CN108831502A - Caution sound identified off-line method and device - Google Patents
Caution sound identified off-line method and device Download PDFInfo
- Publication number
- CN108831502A CN108831502A CN201810422085.0A CN201810422085A CN108831502A CN 108831502 A CN108831502 A CN 108831502A CN 201810422085 A CN201810422085 A CN 201810422085A CN 108831502 A CN108831502 A CN 108831502A
- Authority
- CN
- China
- Prior art keywords
- sound
- caution sound
- caution
- vibration
- matched
- 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
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
- G10L21/16—Transforming into a non-visible representation
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/06—Transformation of speech into a non-audible representation, e.g. speech visualisation or speech processing for tactile aids
- G10L2021/065—Aids for the handicapped in understanding
Abstract
Caution sound identified off-line method and device proposed by the present invention, wherein method include：Obtain the digital signal of current sound；The digital signal is matched with the caution sound sound bank prestored, obtains matched caution sound；According to the caution sound sound bank matched vibration type list in advance, vibration type corresponding with the caution sound is called；Corresponding first vibration is generated according to the vibration type.The present invention provides a kind of caution sound identified off-line method and device suitable for person hard of hearing.
Description
Technical field
The present invention relates to technical field of voice recognition, a kind of caution sound identified off-line method and dress are especially related to
It sets.
Background technique
For the people that hearing has obstacle, the case where relying primarily on vision in trip come around observing, but it is simple
By being limited in scope for visual observation, especially behind and the observation the case where left and right sides be very limited.Have when in hearing
When there is vehicle close to equal unsafe conditions in behind the people of the obstacle or left and right sides, often due to can't hear abnormal caution sound from
And bring biggish security risk.Existing speech recognition technology comparative maturity, although in artificial intelligence equipment, turn over
Translate the extensive use of the fields such as machine, but above-mentioned field generally requires and relies on cloud database to carry out speech recognition, power consumption and at
This is higher, and needs consumed flow.It is existing also to have through offline speech recognition apparatus, but it can only generally be identified less
Number keyword, voice, and the hearing that the sound of its output or text are not suitable for trip has the people of obstacle to use.
Therefore, the mode for how providing the caution sound identified off-line of people for having obstacle suitable for hearing a kind of, which becomes, urgently to be solved
Certainly the problem of.
Summary of the invention
The main object of the present invention be provide a kind of people for having obstacle suitable for hearing caution sound identified off-line method and
Device.
The present invention proposes caution sound identified off-line method, including：
Obtain the digital signal of current sound；
The digital signal is matched with the caution sound sound bank prestored, obtains matched caution sound；
According to the caution sound sound bank matched vibration type list in advance, vibration corresponding with the caution sound is called
Type；
Corresponding first vibration is generated according to the vibration type.
Further, described to match the digital signal with the caution sound sound bank prestored, obtain matched police
Before the step of showing sound, including：
Whether the decibel value for analyzing the digital signal is more than preset decibel value；
If it is not, then matching the digital signal with the caution sound sound bank prestored, matched caution sound is obtained, if
It is then to generate corresponding second vibration.
Further, described the step of generating the corresponding second vibration, including：
Determine digital signal decibel value range locating in matched vibration mapping table in advance, the preparatory matching
Vibration mapping table include different decibel value ranges and vibration type corresponding relationship；
Corresponding vibration type is called in vibration mapping table according to locating decibel value range；
Corresponding second vibration is generated according to the vibration type.
Further, described to match the digital signal with the caution sound sound bank prestored, obtain matched police
The step of showing sound, including：
Caution sound information all in the digital signal and the caution sound sound bank prestored is subjected to sound characteristic one by one
Matching, obtains sound characteristic percent similarity；
Judge whether there is the caution sound information that the sound characteristic percent similarity reaches preset percentage；
If so, determining that the caution sound information is matched caution sound.
Further, described according to the caution sound sound bank matched vibration type list in advance, it calls and the warning
Before the corresponding vibration type of sound, including：
Caution sound information and vibration type Corresponding matching all in the caution sound sound bank are preset according to preset instructions
Vibration type list.
Caution sound identified off-line device proposed by the present invention, including：
Acquiring unit, for obtaining the digital signal of current sound；
Matching unit obtains matched police for matching the digital signal with the caution sound sound bank prestored
Show sound；
Call unit, for calling and the warning according to the caution sound sound bank matched vibration type list in advance
The corresponding vibration type of sound；
First execution unit, for generating corresponding first vibration according to the vibration type.
Further, caution sound identified off-line device further includes：
Analytical unit, whether the decibel value for analyzing the digital signal is more than preset decibel value；
When the decibel value of the digital signal is less than preset decibel value, the matching unit is then executed the number
Word signal is matched with the caution sound sound bank prestored, obtains matched caution sound；
Second execution unit, for when the decibel value of the digital signal is more than preset decibel value, then generating corresponding
Second vibration.
Further, second execution unit, including：
Determining module, for determining digital signal decibel value model locating in matched vibration mapping table in advance
It encloses, the vibration mapping table matched in advance includes the corresponding relationship of different decibel value ranges and vibration type；
Calling module, for calling corresponding vibration type in vibration mapping table according to locating decibel value range；
Execution module, for generating corresponding second vibration according to the vibration type.
Further, the matching unit, including：
Matching module, for by the digital signal with caution sound information all in the caution sound sound bank that prestores one by one
Sound characteristic matching is carried out, sound characteristic percent similarity is obtained；
Judgment module reaches the police of preset percentage for judging whether there is the sound characteristic percent similarity
Show that message ceases；
Determination module, for when there are the caution sound information that the sound characteristic percent similarity reaches preset percentage
When, it is determined that the caution sound information is matched caution sound.
Further, the caution sound identified off-line device further includes：
Default unit, for presetting caution sound information and vibration all in the caution sound sound bank according to preset instructions
The vibration type list of type Corresponding matching.
Beneficial effects of the present invention are：Caution sound identified off-line device is obtained after treatment by the sound outside acquisition
To the digital signal of current sound, caution sound identified off-line device by the digital signal of acquisition and the caution sound sound bank that prestores into
Row matching, when judgement obtains digital signal, there are matching relationships with some caution sound information in the caution sound sound bank that prestores
When, caution sound identified off-line device will generate corresponding first vibration according to the corresponding vibration component of vibration Type Control, to
It reminds hearing to have the people of obstacle, the information of external caution sound can be obtained convenient for the people that hearing has obstacle, to avoid one in time
The generation of a little dangerous situations.
Detailed description of the invention
Fig. 1 is the step schematic diagram of the caution sound identified off-line method in one embodiment of the invention；
Fig. 2 is the step schematic diagram of the caution sound identified off-line method in another embodiment of the present invention；
Fig. 3 is the step schematic diagram of the caution sound identified off-line method in further embodiment of this invention；
Fig. 4 is the structural schematic diagram of the caution sound identified off-line device in one embodiment of the invention；
Fig. 5 is the structural schematic diagram of the caution sound identified off-line device in another embodiment of the present invention；
Fig. 6 is the structural schematic diagram of the second execution unit of the caution sound identified off-line device in one embodiment of the invention；
Fig. 7 is the structural schematic diagram of the matching unit of the caution sound identified off-line device in one embodiment of the invention；
Fig. 8 is the structural schematic diagram of the caution sound identified off-line device in further embodiment of this invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Referring to Fig.1, the caution sound identified off-line method in the present embodiment, including：
Step S1 obtains the digital signal of current sound；
The digital signal is matched with the caution sound sound bank prestored, obtains matched caution sound by step S2；
Step S3, according to the caution sound sound bank, matched vibration type list, calling are opposite with the caution sound in advance
The vibration type answered；
Step S4 generates corresponding first vibration according to the vibration type.
In step sl, the caution sound identified off-line device in the present embodiment, which is specifically as follows earphone and is worn over hearing, obstacle
The number of people on, or wrist-watch is worn over people that hearing has obstacle on hand, can be made into waist pendant and hang over hearing having obstacle
People waistband on.Above-mentioned caution sound identified off-line device obtains current sound by the sound outside acquisition after treatment
Digital signal, for the phonetic warning that external police whistle sound, vehicle whistle sound, bicycle bell sound and other people generate,
Above-mentioned caution sound identified off-line device can pass through the electric signal for being converted to corresponding sound by acquisition above sound, corresponding
The electric signal of sound also needs to amplify processing by the band-pass filter amplifier in caution sound identified off-line device, using
A/D converter (analog-digital converter) carries out analog-to-digital conversion and obtains digital signal.
In step s 2, caution sound identified off-line device carries out above-mentioned digital signal and the caution sound sound bank that prestores
Match, judge whether above-mentioned digital signal matches with the caution sound sound bank prestored, wherein the caution sound sound bank prestored includes each
The sound such as kind police whistle sound, vehicle whistle sound, bicycle bell sound；It further include some caution sounds, wherein the caution sound can be with
For " help ", " getting out of the way ", " danger ", " vehicle come ", " murder ", " robbery ", " hello ", " can recognize you " etc., with
And other need hearing to have the caution sound identified outside the personality of obstacle.When above-mentioned digital signal and the caution sound prestored
When the similarity degree of some caution sound of sound bank is larger, the caution sound sound bank that will judge above-mentioned digital signal Yu prestore
Match, will obtain and the matched caution sound of above-mentioned digital signal.
In step s3, after obtaining caution sound matched with above-mentioned digital signal, for the caution sound sound bank prestored
In different caution sounds, will also be called and above-mentioned caution sound according to above-mentioned caution sound sound bank matched vibration type list in advance
Corresponding vibration type.Specifically, the vibration letter of different vibration types can be generated according to the degree of danger of voice messaging
Number, such as macroseism signal, the different vibration signal such as general vibration signal and light shake signal.For police whistle sound, vehicle whistle
Sound, bicycle bell sound belong to more danger signal, need to generate macroseism signal；For " help ", " getting out of the way ", " danger ", " vehicle
Come ", the voice signal of " murder " and " robbery " et al. can generate macroseism signal or general vibration signal；And for " you
It is good ", the voice signal of " can recognize you " et al., light shake signal can be generated.It is, of course, also possible to according to caution sound information
Degree of danger generates the vibration signal of other different vibration types, such as the vibration letter that long shake signal and short shake signal etc. are different
Number；For the voice signals such as police whistle sound, vehicle whistle sound, bicycle bell sound and for " help ", " getting out of the way ", " danger ",
The voice signal of " vehicle come ", " murders " and " robbery " et al., which can generate to grow, shakes signal, and for " hello ", " one can be recognized
It is lower you " et al. voice signal, short shake signal can be generated.
In step s 4, caution sound identified off-line device will be generated according to the corresponding vibration component of above-mentioned vibration Type Control
Corresponding first vibration can obtain external warning convenient for the people that hearing has obstacle to remind hearing to have the people of obstacle in time
Message breath, to avoid the generation of some dangerous situations.
Referring to Fig. 2, caution sound identified off-line method in another embodiment is described by the digital signal and the police that prestores
Show that sound sound bank is matched, before obtaining the step S2 of matched caution sound, including：
Step S21, whether the decibel value for analyzing the digital signal is more than preset decibel value.
If it is not, thening follow the steps S2, the digital signal is matched with the caution sound sound bank prestored, is matched
Caution sound.
If so, executing step S22, corresponding second vibration is generated.
After caution sound identified off-line device obtains the digital signal of current sound, the decibel of above-mentioned digital signal will be analyzed
Value whether be more than preset value decibel value, the biggish sound of the decibels such as some police whistle sounds, vehicle whistle sound is generally compared
Danger, this kind of sound be need hearing have obstacle people can timely learning sound, wherein preset decibel value can be 110 points
Shellfish.And when the decibel value of the above-mentioned digital signal of caution sound identified off-line device analysis is more than the decibel value of preset value, it can not
By matching above-mentioned digital signal with the caution sound sound bank prestored, the second vibration is generated directly to remind hearing to have barrier
The people hindered, wherein above-mentioned second vibration can be macroseism signal.For when the above-mentioned number letter of caution sound identified off-line device analysis
Number decibel value when being less than the decibel value of preset value, then execute and carry out above-mentioned digital signal and the caution sound sound bank that prestores
Match, obtain matched caution sound, certainly, when the decibel value of the above-mentioned digital signal of caution sound identified off-line device analysis is more than default
When the decibel value of value, corresponding second vibration can also be first generated, is then further continued for above-mentioned digital signal and the warning prestored
Sound sound bank is matched, and is matched when above-mentioned digital signal exists with some caution sound information in the caution sound sound bank prestored
When relationship, the matched caution sound is obtained, caution sound identified off-line device will call corresponding vibration according to the matched caution sound
Type, and corresponding first vibration is generated according to the corresponding vibration component of corresponding vibration Type Control, for replacing above-mentioned the
Two vibrations.
Caution sound identified off-line method in other embodiments, the step S22 for generating corresponding second vibration, packet
It includes：
Step S220 determines digital signal decibel value range locating in matched vibration mapping table in advance, institute
State the corresponding relationship that matched vibration mapping table in advance includes different decibel value ranges and vibration type；
Step S221 calls corresponding vibration type in vibration mapping table according to locating decibel value range；
Step S222 generates corresponding second vibration according to the vibration type.
For being more than the sound of some decibel value, for example, different points can also be preset for the sound more than 110 decibels
The digital signal and vibration type of shellfish value range carry out matched vibration mapping table, and above-mentioned matched vibration mapping table in advance includes
The corresponding relationship of different decibel value ranges and vibration type, specifically, for decibel range between 110 decibels to 130 decibels
The corresponding vibration type of interior value is short shake signal, is long shake for the corresponding vibration type of value in 130 decibels of range above
Signal.For the obtained digital signal about sound, first determine above-mentioned digital signal in preparatory matched vibration mapping table
Locating decibel value range；Corresponding vibration class is called in vibration mapping table according to decibel value range locating for above-mentioned digital signal
Type, such as the decibel value corresponding to above-mentioned digital signal are 120 decibel values, will call corresponding short shake signal, caution sound from
It is corresponding short shake that line identification device, which will control corresponding vibration component according to above-mentioned short shake signal and generate corresponding second vibration,
Signal, to remind hearing to have the people of obstacle.
Caution sound identified off-line method in other embodiments, it is described by the digital signal and the caution sound voice prestored
Library is matched, and the step S2 of matched caution sound is obtained, including：
Step S201 carries out the digital signal and caution sound information all in the caution sound sound bank that prestores one by one
Sound characteristic matching, obtains sound characteristic percent similarity；
Step S202 judges whether there is the warning message that the sound characteristic percent similarity reaches preset percentage
Breath；
Step S203, if so, determining that the caution sound information is matched caution sound.
In step s 201, for the digital signal acquired, by above-mentioned digital signal and the caution sound voice prestored
All caution sound information in library carry out sound characteristic matching one by one, the caution sound sound bank for obtaining above-mentioned digital signal and prestoring
In the corresponding sound characteristic percent similarity of all caution sound information.
In step S202, caution sound identified off-line device is by having judged whether there is the similar percentage of above sound feature
Than the caution sound information for reaching preset percentage, wherein above sound feature mainly includes some parameter informations of digital signal.
In step S203, when caution sound identified off-line device is by judging that there are above sound feature percent similarities
When reaching the caution sound information of preset percentage, it is determined that above-mentioned voice messaging is matched caution sound.
Referring to Fig. 3, caution sound identified off-line method in another embodiment is described preparatory according to the caution sound sound bank
Matched vibration type list, before calling the step S3 for shaking type corresponding with the caution sound, including：
Step S23 presets caution sound information and vibration type all in the caution sound sound bank according to preset instructions
The vibration type list of Corresponding matching.
It, can be according to preset instructions come predetermined warning sound language for the caution sound sound bank in caution sound identified off-line device
All caution sound information and vibration type carry out the vibration type list of Corresponding matching in sound library, so as to according to the warning that prestores
The vibration signal in certain caution sound acquisition of information vibration type list in sound sound bank, convenient for generating corresponding vibration.
Referring to Fig. 4, caution sound identified off-line device in the present embodiment, including：
Acquiring unit 10, for obtaining the digital signal of current sound；
Matching unit 20 obtains matched for matching the digital signal with the caution sound sound bank prestored
Caution sound；
Call unit 30, for calling and the police according to the caution sound sound bank matched vibration type list in advance
Show the corresponding vibration type of sound；
First execution unit 40, for generating corresponding first vibration according to the vibration type.
Caution sound identified off-line device in the present embodiment, which is specifically as follows earphone and is worn over hearing, to be had on the number of people of obstacle,
It can be worn over people that hearing has obstacle for wrist-watch on hand, can be made into waist pendant and hang over the waistband that hearing has the people of obstacle
On.Acquiring unit 10 in above-mentioned caution sound identified off-line device is obtained current after treatment by the sound outside acquisition
The digital signal of sound, the caveat generated for external police whistle sound, vehicle whistle sound, bicycle bell sound and other people
Sound, above-mentioned caution sound identified off-line device can pass through the electric signal for being converted to corresponding sound by acquisition above sound,
The electric signal of corresponding sound also needs to amplify processing by the band-pass filter amplifier in caution sound identified off-line device, then
Analog-to-digital conversion, which is carried out, by A/D converter (analog-digital converter) obtains digital signal.
Matching unit 20 in caution sound identified off-line device by above-mentioned digital signal and the caution sound sound bank that prestores into
Row matching, judges whether above-mentioned digital signal matches with the caution sound sound bank prestored, wherein the caution sound sound bank packet prestored
Include the sound such as various police whistle sounds, vehicle whistle sound, bicycle bell sound；It further include some caution sounds, the wherein caution sound
It can be " help ", " getting out of the way ", " danger ", " vehicle ", " murder ", " robbery ", " hello ", " can recognize you "
Deng and other need hearing to have the caution sound identified outside the personality of obstacle.When above-mentioned digital signal with prestore
When the similarity degree of some caution sound of caution sound sound bank is larger, the caution sound voice that will judge above-mentioned digital signal Yu prestore
Storehouse matching will obtain and the matched caution sound of above-mentioned digital signal.
After obtaining caution sound matched with above-mentioned digital signal, different in the caution sound sound bank that prestores are warned
Sound, call unit 30 will also call and above-mentioned caution sound phase according to above-mentioned caution sound sound bank matched vibration type list in advance
Corresponding vibration type.Specifically, the vibration letter of different vibration types can be generated according to the degree of danger of voice messaging
Number, such as macroseism signal, the different vibration signal such as general vibration signal and light shake signal.For police whistle sound, vehicle whistle
Sound, bicycle bell sound belong to more danger signal, need to generate macroseism signal；For " help ", " getting out of the way ", " danger ", " vehicle
Come ", the voice signal of " murder " and " robbery " et al. can generate macroseism signal or general vibration signal；And for " you
It is good ", the voice signal of " can recognize you " et al., light shake signal can be generated.It is, of course, also possible to according to caution sound information
Degree of danger generates the vibration signal of other different vibration types, such as the vibration letter that long shake signal and short shake signal etc. are different
Number；For the voice signals such as police whistle sound, vehicle whistle sound, bicycle bell sound and for " help ", " getting out of the way ", " danger ",
The voice signal of " vehicle come ", " murders " and " robbery " et al., which can generate to grow, shakes signal, and for " hello ", " one can be recognized
It is lower you " et al. voice signal, short shake signal can be generated.
The first execution unit 40 in caution sound identified off-line device will be according to the corresponding vibration of above-mentioned vibration Type Control
Component generates corresponding first vibration, to remind hearing to have the people of obstacle, can obtain in time convenient for the people that hearing has obstacle outer
The caution sound information in portion, to avoid the generation of some dangerous situations.
Referring to Fig. 5, caution sound identified off-line device in another embodiment further includes：
Analytical unit 21, whether the decibel value for analyzing the digital signal is more than preset decibel value.
When the decibel value of the digital signal is less than preset decibel value, then the execution of matching unit 20 will be described
Digital signal is matched with the caution sound sound bank prestored.
Second execution unit 22, for when the decibel value of the digital signal is more than preset decibel value, then generating phase
The second vibration answered.
After caution sound identified off-line device obtains the digital signal of current sound, analytical unit 21 will analyze above-mentioned number
The decibel value of signal whether be more than preset value decibel value, for the biggish sound of the decibels such as some police whistle sounds, vehicle whistle sound
Sound generally than relatively hazardous, this kind of sound be need hearing have obstacle people can timely learning sound, wherein preset decibel value
It can be 110 decibels.And when the decibel value that analytical unit 21 analyzes above-mentioned digital signal is more than the decibel value of preset value, it can be with
Not by matching above-mentioned digital signal with the caution sound sound bank prestored, second is directly generated by the second execution unit 22
Vibration is to remind hearing to have the people of obstacle, wherein above-mentioned second vibration can be macroseism signal.For being analyzed when analytical unit 21
The decibel value of above-mentioned digital signal be less than preset value decibel value when, then matching unit 20 execute by above-mentioned digital signal with prestore
Caution sound sound bank matched, obtain matched caution sound, certainly, when analytical unit 21 analyze above-mentioned digital signal point
When shellfish value is more than the decibel value of preset value, corresponding second vibration can also be first generated, then be further continued for will be upper for matching unit 20
It states digital signal to be matched with the caution sound sound bank prestored, when in above-mentioned digital signal and the caution sound sound bank prestored
For some caution sound information there are the matched caution sound when matching relationship, is obtained, caution sound identified off-line device will be according to this
The caution sound matched calls corresponding vibration type, and according to the corresponding vibration component generation of corresponding vibration Type Control corresponding the
One vibration, for replacing above-mentioned second vibration.
Referring to Fig. 6, caution sound identified off-line device in other embodiments, second execution unit 22, including：
Determining module 220, for determining digital signal decibel value locating in matched vibration mapping table in advance
Range, the vibration mapping table matched in advance include the corresponding relationship of different decibel value ranges and vibration type；
Calling module 221, for calling corresponding vibration type in vibration mapping table according to locating decibel value range；
Execution module 222, for generating corresponding second vibration according to the vibration type.
For being more than the sound of some decibel value, for example, different points can also be preset for the sound more than 110 decibels
The digital signal and vibration type of shellfish value range carry out matched vibration mapping table, and above-mentioned matched vibration mapping table in advance includes
The corresponding relationship of different decibel value ranges and vibration type, specifically, for decibel range between 110 decibels to 130 decibels
The corresponding vibration type of interior value is short shake signal, is long shake for the corresponding vibration type of value in 130 decibels of range above
Signal.For the obtained digital signal about sound, determining module 220 first determines above-mentioned digital signal in preparatory matched shake
Locating decibel value range in dynamic mapping table；The decibel value range according to locating for above-mentioned digital signal of calling module 221 is shaking
Mapping table calls corresponding vibration type, such as the decibel value corresponding to above-mentioned digital signal is 120 decibel values, will calling pair
The short shake signal answered, execution module 222 will control corresponding vibration component according to above-mentioned short shake signal and generate corresponding second shake
Dynamic i.e. corresponding short shake signal, to remind hearing to have the people of obstacle.
Caution sound identified off-line device referring to Fig. 7, in other embodiments, which is characterized in that the matching unit 20, packet
It includes：
Matching module 201, for by the digital signal and caution sound information all in the caution sound sound bank that prestores
Sound characteristic matching is carried out one by one, obtains sound characteristic percent similarity；
Judgment module 202 reaches the police of preset percentage for judging whether there is the sound characteristic percent similarity
Show that message ceases；
Determination module 203, for when there are the warnings that the sound characteristic percent similarity reaches preset percentage
When message ceases, it is determined that the caution sound information is matched caution sound.
For the digital signal acquired, matching module 201 is by above-mentioned digital signal and the caution sound sound bank that prestores
In all caution sound information carry out sound characteristic matching one by one, obtain above-mentioned digital signal in the caution sound sound bank that prestores
The corresponding sound characteristic percent similarity of all caution sound information.
Judgment module 202 is by judging whether there is the police that above sound feature percent similarity reaches preset percentage
Show that message ceases, wherein above sound feature mainly includes some parameter informations of digital signal.
When judgment module 202 is by judging that there are the warnings that above sound feature percent similarity reaches preset percentage
When message ceases, determination module 203 then determines that above-mentioned voice messaging is matched caution sound.
Referring to Fig. 8, caution sound identified off-line device in another embodiment further includes：
Default unit 23, for presetting caution sound information and shake all in the caution sound sound bank according to preset instructions
The vibration type list of dynamic type Corresponding matching.
For the caution sound sound bank in caution sound identified off-line device, presetting unit 23 can be according to preset instructions come pre-
If in caution sound sound bank all caution sound information and vibration type carry out Corresponding matching vibration type list, so as to according to
The vibration signal in certain caution sound acquisition of information vibration type list in the caution sound sound bank prestored, it is corresponding convenient for generating
Vibration.
In conclusion caution sound identified off-line device obtains current sound by the sound outside acquisition after treatment
Digital signal, caution sound identified off-line device matches the digital signal of acquisition with the caution sound sound bank prestored, when
Judgement obtains digital signal and some caution sound information in the caution sound sound bank that prestores there are when matching relationship, caution sound from
Line identification device will generate corresponding first vibration according to the corresponding vibration component of vibration Type Control, to remind hearing to have barrier
The people hindered can obtain the information of external caution sound convenient for the people that hearing has obstacle, to avoid some dangerous situations in time
Occur.
The above description is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all utilizations
Equivalent structure or equivalent flow shift made by description of the invention and accompanying drawing content is applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of caution sound identified off-line method, which is characterized in that including：
Obtain the digital signal of current sound；
The digital signal is matched with the caution sound sound bank prestored, obtains matched caution sound；
According to the caution sound sound bank matched vibration type list in advance, vibration class corresponding with the caution sound is called
Type；
Corresponding first vibration is generated according to the vibration type.
2. caution sound identified off-line method according to claim 1, which is characterized in that described by the digital signal and pre-
Before the step of caution sound sound bank deposited is matched, obtains matched caution sound, including：
Whether the decibel value for analyzing the digital signal is more than preset decibel value；
If it is not, then matching the digital signal with the caution sound sound bank prestored, matched caution sound is obtained, if so,
Then generate corresponding second vibration.
3. caution sound identified off-line method according to claim 2, which is characterized in that described to generate corresponding second vibration
The step of, including：
Determine digital signal decibel value range locating in matched vibration mapping table in advance, the matched shake in advance
Dynamic mapping table includes the corresponding relationship of different decibel value ranges and vibration type；
Corresponding vibration type is called in vibration mapping table according to locating decibel value range；
Corresponding second vibration is generated according to the vibration type.
4. caution sound identified off-line method according to claim 1, which is characterized in that described by the digital signal and pre-
The step of caution sound sound bank deposited is matched, obtains matched caution sound, including：
Caution sound information all in the digital signal and the caution sound sound bank prestored is subjected to sound characteristic matching one by one,
Obtain sound characteristic percent similarity；
Judge whether there is the caution sound information that the sound characteristic percent similarity reaches preset percentage；
If so, determining that the caution sound information is matched caution sound.
5. caution sound identified off-line method according to claim 1, which is characterized in that described according to the caution sound voice
Library matched vibration type list in advance, before calling vibration type corresponding with the caution sound, including：
The shake of caution sound information and vibration type Corresponding matching all in the caution sound sound bank is preset according to preset instructions
Dynamic type list.
6. a kind of caution sound identified off-line device, which is characterized in that including：
Acquiring unit, for obtaining the digital signal of current sound；
Matching unit obtains matched caution sound for matching the digital signal with the caution sound sound bank prestored；
Call unit, for according to the caution sound sound bank matched vibration type list in advance, calling and the caution sound phase
Corresponding vibration type；
First execution unit, for generating corresponding first vibration according to the vibration type.
7. caution sound identified off-line device according to claim 6, which is characterized in that further include：
Analytical unit, whether the decibel value for analyzing the digital signal is more than preset decibel value；
When the decibel value of the digital signal is less than preset decibel value, then the matching unit, which executes, believes the number
It number is matched with the caution sound sound bank prestored；
Second execution unit, for when the decibel value of the digital signal is more than preset decibel value, then generating corresponding the
Two vibrations.
8. caution sound identified off-line device according to claim 7, which is characterized in that second execution unit, including：
Determining module, for determining digital signal decibel value range locating in matched vibration mapping table in advance, institute
State the corresponding relationship that matched vibration mapping table in advance includes different decibel value ranges and vibration type；
Calling module, for calling corresponding vibration type in vibration mapping table according to locating decibel value range；
Execution module, for generating corresponding second vibration according to the vibration type.
9. caution sound identified off-line device according to claim 6, which is characterized in that the matching unit, including：
Matching module, for carrying out the digital signal and caution sound information all in the caution sound sound bank that prestores one by one
Sound characteristic matching, obtains sound characteristic percent similarity；
Judgment module reaches the caution sound of preset percentage for judging whether there is the sound characteristic percent similarity
Information；
Determination module, for when there are the caution sound information that the sound characteristic percent similarity reaches preset percentage, then
Determine that the caution sound information is matched caution sound.
10. caution sound identified off-line device according to claim 6, which is characterized in that further include：
Default unit, for presetting caution sound information and vibration type all in the caution sound sound bank according to preset instructions
The vibration type list of Corresponding matching.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810422085.0A CN108831502A (en) | 2018-05-04 | 2018-05-04 | Caution sound identified off-line method and device |
| PCT/CN2019/072988 WO2019210713A1 (en) | 2018-05-04 | 2019-01-24 | Warning sound offline identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810422085.0A CN108831502A (en) | 2018-05-04 | 2018-05-04 | Caution sound identified off-line method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108831502A true CN108831502A (en) | 2018-11-16 |
Family
ID=64147376
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810422085.0A Pending CN108831502A (en) | 2018-05-04 | 2018-05-04 | Caution sound identified off-line method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108831502A (en) |
| WO (1) | WO2019210713A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019210713A1 (en) * | 2018-05-04 | 2019-11-07 | 深圳市沃特沃德股份有限公司 | Warning sound offline identification method and device |
| CN112989101A (en) * | 2019-12-18 | 2021-06-18 | 深圳市三诺数字科技有限公司 | Reminding method, device, equipment and storage medium suitable for hearing-impaired patients |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101494816A (en) * | 2009-03-05 | 2009-07-29 | 胡礼斌 | Hearing-aid device and method suitable for anacusia patient |
| JP2010224392A (en) * | 2009-03-25 | 2010-10-07 | Oki Electric Ind Co Ltd | Utterance support device, method, and program |
| CN103000187A (en) * | 2011-09-19 | 2013-03-27 | 英业达股份有限公司 | Portable voice recognition system and voice recognition method |
| CN105939415A (en) * | 2016-04-14 | 2016-09-14 | Tcl集团股份有限公司 | Security alert method and system in external auditory masking state |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11327588A (en) * | 1998-05-08 | 1999-11-26 | Taguchi Tsushin Kogyo Kk | Alarm device for converting sound alarm into another transmission medium |
| US7356473B2 (en) * | 2005-01-21 | 2008-04-08 | Lawrence Kates | Management and assistance system for the deaf |
| CN102499815B (en) * | 2011-10-28 | 2013-07-24 | 东北大学 | Method for assisting deaf people to perceive environmental sound |
| WO2017073850A1 (en) * | 2015-10-26 | 2017-05-04 | 유퍼스트(주) | Notification service provision method for hearing-impaired person and device for executing same |
| CN106817661A (en) * | 2015-11-30 | 2017-06-09 | 深圳富泰宏精密工业有限公司 | Hearing assistant system and electronic installation |
| CN108831502A (en) * | 2018-05-04 | 2018-11-16 | 深圳市沃特沃德股份有限公司 | Caution sound identified off-line method and device |
-
2018
- 2018-05-04 CN CN201810422085.0A patent/CN108831502A/en active Pending
-
2019
- 2019-01-24 WO PCT/CN2019/072988 patent/WO2019210713A1/en active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101494816A (en) * | 2009-03-05 | 2009-07-29 | 胡礼斌 | Hearing-aid device and method suitable for anacusia patient |
| JP2010224392A (en) * | 2009-03-25 | 2010-10-07 | Oki Electric Ind Co Ltd | Utterance support device, method, and program |
| CN103000187A (en) * | 2011-09-19 | 2013-03-27 | 英业达股份有限公司 | Portable voice recognition system and voice recognition method |
| CN105939415A (en) * | 2016-04-14 | 2016-09-14 | Tcl集团股份有限公司 | Security alert method and system in external auditory masking state |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019210713A1 (en) * | 2018-05-04 | 2019-11-07 | 深圳市沃特沃德股份有限公司 | Warning sound offline identification method and device |
| CN112989101A (en) * | 2019-12-18 | 2021-06-18 | 深圳市三诺数字科技有限公司 | Reminding method, device, equipment and storage medium suitable for hearing-impaired patients |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019210713A1 (en) | 2019-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20200380979A1 (en) | Context aware hearing optimization engine | |
| EP1171873B1 (en) | Apparatus and methods for detecting emotions in the human voice | |
| EP3998557B1 (en) | Audio signal processing method and related apparatus | |
| WO2019084214A1 (en) | Separating and recombining audio for intelligibility and comfort | |
| US20230177942A1 (en) | Systems and methods for identifying an acoustic source based on observed sound | |
| CN111508474A (en) | Voice interruption method, electronic equipment and storage device | |
| US20180336000A1 (en) | Contextual sound filter | |
| US20160381450A1 (en) | Contextual information while using headphones | |
| CN106572411A (en) | Noise cancelling control method and relevant device | |
| CN109451390A (en) | A kind of TWS earphone and its control method, device, equipment | |
| Schroeder et al. | Detection and classification of acoustic events for in-home care | |
| WO2008062854A1 (en) | Apparatus and method for detecting sound | |
| US10531178B2 (en) | Annoyance noise suppression | |
| EP3640935B1 (en) | Notification information output method, server and monitoring system | |
| CN108831502A (en) | Caution sound identified off-line method and device | |
| US10955287B2 (en) | System and method of signal processing for use in gunshot detection | |
| CN109451385A (en) | A kind of based reminding method and device based on when using earphone | |
| US10276180B2 (en) | Audio command adaptive processing system and method | |
| CN116156439B (en) | Intelligent wearable electronic intercom interaction system | |
| CN107592129B (en) | A kind of method for early warning and device for wearable device | |
| US8635064B2 (en) | Information processing apparatus and operation method thereof | |
| KR102617126B1 (en) | The artificial intelligence smart earplug | |
| GB2563892A (en) | Sound monitoring system and method | |
| CN114821962A (en) | Triggering method and device for emergency help function, terminal and storage medium | |
| CN112802466A (en) | Artificial intelligence voice interaction service system |
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: 20181116 |
|
| RJ01 | Rejection of invention patent application after publication |