CN109599107A - A kind of method, apparatus and computer storage medium of speech recognition - Google Patents
A kind of method, apparatus and computer storage medium of speech recognition Download PDFInfo
- Publication number
- CN109599107A CN109599107A CN201811496744.1A CN201811496744A CN109599107A CN 109599107 A CN109599107 A CN 109599107A CN 201811496744 A CN201811496744 A CN 201811496744A CN 109599107 A CN109599107 A CN 109599107A
- Authority
- CN
- China
- Prior art keywords
- noise source
- user
- voice
- image information
- sign data
- 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
- 230000033001 locomotion Effects 0.000 claims description 12
- 230000005236 sound signal Effects 0.000 claims description 7
- 230000007613 environmental effect Effects 0.000 claims description 5
- 230000005055 memory storage Effects 0.000 claims 1
- 238000004378 air conditioning Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 206010047924 Wheezing Diseases 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 230000009467 reduction Effects 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 208000037656 Respiratory Sounds Diseases 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 208000033748 Device issues Diseases 0.000 description 1
- 238000010411 cooking Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000033764 rhythmic process Effects 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
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10K—SOUND-PRODUCING DEVICES; METHODS OR DEVICES FOR PROTECTING AGAINST, OR FOR DAMPING, NOISE OR OTHER ACOUSTIC WAVES IN GENERAL; ACOUSTICS NOT OTHERWISE PROVIDED FOR
- G10K11/00—Methods or devices for transmitting, conducting or directing sound in general; Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/16—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general
- G10K11/175—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound
- G10K11/178—Methods or devices for protecting against, or for damping, noise or other acoustic waves in general using interference effects; Masking sound by electro-acoustically regenerating the original acoustic waves in anti-phase
-
- G10L21/0202—
-
- 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)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a kind of method, apparatus of speech recognition and computer storage mediums, to solve the not high technical problem of phonetic recognization rate existing in the prior art.It include: the voice for obtaining smart home device described in user's control, the current scene image information of image capture device acquisition and the sign data of user；According to the current scene image information and the sign data, effective noise source is determined；Wherein, effective noise source is that the noise source of direct interference is generated to the voice that user issues in the current scene；Effective noise source is rejected from the voice, the voice after being denoised, according to smart home device described in the voice control after the denoising.
Description
Technical field
The present invention relates to smart home fields, method, apparatus and computer storage more particularly, to a kind of speech recognition
Medium.
Background technique
In smart home field, the interactive voice man-machine interaction mode novel as one kind of smart home device is wide
General use.
For example, intelligent sound speaker, intelligent sound TV, intelligent sound air-conditioning etc., user can be by voice to it
Controlled.
However, user, when carrying out voice control to smart home device, user may also do other things simultaneously, such as
Running, cooking, blow head hair etc., will necessarily be mixed into other more make an uproar in the collected user speech of smart home device at this time
Sound, this will reduce smart home device to the discrimination of user speech.
In consideration of it, how effectively to improve the phonetic recognization rate of smart home device, become a technology urgently to be resolved
Problem.
Summary of the invention
The present invention provides the method, apparatus and computer storage medium of a kind of speech recognition, to solve in the prior art
The not high technical problem of existing phonetic recognization rate.
In a first aspect, in order to solve the above technical problems, a kind of method of speech recognition provided in an embodiment of the present invention, application
It is as follows in the technical solution of smart home device, this method:
Obtain the voice of smart home device described in user's control, the current scene image information of image capture device acquisition
And the sign data of user；
According to the current scene image information and the sign data, effective noise source is determined；Wherein, described effectively to make an uproar
Source of sound is that the noise source of direct interference is generated to the voice that user issues in the current scene；
Effective noise source is rejected from the voice, the voice after being denoised, according to the language after the denoising
Smart home device described in sound control system.
When obtaining the voice of user's control smart home device, current scene image is also acquired by image capture device
The sign data of information and user；And according to current scene image information and sign data, by current scene to user
The noise source that the voice of sending generates direct interference is determined as effective noise source；And reject effective noise source from voice, it obtains
Voice after must denoising, further according to the voice control smart home device after denoising.To enable smart home device according to really
Effective noise source in the current scene made carries out specific aim noise reduction to user speech, to improve phonetic recognization rate.
Optionally, according to the current scene image information and the sign data, effective noise source is determined, comprising:
According to the current scene image information, the corresponding user of the current scene image information is obtained from database
Noise source set present in scene；
According to the corresponding sign data range of every kind of User Status in the database, judge that the sign data is corresponding
Current user state；
According to the current user state from the noise source set, effective noise source is obtained.
Optionally, according to the current user state from the noise source set, effective noise source, packet are obtained
It includes:
According to the current user state, judge whether that the corresponding noise source of user's noise for generating user is made an uproar from described
It is removed in source of sound set, obtains judging result；Wherein, the sound issued when user's noise is user movement；
If the judging result is to need to remove the corresponding noise source of user's noise, from the noise source set
The corresponding noise source of user's noise is removed, new noise source set is obtained；
The noise source that interference value is less than preset threshold is picked out from the new noise source set, acquisition is described effectively to make an uproar
Source of sound.
Optionally, effective noise source is rejected from the voice, the voice after being denoised, comprising:
After the corresponding audio signal of the effective noise source is done reverse phase processing, the corresponding audio letter of the voice that is added to
Voice in number, after obtaining the denoising.
Optionally, the sign data is acquired particular by wearable device；The wearable device includes movement
The group of any one or more of sensor, biosensor, environmental sensor, skin electric transducer, heart rate sensor, barometer
It closes.
Second aspect, the embodiment of the invention provides a kind of devices for speech recognition, comprising:
Acquiring unit, for obtaining the voice of smart home device described in user's control, image capture device acquisition is worked as
The sign data of preceding scene image information and user；
Determination unit, for determining effective noise source according to the current scene image information and the sign data；Its
In, effective noise source is that the noise source of direct interference is generated to the voice that user issues in the current scene；
Culling unit, for effective noise source to be rejected from the voice, the voice after being denoised, according to institute
Smart home device described in voice control after stating denoising.
Optionally, the determination unit is specifically used for:
According to the current scene image information, the corresponding user of the current scene image information is obtained from database
Noise source set present in scene；
According to the corresponding sign data range of every kind of User Status in the database, judge that the sign data is corresponding
Current user state；
According to the current user state from the noise source set, effective noise source is obtained.
Optionally, the determination unit is also used to:
According to the current user state, judge whether that the corresponding noise source of user's noise for generating user is made an uproar from described
It is removed in source of sound set, obtains judging result；Wherein, the sound issued when user's noise is user movement；
If the judging result is to need to remove the corresponding noise source of user's noise, from the noise source set
The corresponding noise source of user's noise is removed, new noise source set is obtained；
The noise source that interference value is less than preset threshold is picked out from the new noise source set, acquisition is described effectively to make an uproar
Source of sound.
Optionally, the culling unit is specifically used for:
After the corresponding audio signal of the effective noise source is done reverse phase processing, the corresponding audio letter of the voice that is added to
Voice in number, after obtaining the denoising.
Optionally, the sign data includes that movement passes particular by the wearable device that wearable device acquires
The group of any one or more of sensor, biosensor, environmental sensor, skin electric transducer, heart rate sensor, barometer
It closes.
The third aspect, the embodiment of the present invention also provide a kind of device for speech recognition, comprising:
At least one processor, and
The memory being connect at least one described processor；
Wherein, the memory is stored with the instruction that can be executed by least one described processor, described at least one
The instruction that device is stored by executing the memory is managed, the method as described in above-mentioned first aspect is executed.
Fourth aspect, the embodiment of the present invention also provide a kind of computer readable storage medium, comprising:
The computer-readable recording medium storage has computer instruction, when the computer instruction is run on computers
When, so that computer executes the method as described in above-mentioned first aspect.
The technical solution in said one or multiple embodiments through the embodiment of the present invention, the embodiment of the present invention at least have
There is following technical effect:
In embodiment provided by the invention, when obtaining the voice of user's control smart home device, also pass through image
Acquire equipment acquisition current scene image information and the sign data of user；And according to current scene image information and sign
The noise source for generating direct interference to the voice that user issues in current scene is determined as effective noise source by data；And will have
Effect noise source is rejected from voice, the voice after being denoised, further according to the voice control smart home device after denoising.To
Enable smart home device according to effective noise source in the current scene determined, specific aim noise reduction carried out to user speech,
To improve phonetic recognization rate.
Detailed description of the invention
Fig. 1 is a kind of flow chart of audio recognition method provided in an embodiment of the present invention；
Fig. 2 is the schematic diagram that smart home device carries out speech recognition in running scene provided in an embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of speech recognition equipment provided in an embodiment of the present invention.
Specific embodiment
Implementation column of the present invention provides the method, apparatus and computer storage medium of a kind of speech recognition, to solve existing skill
The not high technical problem of phonetic recognization rate present in art.
In order to solve the above technical problems, general thought is as follows for technical solution in the embodiment of the present application:
There is provided a kind of method of speech recognition, comprising: obtain the voice of user's control smart home device, Image Acquisition is set
The current scene image information of standby acquisition and the sign data of user；According to current scene image information and sign data,
Determine effective noise source；Wherein, effective noise source is to generate making an uproar for direct interference to the voice that user issues in current scene
Source of sound, effective noise source is rejected from voice, the voice after being denoised, and is set according to the voice control smart home after denoising
It is standby.
Due in the above scheme, when obtaining the voice of user's control smart home device, also being set by Image Acquisition
The sign data of standby acquisition current scene image information and user；And according to current scene image information and sign data,
The noise source for generating direct interference to the voice that user issues in current scene is determined as effective noise source；And by effective noise
Source is rejected from voice, the voice after being denoised, further according to the voice control smart home device after denoising.To make intelligence
Home equipment can carry out specific aim noise reduction to user speech, to mention according to effective noise source in the current scene determined
High phonetic recognization rate.
In order to better understand the above technical scheme, below by attached drawing and specific embodiment to technical solution of the present invention
It is described in detail, it should be understood that the specific features in the embodiment of the present invention and embodiment are to the detailed of technical solution of the present invention
Thin explanation, rather than the restriction to technical solution of the present invention, in the absence of conflict, the embodiment of the present invention and embodiment
In technical characteristic can be combined with each other.
Referring to FIG. 1, the embodiment of the present invention provides a kind of method of speech recognition, it is applied in smart home device, it should
The treatment process of method is as follows.
Step 101: obtaining the voice of control smart home device, the current scene image letter of image capture device acquisition
Breath and the sign data of user.
With extensive use of the speech recognition technology in smart home device, user is not only allowed to eliminate in intelligence to be controlled
Controller can be needed to find when home equipment, or to bothering for control button is operated on smart home device, allow user at home
When can control smart home device at any time.For example, remote controler can be used in control air-conditioning in the past, it now is possible to pass through voice control
Air-conditioning processed.
In embodiment provided by the invention, image capture device can be camera, ccd sensor, camera etc., image
Acquisition equipment can be the component part of smart home device, be also possible to external image capture device, can also be intelligence
Camera on mobile phone, external image capture device can be communicated by wired mode with smart home device,
It can wirelessly be communicated with smart home device, specifically without limitation.
Optionally, the sign data of user is acquired particular by wearable device.
For example, being sent out when obtaining the voice of user's control smart home device from intelligentized Furniture equipment to wearable device
The sign data of user is acquired out；Or image capture device acquire current scene image information when, from image capture device to
Wearable device issues the sign data of acquisition user；Or wearable device is periodically to smart home device or image capture device
The sign data of report of user.
Specifically, wearable device can be Intelligent bracelet, motion sensor, biosensor, environmental sensor, skin electricity
The combination of any one or more of sensor, heart rate sensor, barometer, wearable device can be by wireless networks, such as
WLAN, bluetooth, Zigbee etc. are communicated with smart home device, and the sign data measured is transferred to smart home
Equipment or image capture device acquisition, are communicated in which way between specific wearable device and smart home device,
This is without limitation.
For example, refer to Fig. 2, when user runs on a treadmill, with the increase of running time, make user produce compared with
More heats, user wants to open air-conditioning, but is unwilling to stop and be controlled with remote controler, at this point, user can run
Phonetic control command is issued to air-conditioning while step, such as says " booting is adjusted to 26 DEG C " to air-conditioning.
At this point, smart home device air-conditioning gets the voice of user's control smart home device air-conditioning, and pass through image
Equipment acquisition current scene image information (running scene) is acquired, while also being used by wearable device (such as Intelligent bracelet) acquisition
The sign data (such as heart rate, movement velocity) at family.
Obtain control smart home device voice, the current scene image information of image capture device acquisition and
After the sign data of user, step 102 can be executed.
Step 102: effective noise source is determined according to current scene image information and sign data；Wherein, effective noise source
To generate the noise source of direct interference to the voice that user issues in the current scene.
Specifically, can be obtained and current scene image information pair from database first according to current scene image information
Noise source set present in the user's scene answered；Its then, further according to the corresponding sign number of every kind of User Status in database
According to range, the corresponding current user state of sign data is judged；Finally, being obtained from noise source set according to current user state
Take effective noise source.
For example, smart home device can be according to the scene image information in database still by taking the example in Fig. 2 as an example
In feature, be compared with the feature in collected current scene image information, as there is the body of user in scene of running
Certain gradient, arm alternately leave ground in swing, foot, and rhythm is identical as the amplitude of arms swing, has race
Step machine etc. is set when the characteristic similarity in the running scene image information in the feature and database in scene image information reaches
When determining threshold value (such as 90%), can determine current scene image information it is corresponding be running scene image information.And then it can be from
It is determined in database, corresponding noise source has the set noise source generated when treadmill work, uses in scene image information of running
The noise source of wheezing etc. in the friction noise source, user that are generated when family moves on a treadmill, and then determine to work as from database
Noise source set present in the corresponding user's scene of preceding scene image information are as follows: set noise source, wheezes and makes an uproar at friction noise source
Source of sound.
Later, according to the corresponding sign data range of every kind of User Status in database (such as heart rate data range), judgement
The corresponding current user state of sign data (i.e. heart rate data) that air-conditioning is obtained by Intelligent bracelet, such as user in database
Heart rate range is that 80-100 corresponds to static or state of being careful when for static state, and user is that heart rate range when jogging is 101-120 pairs
Should jog state, and heart rate range of the user when hurrying up is greater than 121 corresponding states of hurrying up.If the body obtained by Intelligent bracelet
Sign data, that is, heart rate is 90 and illustrates that user's current state is static or state of being careful；If the heart rate measured is 130, illustrate to use
Family current state is state of hurrying up.
After determining the current state of user, according to current user state from noise source set, effective noise is obtained
Source can be accomplished by the following way:
Firstly, judge whether that the user's noise for generating user is removed from noise source set according to current user state,
Obtain judging result；Wherein, the sound issued when user's noise is user movement.
Secondly, removing user if judging result is the corresponding noise source of removal user's noise from noise source set and making an uproar
The corresponding noise source of sound obtains new noise source set.
Finally, picking out the noise source that interference value is less than preset threshold from new noise source set, effective noise is obtained
Source.
For example, the voice influence very little that whoop issues user at this time can if current state is static or state of being careful
To ignore, and then whoop can be removed from noise source set, obtain new noise source collection and be combined into set noise source and rub
The noise source (for example friction noise source) that interference value is less than given threshold is later removed from new noise source set in rubbing noise source
It removes, obtaining effective noise source is set noise source.If current state is state of hurrying up, wheeze what noise source issued user at this time
Voice is affected, and cannot ignore, and obtains new noise source collection and is combined into set noise source and friction noise source, noise source of wheezing,
Therefrom the noise source (noise source of such as wheezing) that interference value is less than given threshold is removed later, effective noise source is obtained and makes an uproar for machine
Source of sound, friction noise source.
After obtaining effective noise source, step 103 can be executed.
Step 103: effective noise source being rejected from voice, the voice after being denoised, according to the voice control after denoising
Smart home device processed.
Specifically, after the corresponding audio signal of effective noise source is done reverse phase processing, the corresponding audio letter of the voice that is added to
Voice in number, after being denoised.
For example, effective noise source is set noise source, friction noise source still by taking the example in Fig. 2 as an example, machine is made an uproar
After the corresponding audio signal of source of sound and friction noise source does reverse phase processing, the language of the user's control that is added to smart home device
In sound, it is cancelled set noise source and friction noise source in the voice of user's control smart home device, to be gone
Except the voice (voice after denoising) after set noise source and friction noise source.Then according to the voice control intelligence after denoising
It can home equipment execution booting and temperature adjusting (adjusting temperature to 26 DEG C).
Based on the same inventive concept, a kind of device for speech recognition is provided in one embodiment of the invention, the device
The specific embodiment of audio recognition method can be found in the description of embodiment of the method part, and overlaps will not be repeated, refer to
Fig. 3, the device include:
Acquiring unit 301, for obtaining the voice of smart home device described in user's control, image capture device acquisition
The sign data of current scene image information and user；
Determination unit 302, for determining effective noise according to the current scene image information and the sign data
Source；Wherein, effective noise source is that the noise of direct interference is generated to the voice that user issues in the current scene
Source；
Culling unit 303, for effective noise source to be rejected from the voice, the voice after being denoised, root
According to smart home device described in the voice control after the denoising.
Optionally, the determination unit 302 is specifically used for:
According to the current scene image information, the corresponding user of the current scene image information is obtained from database
Noise source set present in scene；
According to the corresponding sign data range of every kind of User Status in the database, judge that the sign data is corresponding
Current user state；
According to the current user state from the noise source set, effective noise source is obtained.
Optionally, the determination unit 302 is also used to:
According to the current user state, judge whether that the corresponding noise source of user's noise for generating user is made an uproar from described
It is removed in source of sound set, obtains judging result；Wherein, the sound issued when user's noise is user movement；
If the judging result is to need to remove the corresponding noise source of user's noise, from the noise source set
The corresponding noise source of user's noise is removed, new noise source set is obtained；
The noise source that interference value is less than preset threshold is picked out from the new noise source set, acquisition is described effectively to make an uproar
Source of sound.
Optionally, the culling unit 303 is specifically used for:
After the corresponding audio signal of the effective noise source is done reverse phase processing, the corresponding audio letter of the voice that is added to
Voice in number, after obtaining the denoising.
Optionally, the sign data is acquired particular by wearable device；
The wearable device includes motion sensor, biosensor, environmental sensor, skin electric transducer, heart rate biography
The combination of any one or more of sensor, barometer.
Based on the same inventive concept, a kind of device for speech recognition is provided in the embodiment of the present invention, comprising: at least
One processor, and
The memory being connect at least one described processor；
Wherein, the memory is stored with the instruction that can be executed by least one described processor, described at least one
The instruction that device is stored by executing the memory is managed, audio recognition method as described above is executed.
Based on the same inventive concept, the embodiment of the present invention also mentions a kind of computer readable storage medium, comprising:
The computer-readable recording medium storage has computer instruction, when the computer instruction is run on computers
When, so that computer executes audio recognition method as described above.
In embodiment provided by the invention, when obtaining the voice of user's control smart home device, also pass through image
Acquire equipment acquisition current scene image information and the sign data of user；And according to current scene image information and sign
The noise source for generating direct interference to the voice that user issues in current scene is determined as effective noise source by data；And will have
Effect noise source is rejected from voice, the voice after being denoised, further according to the voice control smart home device after denoising.To
Enable smart home device according to effective noise source in the current scene determined, specific aim noise reduction carried out to user speech,
To improve phonetic recognization rate.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as the production of method, system or computer program
Product.Therefore, in terms of the embodiment of the present invention can be used complete hardware embodiment, complete software embodiment or combine software and hardware
Embodiment form.Moreover, it wherein includes computer available programs generation that the embodiment of the present invention, which can be used in one or more,
The meter implemented in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of code
The form of calculation machine program product.
The embodiment of the present invention be referring to according to the method for the embodiment of the present invention, equipment (system) and computer program product
Flowchart and/or the block diagram describe.It should be understood that can be realized by computer program instructions in flowchart and/or the block diagram
The combination of process and/or box in each flow and/or block and flowchart and/or the block diagram.It can provide these calculating
Processing of the machine program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
Device is to generate a machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute
For realizing the function of being specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
Device.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of method of speech recognition is applied to smart home device characterized by comprising
Obtain user's control described in smart home device voice, image capture device acquisition current scene image information and
The sign data of user；
According to the current scene image information and the sign data, effective noise source is determined；Wherein, effective noise
Source is that the noise source of direct interference is generated to the voice that user issues in the current scene；By effective noise source from
It is rejected in the voice, the voice after being denoised, according to smart home device described in the voice control after the denoising.
2. the method as described in claim 1, which is characterized in that according to the current scene image information and the sign number
According to determining effective noise source, comprising:
According to the current scene image information, user corresponding with the current scene image information is obtained from database
Noise source set present in scape；
According to the corresponding sign data range of every kind of User Status in the database, judge that the sign data is corresponding current
User Status；
According to the current user state from the noise source set, effective noise source is obtained.
3. method according to claim 2, which is characterized in that according to the current user state from the noise source set
In, obtain effective noise source, comprising:
According to the current user state, judge whether the corresponding noise source of user's noise for generating user from the noise source
It is removed in set, obtains judging result；Wherein, the sound issued when user's noise is user movement；
If the judging result is to remove the corresponding noise source of user's noise, from the noise source set described in removal
The corresponding noise source of user's noise obtains new noise source set；
The noise source that interference value is less than preset threshold is picked out from the new noise source set, obtains effective noise
Source.
4. the method as described in any claim of claim 1-3, which is characterized in that by effective noise source from the voice
It rejects, the voice after being denoised, comprising:
After the corresponding audio signal of the effective noise source is done reverse phase processing, be added to the corresponding audio signal of the voice
In, the voice after obtaining the denoising.
5. method as claimed in claim 4, which is characterized in that the sign data is acquired particular by wearable device
's；The wearable device include motion sensor, biosensor, environmental sensor, skin electric transducer, heart rate sensor,
The combination of any one or more of barometer.
6. a kind of device of speech recognition is applied to smart home device characterized by comprising
Acquiring unit, for obtaining the voice of smart home device described in user's control, front court is worked as in image capture device acquisition
The sign data of scape image information and user；
Determination unit, for determining effective noise source according to the scene image information and the sign data；Wherein, described
Effective noise source is that the noise source of direct interference is generated to the voice that user issues in the current scene；
Culling unit, for rejecting effective noise source from the voice, the voice after being denoised is gone according to described
Smart home device described in voice control after making an uproar.
7. device as claimed in claim 6, which is characterized in that the determination unit is specifically used for:
According to the current scene image information, the corresponding user's scene of the current scene image information is obtained from database
Present in noise source set；
According to the corresponding sign data range of every kind of User Status in the database, judge that the sign data is corresponding current
User Status；
According to the current user state from the noise source set, effective noise source is obtained.
8. device as claimed in claim 7, which is characterized in that the determination unit is also used to:
According to the current user state, judge whether the corresponding noise source of user's noise for generating user from the noise source
It is removed in set, obtains judging result；Wherein, the sound issued when user's noise is user movement；
If the judging result is to need to remove the corresponding noise source of user's noise, removed from the noise source set
The corresponding noise source of user's noise, obtains new noise source set；
The noise source that interference value is less than preset threshold is picked out from the new noise source set, obtains effective noise
Source.
9. a kind of device of speech recognition characterized by comprising
At least one processor, and
The memory being connect at least one described processor；
Wherein, the memory is stored with the instruction that can be executed by least one described processor, at least one described processor
By executing the instruction of the memory storage, the method according to claim 1 to 5 is executed.
10. a kind of computer readable storage medium, it is characterised in that:
The computer-readable recording medium storage has computer instruction, when the computer instruction is run on computers,
So that computer executes method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811496744.1A CN109599107A (en) | 2018-12-07 | 2018-12-07 | A kind of method, apparatus and computer storage medium of speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811496744.1A CN109599107A (en) | 2018-12-07 | 2018-12-07 | A kind of method, apparatus and computer storage medium of speech recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109599107A true CN109599107A (en) | 2019-04-09 |
Family
ID=65961547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811496744.1A Pending CN109599107A (en) | 2018-12-07 | 2018-12-07 | A kind of method, apparatus and computer storage medium of speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109599107A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110164425A (en) * | 2019-05-29 | 2019-08-23 | 北京声智科技有限公司 | A kind of noise-reduction method, device and the equipment that can realize noise reduction |
| CN110288989A (en) * | 2019-06-03 | 2019-09-27 | 安徽兴博远实信息科技有限公司 | Voice interactive method and system |
| CN110594970A (en) * | 2019-08-23 | 2019-12-20 | 宁波奥克斯电气股份有限公司 | System and method for controlling air conditioner based on voice of child watch |
| WO2021143599A1 (en) * | 2020-01-15 | 2021-07-22 | 荣耀终端有限公司 | Scene recognition-based speech processing method and apparatus, medium and system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146314B2 (en) * | 2001-12-20 | 2006-12-05 | Renesas Technology Corporation | Dynamic adjustment of noise separation in data handling, particularly voice activation |
| CN102427418A (en) * | 2011-12-09 | 2012-04-25 | 福州海景科技开发有限公司 | Intelligent household system based on speech recognition |
| CN102694891A (en) * | 2011-03-21 | 2012-09-26 | 鸿富锦精密工业（深圳）有限公司 | System and method for removing conversation noises |
| CN106128451A (en) * | 2016-07-01 | 2016-11-16 | 北京地平线机器人技术研发有限公司 | Method for voice recognition and device |
| CN107016996A (en) * | 2017-06-06 | 2017-08-04 | 广东小天才科技有限公司 | A kind of processing method and processing device of voice data |
| CN107257996A (en) * | 2015-03-26 | 2017-10-17 | 英特尔公司 | The method and system of environment sensitive automatic speech recognition |
| CN108320746A (en) * | 2018-02-09 | 2018-07-24 | 杭州智仁建筑工程有限公司 | A kind of intelligent domestic system |
| CN108710485A (en) * | 2018-04-19 | 2018-10-26 | 珠海格力电器股份有限公司 | A kind of information output method, terminal device and readable storage medium storing program for executing |
-
2018
- 2018-12-07 CN CN201811496744.1A patent/CN109599107A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146314B2 (en) * | 2001-12-20 | 2006-12-05 | Renesas Technology Corporation | Dynamic adjustment of noise separation in data handling, particularly voice activation |
| CN102694891A (en) * | 2011-03-21 | 2012-09-26 | 鸿富锦精密工业（深圳）有限公司 | System and method for removing conversation noises |
| CN102427418A (en) * | 2011-12-09 | 2012-04-25 | 福州海景科技开发有限公司 | Intelligent household system based on speech recognition |
| CN107257996A (en) * | 2015-03-26 | 2017-10-17 | 英特尔公司 | The method and system of environment sensitive automatic speech recognition |
| CN106128451A (en) * | 2016-07-01 | 2016-11-16 | 北京地平线机器人技术研发有限公司 | Method for voice recognition and device |
| CN107016996A (en) * | 2017-06-06 | 2017-08-04 | 广东小天才科技有限公司 | A kind of processing method and processing device of voice data |
| CN108320746A (en) * | 2018-02-09 | 2018-07-24 | 杭州智仁建筑工程有限公司 | A kind of intelligent domestic system |
| CN108710485A (en) * | 2018-04-19 | 2018-10-26 | 珠海格力电器股份有限公司 | A kind of information output method, terminal device and readable storage medium storing program for executing |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110164425A (en) * | 2019-05-29 | 2019-08-23 | 北京声智科技有限公司 | A kind of noise-reduction method, device and the equipment that can realize noise reduction |
| CN110288989A (en) * | 2019-06-03 | 2019-09-27 | 安徽兴博远实信息科技有限公司 | Voice interactive method and system |
| CN110594970A (en) * | 2019-08-23 | 2019-12-20 | 宁波奥克斯电气股份有限公司 | System and method for controlling air conditioner based on voice of child watch |
| WO2021143599A1 (en) * | 2020-01-15 | 2021-07-22 | 荣耀终端有限公司 | Scene recognition-based speech processing method and apparatus, medium and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109599107A (en) | A kind of method, apparatus and computer storage medium of speech recognition | |
| CN105902257B (en) | Sleep state analysis method and device, intelligent wearable device | |
| CN106774818B (en) | Gesture recognition method, gesture recognition device and wearable device | |
| CN105868827B (en) | A kind of multi-modal exchange method of intelligent robot and intelligent robot | |
| US20170143226A1 (en) | Action Recognition Method and Device Based on Surface Electromyography Signal | |
| JP2017537370A (en) | Natural motion-based control via wearable and mobile devices | |
| CN110268470A (en) | The modification of audio frequency apparatus filter | |
| CN104731307B (en) | A kind of body-sensing action identification method and human-computer interaction device | |
| CN103353935A (en) | 3D dynamic gesture identification method for intelligent home system | |
| CN110308795B (en) | Dynamic gesture recognition method and system | |
| CN110459214A (en) | Voice interactive method and device | |
| US10901507B2 (en) | Bioelectricity-based control method and apparatus, and bioelectricity-based controller | |
| CN111107278B (en) | Image processing method and device, electronic equipment and readable storage medium | |
| CN105701196A (en) | Intelligent robot oriented audio processing method and intelligent robot | |
| CN104317392B (en) | A kind of information control method and electronic equipment | |
| KR20200094732A (en) | Method and system for classifying time series data | |
| CN108108015A (en) | A kind of action gesture recognition methods based on mobile phone gyroscope and dynamic time warping | |
| CN109840939A (en) | Three-dimensional rebuilding method, device, electronic equipment and storage medium | |
| CN103472915A (en) | Reading control method and reading control device on basis of pupil tracking and display equipment | |
| CN110443113A (en) | A kind of virtual reality Writing method, system and storage medium | |
| Elgendi et al. | Real-time speed detection of hand gesture using, Kinect | |
| Song et al. | Facelistener: Recognizing human facial expressions via acoustic sensing on commodity headphones | |
| Jindal et al. | A comparative analysis of established techniques and their applications in the field of gesture detection | |
| CN209606794U (en) | A kind of wearable device, sound-box device and intelligent home control system | |
| CN110459206A (en) | A kind of speech recognition system and method based on track planning of dual robots identification |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190409 |