CN107945798B - Intelligent voice recognition touch switch and control method thereof - Google Patents
Intelligent voice recognition touch switch and control method thereof Download PDFInfo
- Publication number
- CN107945798B CN107945798B CN201711359889.2A CN201711359889A CN107945798B CN 107945798 B CN107945798 B CN 107945798B CN 201711359889 A CN201711359889 A CN 201711359889A CN 107945798 B CN107945798 B CN 107945798B
- Authority
- CN
- China
- Prior art keywords
- voice
- operational amplifier
- resistor
- module
- standard
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims description 12
- 238000001914 filtration Methods 0.000 claims abstract description 19
- 239000003990 capacitor Substances 0.000 claims description 37
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000013507 mapping Methods 0.000 claims description 17
- 239000013598 vector Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 239000012634 fragment Substances 0.000 claims description 3
- 238000002347 injection Methods 0.000 claims description 3
- 239000007924 injection Substances 0.000 claims description 3
- 238000012806 monitoring device Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 239000000203 mixture Substances 0.000 claims 1
- 238000012544 monitoring process Methods 0.000 claims 1
- 239000002994 raw material Substances 0.000 claims 1
- 230000007547 defect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
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
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques 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
-
- 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
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Signal Processing (AREA)
- Analogue/Digital Conversion (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a voice intelligent recognition touch switch, which comprises a voice acquisition module, a voice recognition module and a voice recognition module, wherein the voice acquisition module is used for acquiring voice signals; the filtering module is used for filtering the collected voice signals; the database module is used for storing standard voice commands; the voice recognition module is used for comparing and recognizing the voice signal with the standard voice command in the database module; and the execution module executes the corresponding command according to the recognition result of the voice recognition module. The invention can improve the defects of the prior art and improve the accuracy of the human voice recognition.
Description
Technical Field
The invention relates to the technical field of intelligent switches, in particular to a voice intelligent recognition touch switch and a control method thereof.
Background
Along with the popularization of intelligent household products, products of the voice intelligent identification switch are diversified in the market. However, the existing voice intelligent identification switch has low identification precision, and in order to improve the identification precision, a recording implementation means is usually used, which leads to poor user experience and inconvenient use.
Disclosure of Invention
The invention aims to provide a voice intelligent recognition touch switch and a control method thereof, which can solve the defects of the prior art and improve the accuracy of voice recognition.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows.
A human voice intelligent identification touch switch comprises,
the voice acquisition module is used for acquiring voice signals;
the filtering module is used for filtering the collected voice signals;
the database module is used for storing standard voice commands;
the voice recognition module is used for comparing and recognizing the voice signal with the standard voice command in the database module;
and the execution module executes the corresponding command according to the recognition result of the voice recognition module.
Preferably, the input terminal of the filter module is connected to the positive input terminal of the first operational amplifier through a first capacitor and a first resistor which are connected in series, the negative input terminal of the first operational amplifier is grounded through a second resistor, the first capacitor and the first resistor are grounded through a third resistor and a fourth resistor which are connected in series, the third resistor and the fourth resistor are grounded through a second capacitor, the output terminal of the first operational amplifier is grounded through a fifth resistor, the third resistor and the fourth resistor are connected to the positive input terminal of the second operational amplifier through a sixth resistor, the negative input terminal of the second operational amplifier is grounded through a seventh resistor, the output terminal of the second operational amplifier is connected to the negative input terminal of the third operational amplifier through an eighth resistor, the positive input terminal of the third operational amplifier is grounded through a ninth resistor, the output terminal of the third operational amplifier is connected to the positive input terminal of the second operational amplifier through a tenth resistor, and the negative input terminal of the third operational amplifier is connected to the output terminal of the third operational amplifier through a third capacitor, the output end of the first operational amplifier is connected to the positive input end of a fourth operational amplifier through an eleventh resistor, the positive input end of the fourth operational amplifier is grounded through a twelfth resistor, the output end of the second operational amplifier is connected to the negative input end of the fourth operational amplifier through a thirteenth resistor, the output ends of the first operational amplifier and the second operational amplifier are grounded through a fourth capacitor, the negative input end of the fourth operational amplifier is connected to the output end of the fourth operational amplifier through a fourteenth resistor, the output end of the fourth operational amplifier is connected to the positive input end of a fifth operational amplifier through a fifteenth resistor and a fifth capacitor which are connected in series, the fifteenth resistor and the fifth capacitor are grounded through a sixth capacitor, the positive input ends of the fifth capacitor and the fifth operational amplifier are grounded through a sixteenth resistor, the fifteenth resistor and the fifth capacitor are connected to the output end of the fifth operational amplifier through a seventeenth resistor, and the negative input end of the fifth operational amplifier is grounded through an eighteenth resistor, the output end of the fifth operational amplifier is connected to the inverting input end of the fifth operational amplifier through a nineteenth resistor, the inverting input end of the fourth operational amplifier is connected to the collector electrode of the triode, the collector electrode of the triode is connected to the base electrode of the triode through a twentieth resistor, the emitter electrode of the triode is connected between the fifth capacitor and a seventeenth resistor, the emitter electrode of the triode is grounded through a seventh capacitor, and the output end of the fifth operational amplifier is connected to the output end through a twenty-first resistor.
A control method of the intelligent human voice recognition touch switch comprises the following steps:
A. the voice acquisition module acquires a voice signal;
B. the collected voice signals enter a voice recognition module after being processed by a filtering module;
C. the voice recognition module decomposes the voice signal, compares the voice signal with a standard voice command in the database module, sends the voice command to the execution module for execution if the same voice command is compared, and then waits for the next voice signal, and directly waits for the next voice signal if the same voice command is not compared.
Preferably, the voice acquisition module performs pre-recognition on the acquired voice signals, identifies repeated signal segments in the voice signals, counts the repetition rate of the signal segments, and eliminates the voice signal segments with the repetition rate exceeding a threshold value or the voice signal segments with the repetition frequency exceeding the threshold value; the repetition rate is as follows,
wherein P is the repetition rate, n is the repetition times, and t is the statistical time.
Preferably, the threshold value of the repetition rate is 2-5 times/second, and the threshold value of the repetition times is 10-20 times.
Preferably, the database module optimizes the stored standard voice command according to the comparison result made by the voice recognition module; firstly, capturing a voice command recognized by a voice recognition module, comparing the recognized voice command with a standard voice command, and temporarily storing the deviation value of a sound wave model of the voice command in a time domain form; secondly, a nearest neighbor algorithm model is established, then the distance between a training sample and a test sample in the model is optimized,
wherein the content of the first and second substances,
the minimum value is summed for the distance,
in order to train the sample to be trained,
in order to test the sample, the sample is,
correcting the training sample according to the minimum distance value; thirdly, obtaining an optimized sound wave model deviation value by utilizing the nearest neighbor algorithm model, and modifying the original standard voice command by using the obtained sound wave model deviation value to obtain the optimized standard voice command.
Preferably, in step C, the recognizing the voice signal by the voice recognition module comprises the following steps,
c1, extracting the characteristic vectors of the standard voice commands in the database module to form a standard matrix, and then establishing a mapping function between different characteristic vectors;
c2, decomposing the voice signal, extracting the feature vectors of the decomposed voice fragments to form a comparison matrix, and then establishing a mapping function between different feature vectors;
c3, comparing the comparison matrix with the standard matrix to obtain the deviation degree of the same comparison matrix and different standard matrices and the corresponding linear deviation degree of the mapping function;
and C4, carrying out weighted summation on the matrix deviation degree and the linear deviation degree of the mapping function, if a standard voice command with a summation result smaller than a judgment threshold value exists, regarding the standard voice command with the minimum summation result as an execution command, and if the standard voice command with the summation result smaller than the judgment threshold value does not exist, ending the recognition and waiting for the next voice signal.
Preferably, the weighted value of the matrix deviation degree is 0.65, and the weighted value of the mapping function linear deviation degree is 0.35.
Adopt the beneficial effect that above-mentioned technical scheme brought to lie in: the invention establishes the standard voice command database, does not need a user to input the voice command, and is convenient for the user to use. The voice signal is filtered by the voice acquisition module and the filtering module, so that background noise and interference signals contained in the voice signal can be effectively removed. The filtering module can effectively eliminate the pulse interference contained in the audio by designing two audio processing channels, and the strength and the distortion degree of the output audio signal are ensured. The database module can effectively improve the recognition degree of the voice command by optimizing the standard voice command in real time. When a voice signal is identified, the deviation degree of the voice signal from a standard matrix and the linear deviation degree of a mapping function are used as identification objects, an optimized identification target value is obtained through weighted summation, and the problem that a single identification target easily causes large identification errors is solved.
Drawings
FIG. 1 is a schematic diagram of one embodiment of the present invention.
Fig. 2 is a circuit diagram of a filtering module according to an embodiment of the present invention.
In the figure: 1. a voice acquisition module; 2. a filtering module; 3. a database module; 4. a voice recognition module; 5. an execution module; IN, input end; OUT, an output terminal; r1, a first resistor; r2, a second resistor; r3, third resistor; r4, fourth resistor; r5, fifth resistor; r6, sixth resistor; r7, seventh resistor; r8, eighth resistor; r9, ninth resistor; r10, tenth resistor; r11, eleventh resistor; r12, twelfth resistor; r13, thirteenth resistor; r14, fourteenth resistance; r15, fifteenth resistor; r16, sixteenth resistor; r17, seventeenth resistor; r18, eighteenth resistor; r19, nineteenth resistor; r20, twentieth resistor; r21, twenty-first resistance; c1, a first capacitance; c2, a second capacitor; c3, a third capacitance; c4, a fourth capacitance; c5, a fifth capacitance; c6, a sixth capacitor; c7, a seventh capacitance; a1, a first operational amplifier; a2, a second operational amplifier; a3, a third operational amplifier; a4, a fourth operational amplifier; a5, a fifth operational amplifier; q, a triode; VCC, feedback signal injection end.
Detailed Description
The standard parts used in the invention can be purchased from the market, the special-shaped parts can be customized according to the description and the description of the attached drawings, and the specific connection mode of each part adopts the conventional means of mature bolts, rivets, welding, sticking and the like in the prior art, and the detailed description is not repeated.
Referring to fig. 1-2, one embodiment of the present invention includes a voice acquisition module 1 for acquiring a voice signal;
the filtering module 2 is used for filtering the collected voice signals;
the database module 3 is used for storing standard voice commands;
the voice recognition module 4 is used for comparing and recognizing the voice signal with the standard voice command in the database module 3;
and the execution module 5 executes the corresponding command according to the recognition result of the voice recognition module 4.
An input terminal IN of the filter module 2 is connected to a positive-phase input terminal of the first operational amplifier a1 through a first capacitor C1 and a first resistor R1 connected IN series, an inverting input terminal of the first operational amplifier a1 is connected to ground through a second resistor R2, a voltage between the first capacitor C1 and the first resistor R1 is connected to ground through a third resistor R3 and a fourth resistor R4 connected IN series, a voltage between the third resistor R3 and the fourth resistor R4 is connected to ground through a second capacitor C2, an output terminal of the first operational amplifier a1 is connected to ground through a fifth resistor R5, a voltage between the third resistor R3 and the fourth resistor R4 is connected to a positive-phase input terminal of the second operational amplifier a4 through a sixth resistor R4, an inverting input terminal of the second operational amplifier a4 is connected to ground through a seventh resistor R4, an output terminal of the second operational amplifier a4 is connected to a positive-phase input terminal of the third operational amplifier a4 through an eighth resistor R4, a positive-phase input terminal of the third operational amplifier a4 is connected to a tenth positive-phase input terminal of the ninth operational amplifier 4, an inverting input terminal of the third operational amplifier A3 is connected to an output terminal of the third operational amplifier A3 through a third capacitor C3, an output terminal of the first operational amplifier a1 is connected to a non-inverting input terminal of the fourth operational amplifier A4 through an eleventh resistor R11, a non-inverting input terminal of the fourth operational amplifier A4 is grounded through a twelfth resistor R12, an output terminal of the second operational amplifier a2 is connected to an inverting input terminal of the fourth operational amplifier A4 through a thirteenth resistor R13, an output terminal of the first operational amplifier a1 and an output terminal of the second operational amplifier a2 are grounded through a fourth capacitor C4, an inverting input terminal of the fourth operational amplifier A4 is connected to an output terminal of the fourth operational amplifier A4 through a fourteenth resistor R14, an output terminal of the fourth operational amplifier A4 is connected to a non-inverting input terminal of the fifth operational amplifier a 15 through a fifteenth resistor R15 and a fifth capacitor C15 connected in series, a positive inverting input terminal of the fifth operational amplifier a 15 is grounded through a sixteenth resistor R15 and a sixteenth capacitor C15, the fifteenth resistor R15 and the fifth capacitor C5 are connected to the output end of the fifth operational amplifier A5 through a seventeenth resistor R17, the inverting input end of the fifth operational amplifier A5 is grounded through an eighteenth resistor R18, the output end of the fifth operational amplifier A5 is connected to the inverting input end of the fifth operational amplifier A5 through a nineteenth resistor R19, the inverting input end of the fourth operational amplifier A4 is connected to the collector of the triode Q, the collector of the triode Q is connected to the base of the triode Q through a twentieth resistor R20, the emitter of the triode Q is connected between the fifth capacitor C5 and the seventeenth resistor R17, the emitter of the triode Q is grounded through a seventh capacitor C7, and the output end of the fifth operational amplifier A5 is connected to the output end OUT through a twenty-first resistor R21.
In addition, on the basis of the filter circuit, a voice signal monitoring device is arranged on the voice recognition module 4 to monitor the voice signal after filtering, and when an unfiltered continuous interference signal occurs, a feedback compensation signal is injected through a feedback signal injection end VCC arranged between the fifteenth resistor R15 and the fifth capacitor C5, so that feedback compensation is performed on the voice signal, and the problem that the filter circuit has a poor dynamic interference filtering effect is solved.
A control method of the intelligent human voice recognition touch switch comprises the following steps:
A. the voice acquisition module 1 acquires voice signals;
B. the collected voice signals are processed by the filtering module 2 and then enter the voice recognition module 4;
C. the voice recognition module 4 decomposes the voice signal, compares the voice signal with a standard voice command in the database module 3, sends the voice command to the execution module 5 for execution if the same voice command is compared, then waits for the next voice signal, and directly waits for the next voice signal if the same voice command is not compared.
The voice acquisition module 1 pre-identifies the acquired voice signals, identifies repeated signal segments in the voice signals, counts the repetition rate of the signal segments, and eliminates the voice signal segments with the repetition rate exceeding a threshold value or the voice signal segments with the repetition times exceeding the threshold value; the repetition rate is as follows,
wherein P is the repetition rate, n is the repetition times, and t is the statistical time.
The threshold value of repetition rate is 3 times/second, and the threshold value of repetition times is 15 times.
The database module 3 optimizes the stored standard voice command according to the comparison result made by the voice recognition module 4; firstly, capturing a voice command recognized by the voice recognition module 4, comparing the recognized voice command with a standard voice command, and temporarily storing the deviation value of the sound wave model of the voice command according to a time domain form; secondly, a nearest neighbor algorithm model is established, then the distance between a training sample and a test sample in the model is optimized,
wherein the content of the first and second substances,
the minimum value is summed for the distance,
in order to train the sample to be trained,
in order to test the sample, the sample is,
correcting the training sample according to the minimum distance value; thirdly, obtaining an optimized sound wave model deviation value by utilizing the nearest neighbor algorithm model, and modifying the original standard voice command by using the obtained sound wave model deviation value to obtain the optimized standard voice command.
In step C, the step of recognizing the voice signal by the voice recognition module 4 comprises the following steps,
c1, extracting the characteristic vectors of the standard voice commands in the database module 3 to form a standard matrix, and then establishing a mapping function between different characteristic vectors;
c2, decomposing the voice signal, extracting the feature vectors of the decomposed voice fragments to form a comparison matrix, and then establishing a mapping function between different feature vectors;
c3, comparing the comparison matrix with the standard matrix to obtain the deviation degree of the same comparison matrix and different standard matrices and the corresponding linear deviation degree of the mapping function;
and C4, carrying out weighted summation on the matrix deviation degree and the linear deviation degree of the mapping function, if a standard voice command with a summation result smaller than a judgment threshold value exists, regarding the standard voice command with the minimum summation result as an execution command, and if the standard voice command with the summation result smaller than the judgment threshold value does not exist, ending the recognition and waiting for the next voice signal.
The weight of the matrix deviation degree is 0.65, and the weight of the mapping function linear deviation degree is 0.35.
On the basis, the inventor finds that the weighting values of the matrix deviation degree and the mapping function linear deviation degree can be further optimized. And for a standard matrix formed by the optimized standard voice commands, the weight value of the deviation degree is increased, the increased amplitude is in direct proportion to the optimization times of the standard voice commands, and the maximum weight value of the matrix deviation degree is not more than 0.85.
In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, are merely for convenience of description of the present invention, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. The utility model provides a voice intelligent recognition touch switch which characterized in that: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
the voice acquisition module (1) is used for acquiring voice signals;
the filtering module (2) is used for filtering the collected voice signals;
the database module (3) is used for storing standard voice commands;
the voice recognition module (4) is used for comparing and recognizing the voice signal with the standard voice command in the database module (3);
the execution module (5) executes a corresponding command according to the recognition result of the voice recognition module (4);
an input end (IN) of the filtering module (2) is connected to a positive phase input end of a first operational amplifier (A1) through a first capacitor (C1) and a first resistor (R1) which are connected IN series, an inverting input end of the first operational amplifier (A1) is grounded through a second resistor (R2), a position between the first capacitor (C1) and the first resistor (R1) is grounded through a third resistor (R3) and a fourth resistor (R4) which are connected IN series, a position between the third resistor (R3) and the fourth resistor (R4) is grounded through a second capacitor (C2), an output end of the first operational amplifier (A1) is grounded through a fifth resistor (R5), a position between the third resistor (R3) and the fourth resistor (R4) is connected to an input end of the second operational amplifier (A2) through a sixth resistor (R6), an input end of the second operational amplifier (A2) is connected to an inverting input end of the eighth operational amplifier (A599) through a seventh resistor (R5928), the positive input terminal of the third operational amplifier (A3) is grounded through a ninth resistor (R9), the output terminal of the third operational amplifier (A3) is connected to the positive input terminal of the second operational amplifier (A2) through a tenth resistor (R10), the inverting input terminal of the third operational amplifier (A3) is connected to the output terminal of the third operational amplifier (A3) through a third capacitor (C3), the output terminal of the first operational amplifier (A1) is connected to the positive input terminal of the fourth operational amplifier (A4) through an eleventh resistor (R11), the positive input terminal of the fourth operational amplifier (A4) is grounded through a twelfth resistor (R12), the output terminal of the second operational amplifier (A2) is connected to the inverting input terminal of the fourth operational amplifier (A4) through a thirteenth resistor (R13), the output terminal of the first operational amplifier (A1) and the output terminal of the second operational amplifier (A2) are grounded through a fourth resistor (C4), and the inverting input terminal of the fourth operational amplifier (A4) is connected to the fourteenth input terminal of the fourth operational amplifier (A5857323), an output end of a fourth operational amplifier (A4) is connected to a positive input end of a fifth operational amplifier (A5) through a fifteenth resistor (R15) and a fifth capacitor (C5) which are connected in series, the fifth resistor (R15) and the fifth capacitor (C5) are grounded through a sixth capacitor (C6), positive input ends of the fifth capacitor (C5) and the fifth operational amplifier (A5) are grounded through a sixteenth resistor (R16), an output end of the fifth operational amplifier (A15) and the fifth capacitor (C5) are connected to an output end of a fifth operational amplifier (A5) through a seventeenth resistor (R17), an inverting input end of the fifth operational amplifier (A5) is grounded through an eighteenth resistor (R18), an output end of the fifth operational amplifier (A5) is connected to an inverting input end of a fifth operational amplifier (A5) through a nineteenth resistor (R19), an inverting input end (A4) of the fourth operational amplifier (A4) is connected to an inverting input end of a triode (Q), and a collector (Q) is connected to a collector (Q) of a triode (20), an emitter of the triode (Q) is connected between the fifth capacitor (C5) and the seventeenth resistor (R17), the emitter of the triode (Q) is grounded through the seventh capacitor (C7), and the output end of the fifth operational amplifier (A5) is connected to the output end (OUT) through the twenty-first resistor (R21); and a voice signal monitoring device is arranged on the voice recognition module (4) and is used for monitoring the voice signal after filtering processing, and when an unfiltered continuous interference signal appears, a feedback compensation signal is injected through a feedback signal injection end (VCC) arranged between a fifteenth resistor (R15) and a fifth capacitor (C5), so that the voice signal is subjected to feedback compensation.
2. The control method of the human voice intelligent recognition touch switch as claimed in claim 1, characterized by comprising the following steps:
A. the voice acquisition module (1) acquires a voice signal;
B. the collected voice signals enter a voice recognition module (4) after being processed by a filtering module (2);
C. the voice recognition module (4) decomposes the voice signal, compares the voice signal with a standard voice command in the database module (3), sends the voice command to the execution module (5) for execution if the same voice command is compared, then waits for the next voice signal, and directly waits for the next voice signal if the same voice command is not compared.
3. The control method of the human voice intelligent recognition touch switch according to claim 2, characterized in that: the voice acquisition module (1) pre-identifies the acquired voice signals, identifies repeated signal segments in the voice signals, counts the repetition rate of the signal segments, and eliminates the voice signal segments with the repetition rate exceeding a threshold value or the voice signal segments with the repetition frequency exceeding the threshold value; the repetition rate is as follows,
wherein P is the repetition rate, n is the repetition times, and t is the statistical time.
4. The control method of the human voice intelligent recognition touch switch according to claim 3, characterized in that: the threshold value of the repetition rate is 2-5 times/second, and the threshold value of the repetition times is 10-20 times.
5. The control method of the human voice intelligent recognition touch switch according to claim 2, characterized in that: the database module (3) optimizes the stored standard voice command according to the comparison result made by the voice recognition module (4); firstly, capturing a voice command recognized by a voice recognition module (4), comparing the recognized voice command with a standard voice command, and temporarily storing the deviation value of a sound wave model of the voice command according to a time domain form; secondly, a nearest neighbor algorithm model is established, then the distance between a training sample and a test sample in the model is optimized,
wherein Q isminIs the minimum sum of distances, x' is the training sample, xnCorrecting the training sample for the test sample according to the minimum distance value; thirdly, obtaining an optimized sound wave model deviation value by utilizing the nearest neighbor algorithm model, and modifying the original standard voice command by using the obtained sound wave model deviation value to obtain the optimized standard voice command.
6. The control method of the human voice intelligent recognition touch switch according to claim 5, characterized in that: in the step C, the voice recognition module (4) recognizes the voice signal and comprises the following steps,
c1, extracting the characteristic vectors of the standard voice commands in the database module (3) to form a standard matrix, and then establishing a mapping function between different characteristic vectors;
c2, decomposing the voice signal, extracting the feature vectors of the decomposed voice fragments to form a comparison matrix, and then establishing a mapping function between different feature vectors;
c3, comparing the comparison matrix with the standard matrix to obtain the deviation degree of the same comparison matrix and different standard matrices and the corresponding linear deviation degree of the mapping function;
and C4, carrying out weighted summation on the matrix deviation degree and the linear deviation degree of the mapping function, if a standard voice command with a summation result smaller than a judgment threshold value exists, regarding the standard voice command with the minimum summation result as an execution command, and if the standard voice command with the summation result smaller than the judgment threshold value does not exist, ending the recognition and waiting for the next voice signal.
7. The control method of the human voice intelligent recognition touch switch according to claim 6, characterized in that: the weighted value of the matrix deviation degree is 0.65, and the weighted value of the mapping function linear deviation degree is 0.35.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711359889.2A CN107945798B (en) | 2017-12-18 | 2017-12-18 | Intelligent voice recognition touch switch and control method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711359889.2A CN107945798B (en) | 2017-12-18 | 2017-12-18 | Intelligent voice recognition touch switch and control method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107945798A CN107945798A (en) | 2018-04-20 |
| CN107945798B true CN107945798B (en) | 2021-06-15 |
Family
ID=61944540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711359889.2A Active CN107945798B (en) | 2017-12-18 | 2017-12-18 | Intelligent voice recognition touch switch and control method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107945798B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110473540B (en) * | 2019-08-29 | 2022-05-31 | 京东方科技集团股份有限公司 | Voice interaction method and system, terminal device, computer device and medium |
| CN111026320B (en) * | 2019-12-26 | 2022-05-27 | 腾讯科技（深圳）有限公司 | Multi-mode intelligent text processing method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2396590Y (en) * | 1999-10-13 | 2000-09-13 | 北京华日昌科贸开发有限公司 | Electronic voice switch device for semiconductor video recording system |
| JP2009253709A (en) * | 2008-04-08 | 2009-10-29 | Kenwood Corp | Signal amplifier |
| CN201589951U (en) * | 2009-09-25 | 2010-09-22 | 南京信息工程大学 | Socket type power switch with voice recognition function |
| CN204145720U (en) * | 2014-07-22 | 2015-02-04 | 歌尔声学股份有限公司 | A kind of noise canceller circuit of audio devices |
| CN106849902A (en) * | 2016-12-28 | 2017-06-13 | 宁波斯凯勒智能科技有限公司 | A kind of multi-stage filter circuit |
| CN107249116A (en) * | 2017-08-09 | 2017-10-13 | 成都全云科技有限公司 | Noise echo eliminating device based on video conference |
-
2017
- 2017-12-18 CN CN201711359889.2A patent/CN107945798B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2396590Y (en) * | 1999-10-13 | 2000-09-13 | 北京华日昌科贸开发有限公司 | Electronic voice switch device for semiconductor video recording system |
| JP2009253709A (en) * | 2008-04-08 | 2009-10-29 | Kenwood Corp | Signal amplifier |
| CN201589951U (en) * | 2009-09-25 | 2010-09-22 | 南京信息工程大学 | Socket type power switch with voice recognition function |
| CN204145720U (en) * | 2014-07-22 | 2015-02-04 | 歌尔声学股份有限公司 | A kind of noise canceller circuit of audio devices |
| CN106849902A (en) * | 2016-12-28 | 2017-06-13 | 宁波斯凯勒智能科技有限公司 | A kind of multi-stage filter circuit |
| CN107249116A (en) * | 2017-08-09 | 2017-10-13 | 成都全云科技有限公司 | Noise echo eliminating device based on video conference |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107945798A (en) | 2018-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107945798B (en) | Intelligent voice recognition touch switch and control method thereof | |
| CN102779234B (en) | A kind of cardiogram classification processing method and device | |
| CN102169690A (en) | Voice signal recognition system and method based on surface myoelectric signal | |
| CN110473547B (en) | Speech recognition method | |
| CN101766509B (en) | Real-time control method for artificial limb based on single-point acquiring muscle signals | |
| CN202383462U (en) | Electronic fence system for defining working range of automatic mower | |
| CN104409080A (en) | Voice end node detection method and device | |
| CN107153155A (en) | A kind of cable local discharge signal characteristic vector extracting method | |
| CN111973177A (en) | Sleep assisting system and method based on portable electroencephalogram equipment | |
| CN107561420A (en) | A kind of cable local discharge signal characteristic vector extracting method based on empirical mode decomposition | |
| CN109036810B (en) | Transformer active noise reduction device and control method | |
| CN110108929B (en) | Anti-interference lightning current collecting device | |
| CN109059992B (en) | On-line monitoring system and monitoring method of poultry house environment sensor | |
| CN210697629U (en) | Signal acquisition device, mobile terminal and signal analysis system | |
| CN105054941A (en) | Fetus heart rate extraction device and method based on body vibration signals | |
| CN105824321A (en) | Four-axis aircraft control system and method based on surface electromyogram signals | |
| CN204101170U (en) | Hummer testing circuit | |
| CN203101409U (en) | DSP (digital signal processor)-based intelligent grain insect detecting device | |
| CN108742695B (en) | DSP-based heart sound analyzer and analysis method thereof | |
| CN102184015A (en) | Method for quickly sorting EEG (Electroencephalo-graph) signal based on threshold analysis | |
| CN111887836A (en) | Electrocardiosignal processing method, device and equipment | |
| CN109009090A (en) | A kind of brain signal figure display methods based on histogram | |
| CN203414390U (en) | DSP Builder-based blood cell pulse signal detection and statistics device | |
| CN105982665A (en) | Method for collecting brain wave signals from human ear auditory meatus | |
| CN206818861U (en) | A kind of Radar Signal Processing System based on Ethernet |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210526 Address after: 325600 Zhanzhan Road, leshang Pioneer Park, yueqingwan port area, Yueqing City, Wenzhou City, Zhejiang Province Applicant after: ZHEJIANG IGOTO ELECTRIC Co.,Ltd. Address before: 125001 unit 1, building 2-11, LVYE villa, No.2, Lianyang Road, Lianshan District, Huludao City, Liaoning Province Applicant before: Tang Bingzhen |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |