pass through background and current shooting in concrete steps s212
To foreground image carry out Difference Calculation, obtain the difference portion of two width images, according to changing obvious region in error image
(for example, difference exceedes certain threshold value and area exceedes the region of certain threshold value), you can obtain prospect as shown in Figure 2 c, front
Scape i.e. user, the candidate direction of the Sounnd source direction being also simultaneously.
Further, logical candidate direction calculated image recognition algorithm and was calculated, and step s200 also includes:
Step s221, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains intelligence
The current foreground image of energy equipment;
Step s222, obtain the position of the foreground object in current foreground image by image recognition algorithm, according to foreground object
Position acquisition Sounnd source direction candidate direction.
In another example, it is possible to use existing face recognition module carries out detecting possible sound source direction.Obtain
After obtaining the position of foreground object (namely " people "), this position refers to putting down of the scope of view-finder photographing in photographic head
The position (rectangular coordinate system, pixel represents) in face, by candidate direction computing module, is scaled the angular range in Fig. 3.
Simply, can be tried to achieve by tabling look-up, that is, realize calculating the plane of scope of view-finder every picture
The angular range of element, is stored in form.The corresponding angular range of pixel further according to the scope of each object adds up.
After step s300, smart machine are calculated by location algorithm according to the candidate direction of Sounnd source direction, tie according to calculating
Fruit positions the Sounnd source direction of voice signal.
When being embodied as, calculate the possible source direction of voice, on the basis of existing computational methods, according to candidate side
To reducing the scope in possible direction further, and obtain more accurate beam direction.Wherein beam direction is generally voice
The Sounnd source direction of signal.Specifically, after being checked using doa, you can obtain optimal Sounnd source direction, complete positioning and appoint
Business, related voice signal can export and carry out subsequent treatment, such as speech recognition.
According to Sounnd source direction, voice signal is further amplified and denoising, the voice of beamformer output molding.
In further embodiment, step s300 specifically includes:
After step s301, smart machine obtain the candidate direction of Sounnd source direction, checked by direction of arrival algorithm;
Step s302, the Sounnd source direction according to checking computation results positioning voice signal.
When being embodied as, wherein beam forming concrete grammar is the multi-path voice according to aforesaid plurality of microphone input, meter
Optimal synthetic speech signal.For example there is one signal of signal to noise ratio highest.There are much existing ripples at present
Beam forming algorithm, the such as adaptive algorithm based on direction estimation, such as linearly constrained minimum variance, maximum-likelihood criterion and
Maximum signal noise ratio principle, the method such as Multiple Signal Classification (music), ESPRIT Signal parameter estimation (esprit).These
Method generally requires and Sounnd source direction is scanned for, and according to the difference of precision, these methods are required for larger computing and measure testing
Calculate the sound source in different possible directions.
Taking by music (multiple signal classification) algorithm, candidate direction is checked as a example
It is introduced,
When in prior art, the wave beam of search maximum possible reaches direction (doa direction of arrival), it is right to need
All possible deflection scans for.Candidate angle scope 0 as shown in Figure 3, is designated as d0, and typical scope is 160 degree.Search
During rope d0, every 2 degree are once calculated, and need 81 calculating.(notice that calculating here refers to, check this angle corresponding
In music algorithm, estimation space spectrum parameter, consumes larger computing resource every time).
In the technical program, by precalculated candidate direction module, decrease the amount of calculation of substantial amounts of search, especially
It is under the scene needing in real time to calculate Sounnd source direction, the optimal direction of the acquisition of more precise and high efficiency.
It is exactly briefly several directions according to candidate direction it is only necessary to checking candidate angle scope 1 as shown in Figure 3
(being designated as d1) and candidate angle scope 2(are designated as d2).For example, about 5 degree of d1 and d2 scope, need to calculate (altogether for 3 times every direction
6 times) calculate.Can see that comparing rudimentary algorithm needs 81 calculating, decreases a lot of amounts of calculation.
In the exemplary embodiment, device can be by one or more application specific integrated circuits (asic), digital signal
Processor (dsp), digital signal processing appts (dspd), PLD (pld), field programmable gate array
(fpga), controller, microcontroller, microprocessor or other electronic components are realized, for executing said method.
In the exemplary embodiment, a kind of non-provisional computer-readable recording mediums including instruction are additionally provided, for example
Including the memorizer of instruction, above-mentioned instruction can be by the computing device of device to complete said method.For example, described non-transitory
Computer-readable recording medium can be rom, random access memory (ram), cd-rom, tape, floppy disk and optical data storage
Equipment etc..
Present invention also offers a kind of principle of work and power frame of the preferred embodiment of the speech-oriented system based on smart machine
Figure, as shown in figure 5, system includes:
Voice signal acquisition module 100, after opening for smart machine, obtains voice signal in real time;Concrete such as embodiment of the method
Shown.
Candidate direction computing module 200, for when smart machine detects voice signal, obtaining the current of smart machine
Foreground image, obtains the candidate direction of Sounnd source direction according to current foreground image;Specifically as shown in embodiment of the method.
Locating module 300, after being calculated by location algorithm according to the candidate direction of Sounnd source direction for smart machine, according to
Result of calculation positions the Sounnd source direction of voice signal;Specifically as shown in embodiment of the method.
The described speech-oriented system based on smart machine, wherein,
Described system also includes:
Background image acquisition module, the foreground image obtaining in advance without foreground object for intelligence as background image and stores;
Specifically as shown in embodiment of the method.
Described candidate direction computing module specifically includes:
First foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;Specifically as shown in embodiment of the method.
First candidate direction computing unit, carry out by the background image obtaining current foreground image He prestore based on
Calculate, obtain the candidate direction of Sounnd source direction according to result of calculation;Specifically as shown in embodiment of the method.
The described speech-oriented system based on smart machine, wherein,
Described voice signal acquisition module is additionally operable to after smart machine opens, by several at certain intervals and arrangement form
Speech transducer collection voice signal;Specifically as shown in embodiment of the method.
Described candidate direction computing module specifically includes:
Second foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;Specifically as shown in embodiment of the method.
Second candidate direction computing unit, for obtaining the foreground object in current foreground image by image recognition algorithm
Position, the candidate direction of the position acquisition Sounnd source direction according to foreground object;Specifically as shown in embodiment of the method.
The described speech-oriented system based on smart machine, wherein, described locating module specifically includes:
Computing unit, is obtained after the candidate direction of Sounnd source direction for smart machine, is checked by direction of arrival algorithm;Tool
Body is as shown in embodiment of the method.
Positioning unit, for positioning the Sounnd source direction of voice signal according to checking computation results;Specifically as shown in embodiment of the method.
In sum, the invention provides a kind of speech-oriented method and system based on smart machine, method includes: intelligence
After energy opening of device, obtain voice signal in real time;When smart machine detects voice signal, obtain smart machine current before
Scape image, obtains the candidate direction of Sounnd source direction according to current foreground image;Smart machine is according to the candidate direction of Sounnd source direction
The Sounnd source direction of voice signal after calculating, is positioned by location algorithm according to result of calculation.The present invention combines image acquisition and calculates
Obtain the candidate direction of voice signal in advance, by algorithm, candidate direction is checked, obtain optimal Sounnd source direction, reduce
The complexity calculating, improves the location efficiency of voice signal.
It should be appreciated that the application of the present invention is not limited to above-mentioned citing, for those of ordinary skills, can
To be improved according to the above description or to convert, all these modifications and variations all should belong to the guarantor of claims of the present invention
Shield scope.
Claims (10)
1. a kind of speech-oriented method based on smart machine is it is characterised in that methods described includes:
After a, smart machine are opened, obtain voice signal in real time;
B, when smart machine detects voice signal, obtain smart machine current foreground image, according to current foreground image
Obtain the candidate direction of Sounnd source direction;
After c, smart machine are calculated by location algorithm according to the candidate direction of Sounnd source direction, voice letter is positioned according to result of calculation
Number Sounnd source direction.
2. the speech-oriented method based on smart machine according to claim 1 is it is characterised in that described a specifically includes:
After a1, smart machine are opened, gather voice letter with the speech transducer of arrangement form at certain intervals by several
Number.
3. the speech-oriented method based on smart machine according to claim 1 is it is characterised in that include before described a:
The foreground image that s, smart machine obtain in advance without foreground object as background image and stores.
4. the speech-oriented method based on smart machine according to claim 3 is it is characterised in that described b specifically includes:
B11, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains smart machine
Current foreground image;
B12, the background image obtaining current foreground image and prestoring are calculated, and obtain Sounnd source direction according to result of calculation
Candidate direction.
5. the speech-oriented method based on smart machine according to claim 1 is it is characterised in that described b specifically includes:
B21, when smart machine detects voice signal, smart machine passes through image or video capture device and obtains smart machine
Current foreground image;
B22, obtain the position of the foreground object in current foreground image by image recognition algorithm, according to the position of foreground object
Obtain the candidate direction of Sounnd source direction.
6. the speech-oriented method based on smart machine according to claim 4 or 5 is it is characterised in that described c specifically wraps
Include:
After c1, smart machine obtain the candidate direction of Sounnd source direction, checked by direction of arrival algorithm;
C2, the Sounnd source direction according to checking computation results positioning voice signal.
7. a kind of speech-oriented system based on smart machine is it is characterised in that system includes:
Voice signal acquisition module, after opening for smart machine, obtains voice signal in real time;
Candidate direction computing module, for when smart machine detects voice signal, obtaining the current foreground picture of smart machine
Picture, obtains the candidate direction of Sounnd source direction according to current foreground image;
Locating module, after being calculated by location algorithm according to the candidate direction of Sounnd source direction for smart machine, is tied according to calculating
Fruit positions the Sounnd source direction of voice signal.
8. the speech-oriented system based on smart machine according to claim 7 it is characterised in that
Described system also includes:
Background image acquisition module, the foreground image obtaining in advance without foreground object for intelligence as background image and stores;
Described candidate direction computing module specifically includes:
First foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;
First candidate direction computing unit, the background image for obtaining current foreground image He prestore is calculated, root
Obtain the candidate direction of Sounnd source direction according to result of calculation.
9. the speech-oriented system based on smart machine according to claim 7 it is characterised in that
Described voice signal acquisition module is additionally operable to after smart machine opens, by several at certain intervals and arrangement form
Speech transducer collection voice signal;
Described candidate direction computing module specifically includes:
Second foreground image acquiring unit, for when smart machine detects voice signal, smart machine passes through image or regards
Frequency collecting device obtains the current foreground image of smart machine;
Second candidate direction computing unit, for obtaining the position of the foreground object in current foreground image by image recognition algorithm
Put, the candidate direction of the position acquisition Sounnd source direction according to foreground object.
10. the speech-oriented system based on smart machine according to claim 8 or claim 9 is it is characterised in that described positioning mould
Block specifically includes:
Computing unit, is obtained after the candidate direction of Sounnd source direction for smart machine, is checked by direction of arrival algorithm;
Positioning unit, for positioning the Sounnd source direction of voice signal according to checking computation results.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610760099.4A CN106338711A (en) | 2016-08-30 | 2016-08-30 | Voice directing method and system based on intelligent equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610760099.4A CN106338711A (en) | 2016-08-30 | 2016-08-30 | Voice directing method and system based on intelligent equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106338711A true CN106338711A (en) | 2017-01-18 |
Family
ID=57823259
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610760099.4A Pending CN106338711A (en) | 2016-08-30 | 2016-08-30 | Voice directing method and system based on intelligent equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106338711A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108957392A (en) * | 2018-04-16 | 2018-12-07 | 深圳市沃特沃德股份有限公司 | Sounnd source direction estimation method and device |
CN108983149A (en) * | 2018-07-30 | 2018-12-11 | 中国空气动力研究与发展中心低速空气动力研究所 | A kind of pivoting microphone sound localization method |
CN109286790A (en) * | 2018-10-17 | 2019-01-29 | 杭州兆华电子有限公司 | A kind of orientation monitoring system and its monitor method based on auditory localization |
CN109672465A (en) * | 2018-12-25 | 2019-04-23 | 华为终端有限公司 | A kind of method, equipment and system adjusting antenna gain |
CN110716180A (en) * | 2019-10-17 | 2020-01-21 | 北京华捷艾米科技有限公司 | Audio positioning method and device based on face detection |
CN111402912A (en) * | 2020-02-18 | 2020-07-10 | 云知声智能科技股份有限公司 | Voice signal noise reduction method and device |
CN111551921A (en) * | 2020-05-19 | 2020-08-18 | 北京中电慧声科技有限公司 | Sound source orientation system and method based on sound image linkage |
CN113596240A (en) * | 2021-07-27 | 2021-11-02 | Oppo广东移动通信有限公司 | Recording method, recording device, electronic equipment and computer readable medium |
Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6965541B2 (en) * | 2002-12-24 | 2005-11-15 | The Johns Hopkins University | Gun shot digital imaging system |
CN101799876A (en) * | 2010-04-20 | 2010-08-11 | 王巍 | Video/audio intelligent analysis management control system |
CN101820565A (en) * | 2009-02-27 | 2010-09-01 | 本田技研工业株式会社 | The deduction method and the device thereof of source of sound |
US7792326B2 (en) * | 2006-07-11 | 2010-09-07 | Compal Electronics, Inc. | Method of tracking vocal target |
CN202095037U (en) * | 2011-04-29 | 2011-12-28 | 中国计量学院 | Real-time three-dimensional sound source image monitoring system |
CN102945672A (en) * | 2012-09-29 | 2013-02-27 | 深圳市国华识别科技开发有限公司 | Voice control system for multimedia equipment, and voice control method |
CN103235287A (en) * | 2013-04-17 | 2013-08-07 | 华北电力大学(保定) | Sound source localization camera shooting tracking device |
CN103428406A (en) * | 2012-05-23 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for analyzing monitoring video |
CN103581603A (en) * | 2012-07-24 | 2014-02-12 | 联想(北京)有限公司 | Multimedia data transmission method and electronic equipment |
CN104202694A (en) * | 2014-07-31 | 2014-12-10 | 广东美的制冷设备有限公司 | Method and system of orientation of voice pick-up device |
CN104269172A (en) * | 2014-07-31 | 2015-01-07 | 广东美的制冷设备有限公司 | Voice control method and system based on video positioning |
CN105554443A (en) * | 2015-12-04 | 2016-05-04 | 浙江宇视科技有限公司 | Method and device for positioning abnormal sound source in video image |
CN105763787A (en) * | 2014-12-19 | 2016-07-13 | 索尼公司 | Image forming method, device and electric device |
-
2016
- 2016-08-30 CN CN201610760099.4A patent/CN106338711A/en active Pending
Patent Citations (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6965541B2 (en) * | 2002-12-24 | 2005-11-15 | The Johns Hopkins University | Gun shot digital imaging system |
US7792326B2 (en) * | 2006-07-11 | 2010-09-07 | Compal Electronics, Inc. | Method of tracking vocal target |
CN101820565A (en) * | 2009-02-27 | 2010-09-01 | 本田技研工业株式会社 | The deduction method and the device thereof of source of sound |
CN101799876A (en) * | 2010-04-20 | 2010-08-11 | 王巍 | Video/audio intelligent analysis management control system |
CN202095037U (en) * | 2011-04-29 | 2011-12-28 | 中国计量学院 | Real-time three-dimensional sound source image monitoring system |
CN103428406A (en) * | 2012-05-23 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for analyzing monitoring video |
CN103581603A (en) * | 2012-07-24 | 2014-02-12 | 联想(北京)有限公司 | Multimedia data transmission method and electronic equipment |
CN102945672A (en) * | 2012-09-29 | 2013-02-27 | 深圳市国华识别科技开发有限公司 | Voice control system for multimedia equipment, and voice control method |
CN103235287A (en) * | 2013-04-17 | 2013-08-07 | 华北电力大学(保定) | Sound source localization camera shooting tracking device |
CN104202694A (en) * | 2014-07-31 | 2014-12-10 | 广东美的制冷设备有限公司 | Method and system of orientation of voice pick-up device |
CN104269172A (en) * | 2014-07-31 | 2015-01-07 | 广东美的制冷设备有限公司 | Voice control method and system based on video positioning |
CN105763787A (en) * | 2014-12-19 | 2016-07-13 | 索尼公司 | Image forming method, device and electric device |
CN105554443A (en) * | 2015-12-04 | 2016-05-04 | 浙江宇视科技有限公司 | Method and device for positioning abnormal sound source in video image |
Non-Patent Citations (3)
Title |
---|
应怀樵 编: "《现代振动与噪声技术(第4卷)》", 31 October 2005, 航空工业出版社 * |
高冠东 编: "《监狱智能视频监控技术及应用》", 29 February 2016, 河北大学出版社 * |
黄石磊 等: "一种改进的语音质量感知评估算法", 《计算机工程》 * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108957392A (en) * | 2018-04-16 | 2018-12-07 | 深圳市沃特沃德股份有限公司 | Sounnd source direction estimation method and device |
CN108983149A (en) * | 2018-07-30 | 2018-12-11 | 中国空气动力研究与发展中心低速空气动力研究所 | A kind of pivoting microphone sound localization method |
CN109286790A (en) * | 2018-10-17 | 2019-01-29 | 杭州兆华电子有限公司 | A kind of orientation monitoring system and its monitor method based on auditory localization |
CN109286790B (en) * | 2018-10-17 | 2021-05-07 | 杭州兆华电子有限公司 | Directional monitoring system based on sound source positioning and monitoring method thereof |
CN109672465A (en) * | 2018-12-25 | 2019-04-23 | 华为终端有限公司 | A kind of method, equipment and system adjusting antenna gain |
CN109672465B (en) * | 2018-12-25 | 2022-05-31 | 华为终端有限公司 | Method, equipment and system for adjusting antenna gain |
CN110716180A (en) * | 2019-10-17 | 2020-01-21 | 北京华捷艾米科技有限公司 | Audio positioning method and device based on face detection |
CN110716180B (en) * | 2019-10-17 | 2022-03-15 | 北京华捷艾米科技有限公司 | Audio positioning method and device based on face detection |
CN111402912A (en) * | 2020-02-18 | 2020-07-10 | 云知声智能科技股份有限公司 | Voice signal noise reduction method and device |
CN111551921A (en) * | 2020-05-19 | 2020-08-18 | 北京中电慧声科技有限公司 | Sound source orientation system and method based on sound image linkage |
CN113596240A (en) * | 2021-07-27 | 2021-11-02 | Oppo广东移动通信有限公司 | Recording method, recording device, electronic equipment and computer readable medium |
CN113596240B (en) * | 2021-07-27 | 2022-08-12 | Oppo广东移动通信有限公司 | Recording method, recording device, electronic equipment and computer readable medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106338711A (en) | Voice directing method and system based on intelligent equipment | |
US11398235B2 (en) | Methods, apparatuses, systems, devices, and computer-readable storage media for processing speech signals based on horizontal and pitch angles and distance of a sound source relative to a microphone array | |
EP3703052B1 (en) | Echo cancellation method and apparatus based on time delay estimation | |
US20170193282A1 (en) | Presence Sensing | |
WO2020094091A1 (en) | Image capturing method, monitoring camera, and monitoring system | |
US20190370551A1 (en) | Object detection and tracking delay reduction in video analytics | |
CN110716180B (en) | Audio positioning method and device based on face detection | |
CN110287907B (en) | Object detection method and device | |
CN103793719A (en) | Monocular distance-measuring method and system based on human eye positioning | |
US11212613B2 (en) | Signal processing device and signal processing method | |
CN105578097A (en) | Video recording method and terminal | |
CN109492577A (en) | A kind of gesture identification method, device and electronic equipment | |
US20220270266A1 (en) | Foreground image acquisition method, foreground image acquisition apparatus, and electronic device | |
WO2021052020A1 (en) | Vibration testing system | |
CN109922311B (en) | Monitoring method, device, terminal and storage medium based on audio and video linkage | |
CN107016324B (en) | Fingerprint image processing method and fingerprint detection equipment | |
CN110505398A (en) | A kind of image processing method, device, electronic equipment and storage medium | |
Savran et al. | Energy and computation efficient audio-visual voice activity detection driven by event-cameras | |
CN112183431A (en) | Real-time pedestrian number statistical method and device, camera and server | |
CN111739554A (en) | Acoustic imaging frequency determination method, device, equipment and storage medium | |
CN108875506A (en) | Face shape point-tracking method, device and system and storage medium | |
CN108875488B (en) | Object tracking method, object tracking apparatus, and computer-readable storage medium | |
WO2022227916A1 (en) | Image processing method, image processor, electronic device, and storage medium | |
CN112015364A (en) | Method and device for adjusting pickup sensitivity | |
KR102468654B1 (en) | Method of heart rate estimation based on corrected image and apparatus thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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: 20170118 |
|
RJ01 | Rejection of invention patent application after publication |