CN106128455A - Based on the speech recognition system under bone conduction high-noise environment - Google Patents
Based on the speech recognition system under bone conduction high-noise environment Download PDFInfo
- Publication number
- CN106128455A CN106128455A CN201610792166.0A CN201610792166A CN106128455A CN 106128455 A CN106128455 A CN 106128455A CN 201610792166 A CN201610792166 A CN 201610792166A CN 106128455 A CN106128455 A CN 106128455A
- Authority
- CN
- China
- Prior art keywords
- module
- voice
- signal
- speech recognition
- speech
- 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
- 210000000988 bone and bone Anatomy 0.000 title claims abstract description 15
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000004891 communication Methods 0.000 claims abstract description 19
- 238000001914 filtration Methods 0.000 claims abstract description 18
- 238000006243 chemical reaction Methods 0.000 claims abstract description 16
- 102000001253 Protein Kinase Human genes 0.000 claims abstract description 4
- 108060006633 protein kinase Proteins 0.000 claims abstract description 4
- 230000001105 regulatory effect Effects 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 19
- 238000000034 method Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 14
- 230000001276 controlling effect Effects 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 6
- 238000003672 processing method Methods 0.000 claims description 4
- 238000003860 storage Methods 0.000 claims description 4
- 230000003321 amplification Effects 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 abstract description 13
- 238000001228 spectrum Methods 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000013461 design Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000013144 data compression Methods 0.000 description 3
- 230000010365 information processing Effects 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 230000003750 conditioning effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004899 motility Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- 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
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
Abstract
The invention discloses a kind of based on the speech recognition system under bone conduction high-noise environment, it includes transmitter, speech reception module, drive circuit module, logic switch amount I/O interface, controls setting module, duty display module, communication interface modules, D/A conversion module, pid control module, AD conversion module, DSP module, speech processing module, digital filtering module, audio playback module, Signal-regulated kinase, receiver.The functions such as the present invention has noise suppressed, volume is automatically adjusted, voice stores, speech recognition, utilize the digital signal processing capability that DSP is powerful, complete source of sound ripple spectrum component analysis, then being filtered according to desired frequency characteristic, the closed-loop volume that voice signal amplitude carries out discrete magnitude simultaneously is automatically adjusted.
Description
Technical field
The present invention relates to a kind of speech recognition system, particularly relate to a kind of based on the voice under bone conduction high-noise environment
Identification system.
Background technology
Voice is a kind of mode of Human communication's most convenient, then can bring great convenience, at present as realized operation with voice
General speech recognition system, uses analog filtering and MCU (single-chip microcomputer) speech processes, owing to analog filtering device volume is big,
Filter noise range to be restricted by component parameters, adjust inconvenience, filter noise range the most influenced by ambient temperature, and several
Word voice stores, speech recognition technology, is affected greatly by environment noise and volume, and the weak effect of speech recognition causes and identifies
Mistake, causes operation or controls error.
Summary of the invention
For defect of the prior art, it is an object of the invention to provide a kind of based on the language under bone conduction high-noise environment
Sound identification system, its overcome existing apparatus filtering adjust inconvenience, poor universality, volume is big, structure is complicated, by environment noise and
The impact of volume is big, the weak effect of speech recognition, easily cause identification and make mistakes, cause operation or control the problems such as error.
According to an aspect of the present invention, it is provided that a kind of based on the speech recognition system under bone conduction high-noise environment, its
It is characterised by, comprising:
Transmitter, is used for sending voice;
Speech reception module, is used for receiving training voice or identification voice;Receive training voice, store the voice of user
Order;Receive identification voice, and multiple training speech comparison identifications, and by function setting, exported not by switching value I/O interface
Same operation logic signal;
Drive circuit module, for the signal of amplification control circuit;
Logic switch amount I/O interface, for performing different operation, control circuit realizes the control to plurality of devices, is used for
Pressure, flow, temperature are adjusted;
Control setting module, be used for controlling, setting work;
Duty display module, is used for showing whole system working condition;
Communication interface modules, is used for connecting communication equipment;
Speech processing module, is used for processing voice signal;Speech processing module includes interconnective audio playback module
And DSP module, audio playback module is used for realizing voice communication clearly or after to operational order identification, enters by function setting
During row corresponding operating, for improving operation accuracy further, after each instruction receives, system can pass through audio playback feedback
To user, it is simple to confirm in time;DSP module is used for receiving training voice or identification voice;Environment noise is filtered, disappears
Make an uproar and the closed-loop volume of voice signal is automatically adjusted, make voice not affected by environment noise and volume, it is achieved to communicate clearly
Or correct voice training stores and speech recognition;DSP module includes D/A conversion module, digital filtering module, pid control module
And AD conversion module, D/A conversion module is used for converting digital signals into analogue signal;Digital filtering module is for entering signal
Row processes;Pid control module is used for measuring, compare and perform actual value and expected value;AD conversion module is for by analogue signal
Be converted to digital signal;
Signal-regulated kinase, makes voice signal reach A/D requirement;
Receiver, is used for receiving voice.
Preferably, the work process that described DSP module comprises the following steps:
Step one, receives voice signal;
Step 2, carries out DA conversion;
Step 3, carries out digital filtering process;
Step 4, storage identifies signal;
Step 5, through switching value IO, is adjusted controlling;
Step 6, is AD converted and plays back.
Preferably, described DSP module uses mathematical operation processing method to realize correct identification, and carries out corresponding operating.
Preferably, described DSP module uses TI company's T MS320LF2407A controller, and TMS320LF2407A is with general
CPU compares with microprocessor, and price is low, be widely used, have the strongest digital information processing function, and it collects C2xx kernel enhancement mode
TMS320 design structure and low-power consumption, high-performance, optimization peripheral circuit are in one, and 16 position digital signals process, with hardware multiplication
Musical instruments used in a Buddhist or Taoist mass and multiply-add instruction, every instruction completes at 25ns, and voice is changed at digital information through 10 speed A/D converter voices
Reason ability as basis invention provides suitable hardware foundation;DSP has program control property, repeatability, good in anti-interference performance, it is possible to realize certainly
Adaptive algorithm, data compression, it has the irreplaceable characteristic of analogue signal.
Preferably, described pid control module has closed-loop volume and is automatically adjusted, and closed-loop volume is automatically adjusted employing discrete magnitude
PID regulates, and makes voice not affected by environment noise and volume.
Preferably, described pid control module has voice training, for carrying out point from the DSP speech samples to collecting
During analysis processes, extract voice characteristics information, set up a characteristic model, be stored in memorizer.
Preferably, described pid control module has speech recognition, for carrying out class from the DSP speech samples to collecting
As in analyzing and processing, extract the characteristic information of voice, then this feature information model believed with existing characteristic model
Number contrasting, if the two has reached certain matching degree, then the voice inputted is identified, and acts accordingly.
The most progressive effect of the present invention is: the present invention, without any analog filter, device fully-integratedization, uses
DSP module and correlation technique, device performance is excellent;Saying from economic angle, what the present invention used is only one piece of general dsp mould
Block and the peripheral circuit of a small amount of normal electronics composition, owing to using general electrical components, device cost is cheap, from application
Angle is said, voice is a kind of mode of Human communication's most convenient, realizes communication clearly with voice or corresponding operation has very
Wide application prospect.The present invention is easy to adjust, and versatility is good, and volume is little, simple in construction, is not affected by environment noise and volume,
Speech recognition effective, does not results in identification and makes mistakes, cause operation or control error.The present invention has noise suppressed, volume
Be automatically adjusted, voice storage, the function such as speech recognition, utilize the digital signal processing capability that DSP is powerful, complete source of sound ripple frequency spectrum
Component analysis, is then filtered according to desired frequency characteristic, and voice signal amplitude carries out the closed loop of discrete magnitude simultaneously
Volume is automatically adjusted.
Accompanying drawing explanation
By the detailed description non-limiting example made with reference to the following drawings of reading, the further feature of the present invention,
Purpose and advantage will become more apparent upon:
Fig. 1 is present invention principle framework based on the speech recognition system under bone conduction high-noise environment figure.
Fig. 2 is the flow chart of the work process of pid control module in the present invention.
Detailed description of the invention
Below in conjunction with specific embodiment, the present invention is described in detail.Following example will assist in the technology of this area
Personnel are further appreciated by the present invention, but limit the present invention the most in any form.It should be pointed out that, the ordinary skill to this area
For personnel, without departing from the inventive concept of the premise, it is also possible to make some deformation and improvement.These broadly fall into the present invention
Protection domain.
As it is shown in figure 1, the invention discloses a kind of based on the speech recognition system under bone conduction high-noise environment, it includes
Transmitter, speech reception module, drive circuit module, logic switch amount IO (input and output) interface, control setting module, work
State display module, communication interface modules, DA (digital-to-analogue) modular converter, PID (proportional-integral-differential) control module, AD (mould
Number) modular converter, DSP (Digital Signal Processing) module, speech processing module, digital filtering module, audio playback module, signal
Conditioning module, receiver, wherein:
Transmitter, is used for sending voice;
Speech reception module, is used for receiving training voice or identification voice;Receive training voice, store the voice of user
Order;Receive identification voice, and multiple training speech comparison identifications, and by function setting, exported not by switching value I/O interface
Same operation logic signal;
Drive circuit module, for the signal of amplification control circuit;
Logic switch amount I/O interface, for performing different operation, control circuit realizes the control to plurality of devices, is used for
Pressure, flow, temperature are adjusted;
Control setting module, be used for controlling, setting work;
Duty display module, is used for showing whole system working condition;
Communication interface modules, is used for connecting communication equipment;
Speech processing module, is used for processing voice signal;Speech processing module includes interconnective audio playback module
And DSP module, audio playback module is used for realizing voice communication clearly or after to operational order identification, enters by function setting
During row corresponding operating, for improving operation accuracy further, after each instruction receives, system can pass through audio playback feedback
To user, it is simple to confirm in time, control accuracy is made to reach 100%;
DSP module is used for receiving training voice or identification voice;Environment noise is filtered, de-noising and voice signal
Closed-loop volume is automatically adjusted, and makes voice not affected by environment noise and volume, it is achieved communication clearly or correct voice instruction
Practice and store and speech recognition;DSP module includes D/A conversion module, digital filtering module, pid control module and AD conversion module,
D/A conversion module is used for converting digital signals into analogue signal;Digital filtering module is for processing signal;PID controls
Module is used for measuring, compare and perform actual value and expected value;AD conversion module is used for converting analog signals into digital signal;
After speech recognition, it is achieved multiple operating function, operate for different phonetic order or control;For just improving further
Really rate, after each instruction receives, system can feed back to user by audio playback, it is simple to confirms in time, makes control just
Really rate reaches 100%, and is operated to control interface to perform next step with logical order by system;Whole system working condition by
Display indicates;Whole system can also be applied to multi-platform control by communication interface, can move towards industrial robot and control to answer
Use field;
Signal-regulated kinase, makes voice signal reach A/D requirement;
Receiver, is used for receiving voice.
As in figure 2 it is shown, the work process that DSP module comprises the following steps:
Step one, receives voice signal;
Step 2, carries out DA conversion;
Step 3, carries out digital filtering process;
Step 4, storage identifies signal;
Step 5, through switching value IO, is adjusted controlling;
Step 6, is AD converted and plays back.
Described DSP module uses mathematical operation processing method to realize correct identification, and carries out corresponding operating.
Described DSP module uses TI company's T MS320LF2407A controller, TMS320LF2407A and general CPU and micro-
Processor (MCU) is compared, and price is low, be widely used, have the strongest digital information processing function, and it collects C2xx kernel enhancement mode
TMS320 design structure and low-power consumption, high-performance, optimization peripheral circuit are in one, and 16 position digital signals process, with hardware multiplication
Musical instruments used in a Buddhist or Taoist mass and multiply-add instruction, every instruction completes at 25ns, and voice is changed at digital information through 10 speed A/D converter voices
Reason ability as basis invention provides suitable hardware foundation;DSP has program control property, repeatability, good in anti-interference performance, it is possible to realize certainly
Adaptive algorithm, data compression, it has the irreplaceable characteristic of analogue signal.DSP module carries out mathematics fortune to sound source signal
Calculate processing method and reach frequency domain filtering, and voice signal amplitude is carried out discrete magnitude closed-loop volume be automatically adjusted.DSP module
To the speech samples collected after filtering and volume are automatically adjusted, it is analyzed processing, improves matching degree, reach correct
Speech recognition.
There is provided on high-speed digital video camera hardware foundation in DSP module, carry out significant digits signal processing technology exploitation,
By research human speech characteristic, carry out trickle spectrum analysis, produce when various environment noises and various armament-related work simultaneously
Raw noise carries out frequency spectrum research and statistical analysis, identifies the difference of environment noise and human speech, on this basis, draws
Digital filtering transmission function, is filtered, suppresses environment noise, develop speech recognition software and carry out speech recognition;Voice is known
Other to as if voice, determine that the key parameter of source of sound quality is " sample rate ", as long as selecting sample rate > 48KHz just can reach CD
Tonequality, it is achieved the high tone quality precise acquisition to voice, it is ensured that the precision that overall data process is analyzed;General MCU is to be difficult to reach
Arrive, and made Digital Signal Processing by DSP module further, complete noise suppressed, numerical value PID control, data compression technique etc.;
Speech processing is required that the digital information processing with DSP module combines and realizes high-quality voice communications with correct
Speech recognition.
Voice signal is transformed from the time domain to frequency domain form and generally uses Fourier transform, for discretization by DSP module
For periodic signal, fast fourier transform (FFT) is one of best algorithm analyzing speech waveform, utilizes FFT energy directly
Obtain each spectrum component contained by waveform.FFT is a kind of fast algorithm of discrete Fourier transform (DFT) (DFT);Owing to we are at meter
When calculating DFT, complex multiplication need to be with four real multiplications and secondary real addition;One time complex addition then needs secondary real add
Method;Every one X (k) of computing needs 4N complex multiplication and 2N+2 (N-1)=2 (2N-1) secondary real addition;So whole DFT fortune
Calculate and altogether need 4N2 real multiplications and N*2 (2N-1)=2N (2N-1) secondary real addition;Consequently, it is possible to multiplication when calculating
Number and addition number of times are all directly proportional with N2, and when N is the biggest, operand is huge, so that improve the algorithm to DFT
Reduce arithmetic speed;Symmetry according to Fourier transform and periodicity, by DFT computing, some merges and finite term letter
Change.
If sequence length is N=2L, L is integer;By sequence x (n) of N=2L (n=0,1 ..., N-1), by the odd even of N
Being divided into two groups, say, that the DFT of a N point is resolved into the DFT of two N/2 points by us, they are combined into one again
N point DFT as expressed by following formula (1):
Above formula FFT sequence of operations is chaotic, but through analyzing, signal flow has the i.e. bit code of certain rule to be inverted, this
Bright employ bit-reversed subtract connect addressing system and multiply accumulating, shift accumulated instruction realization;
Utilize fast fourier transform (FFT) that input speech signal carries out discrete Fourier transform (DFT) and analyze its frequency spectrum, so
After utilize Finite Impulse Response filter, filter design method be utilize filtering theory design meet require transfer function H (s), depend on
Obtain corresponding digital filter transfer function H (z) according to H (s), be filtered according to desired frequency characteristic, this way
There is preferable frequency selectivity and motility.
Described pid control module has closed-loop volume and is automatically adjusted, and closed-loop volume is automatically adjusted employing discrete magnitude PID and adjusts
Joint, makes voice not affected by environment noise and volume.
Described pid control module has voice training, and voice training is for carrying out point from the DSP speech samples to collecting
During analysis processes, extract voice characteristics information, set up a characteristic model, be stored in memorizer.
Described pid control module has speech recognition, and speech recognition is for carrying out class from the DSP speech samples to collecting
As in analyzing and processing, extract the characteristic information of voice, then this feature information model believed with existing characteristic model
Number contrasting, if the two has reached certain matching degree, then the voice inputted is identified, and acts accordingly.
The present invention is by research human speech characteristic, simultaneously by producing during to the various noises of environment and various armament-related work
Raw noise carries out frequency spectrum research and statistical analysis, identifies the difference of environment noise and human speech, utilizes the number that DSP is powerful
Word signal handling capacity, completes source of sound ripple spectrum component analysis, is then filtered according to desired frequency characteristic, the most right
Voice signal amplitude carries out the closed-loop volume of discrete magnitude and is automatically adjusted, and makes voice not affected by environment noise and volume, it is achieved
High standard voice communication, by with training speech comparison, it is achieved speech recognition, and operate on request and control.Owing to adopting
With DSP Digital Signal Processing effectively, use mathematical algorithms that sound source signal is processed, eliminate the devices such as wave filter
Part, overcomes existing identification voice device poor universality, volume is big, structure is complicated, high in cost of production problem, achieves device volume
Little, simple in construction, frequency select conveniently, use the beneficial effects such as flexible, function admirable.
Above the specific embodiment of the present invention is described.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make various deformation or amendment within the scope of the claims, this not shadow
Ring the flesh and blood of the present invention.
Claims (7)
1. one kind based on the speech recognition system under bone conduction high-noise environment, it is characterised in that comprising:
Transmitter, is used for sending voice;
Speech reception module, is used for receiving training voice or identification voice;Receive training voice, store the voice life of user
Order;Receive identification voice, and multiple training speech comparison identifications, and by function setting, by switching value I/O interface output difference
Operation logic signal;
Drive circuit module, for the signal of amplification control circuit;
Logic switch amount I/O interface, for performing different operation, control circuit realizes the control to plurality of devices, for pressure
Power, flow, temperature are adjusted;
Control setting module, be used for controlling, setting work;
Duty display module, is used for showing whole system working condition;
Communication interface modules, is used for connecting communication equipment;
Speech processing module, is used for processing voice signal;Speech processing module includes interconnective audio playback module and DSP
Module, audio playback module is used for realizing voice communication clearly or after to operational order identification, carries out phase by function setting
When should operate, for improving operation accuracy further, after each instruction receives, system can feed back to make by audio playback
User, it is simple to confirm in time;DSP module is used for receiving training voice or identification voice;Environment noise is filtered, de-noising and
The closed-loop volume of voice signal is automatically adjusted, and makes voice not affected by environment noise and volume, it is achieved clearly communication or just
True voice training stores and speech recognition;DSP module includes D/A conversion module, digital filtering module, pid control module and AD
Modular converter, D/A conversion module is used for converting digital signals into analogue signal;Digital filtering module is to signal
Reason;Pid control module is used for measuring, compare and perform actual value and expected value;AD conversion module is for changing analogue signal
For digital signal;
Signal-regulated kinase, makes voice signal reach A/D requirement;
Receiver, is used for receiving voice.
2. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
The work process that DSP module comprises the following steps:
Step one, receives voice signal;
Step 2, carries out DA conversion;
Step 3, carries out digital filtering process;
Step 4, storage identifies signal;
Step 5, through switching value IO, is adjusted controlling;
Step 6, is AD converted and plays back.
3. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
DSP module uses mathematical operation processing method to realize correct identification, and carries out corresponding operating.
4. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
DSP module uses TI company's T MS320LF2407A controller.
5. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
Pid control module has closed-loop volume and is automatically adjusted, and closed-loop volume is automatically adjusted employing discrete magnitude PID regulation, makes voice not be subject to
Environment noise and the impact of volume.
6. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
Pid control module has voice training, for being analyzed processing to the speech samples collected from DSP, extracts voice
Characteristic information, sets up a characteristic model, is stored in memorizer.
7. as claimed in claim 1 based on the speech recognition system under bone conduction high-noise environment, it is characterised in that described
Pid control module has speech recognition, for from the analyzing and processing that the speech samples collected is similar to by DSP, extracts
Go out the characteristic information of voice, then this feature information model is contrasted with existing characteristic model signal, if the two
Reached certain matching degree, then the voice inputted is identified, and acts accordingly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610792166.0A CN106128455A (en) | 2016-08-31 | 2016-08-31 | Based on the speech recognition system under bone conduction high-noise environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610792166.0A CN106128455A (en) | 2016-08-31 | 2016-08-31 | Based on the speech recognition system under bone conduction high-noise environment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106128455A true CN106128455A (en) | 2016-11-16 |
Family
ID=57271377
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610792166.0A Pending CN106128455A (en) | 2016-08-31 | 2016-08-31 | Based on the speech recognition system under bone conduction high-noise environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106128455A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107886967A (en) * | 2017-11-18 | 2018-04-06 | 中国人民解放军陆军工程大学 | A kind of bone conduction sound enhancement method of depth bidirectional gate recurrent neural network |
CN109660899A (en) * | 2018-12-28 | 2019-04-19 | 广东思派康电子科技有限公司 | The bone vocal print test earphone of computer readable storage medium and the application medium |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN203311843U (en) * | 2013-07-08 | 2013-11-27 | 北京顶亮科技有限公司 | Device correctly identifying voice in loud noise environment |
-
2016
- 2016-08-31 CN CN201610792166.0A patent/CN106128455A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN203311843U (en) * | 2013-07-08 | 2013-11-27 | 北京顶亮科技有限公司 | Device correctly identifying voice in loud noise environment |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107886967A (en) * | 2017-11-18 | 2018-04-06 | 中国人民解放军陆军工程大学 | A kind of bone conduction sound enhancement method of depth bidirectional gate recurrent neural network |
CN109660899A (en) * | 2018-12-28 | 2019-04-19 | 广东思派康电子科技有限公司 | The bone vocal print test earphone of computer readable storage medium and the application medium |
CN109660899B (en) * | 2018-12-28 | 2020-06-05 | 广东思派康电子科技有限公司 | Computer readable storage medium and bone voiceprint detection earphone applying same |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101154215B (en) | Fast Fourier transform hardware structure based on three cubed 2 frequency domain sampling | |
JP6571736B2 (en) | Method and apparatus for dominant signal detection | |
CN101614760A (en) | A kind of energy monitoring apparatus | |
US11715456B2 (en) | Serial FFT-based low-power MFCC speech feature extraction circuit | |
CN106128455A (en) | Based on the speech recognition system under bone conduction high-noise environment | |
CN206558213U (en) | Based on the speech recognition system under osteoacusis complex environment | |
Moss et al. | Real-time FPGA-based anomaly detection for radio frequency signals | |
CN203311843U (en) | Device correctly identifying voice in loud noise environment | |
CN102841880B (en) | A kind of method based on the refinement of digital signal processor fast frequency spectrum | |
CN103176949B (en) | Realize circuit and the method for FFT/IFFT conversion | |
CN112669819A (en) | Extremely-low-power-consumption voice feature extraction circuit based on non-overlapping framing and serial FFT (fast Fourier transform) | |
CN107644004B (en) | Digital signal processing method and device based on discrete fractional Fourier transform fast calculation method | |
CN115238519A (en) | Research and development system for digital signal processing algorithm | |
WO2008123317A1 (en) | Signal processing device, echo canceller, and signal processing method | |
Ibrahim et al. | The evolution of digital signal processors | |
CN103926567B (en) | high-speed real-time pulse compression algorithm | |
CN108510996A (en) | A kind of iteratively faster adaptive filter method | |
US7904497B2 (en) | Hardware arithmetic engine for lambda rule computations | |
Arslan et al. | A design of DSPIC based signal monitoring and processing system | |
KR20100088513A (en) | Multi-channel down-converter and multi-channel down-converting method for generating narrowband channel using fast fourier transform | |
CN204480660U (en) | A kind of Computer Distance Education system | |
CN110808935B (en) | Accurate and efficient implementation method and device for autocorrelation operation of linear frequency modulation signal | |
Oshana | Overview of digital signal processing algorithms, part II-Part 10 in a series of tutorials in instrumentation and measurement | |
CN101546560A (en) | Audio coding and decoding device and coding and decoding method | |
CN103811011B (en) | Audio sine wave detection method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161116 |
|
RJ01 | Rejection of invention patent application after publication |