CN117690435A - Intelligent voice recognition electric switch for curtain control - Google Patents
Intelligent voice recognition electric switch for curtain control Download PDFInfo
- Publication number
- CN117690435A CN117690435A CN202410157967.4A CN202410157967A CN117690435A CN 117690435 A CN117690435 A CN 117690435A CN 202410157967 A CN202410157967 A CN 202410157967A CN 117690435 A CN117690435 A CN 117690435A
- Authority
- CN
- China
- Prior art keywords
- user
- audio
- control
- electric signal
- circuit
- 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
- 238000013507 mapping Methods 0.000 claims abstract description 33
- 238000003062 neural network model Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 13
- 238000004891 communication Methods 0.000 claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 34
- 230000005236 sound signal Effects 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 125000004122 cyclic group Chemical group 0.000 claims description 4
- 238000011156 evaluation Methods 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 abstract description 9
- 230000006870 function Effects 0.000 description 6
- 238000013473 artificial intelligence Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention relates to an intelligent voice recognition electric switch for curtain control, which comprises a local MCU, a local storage circuit and a network communication circuit which are electrically connected, wherein the local MCU is also used for the network communication circuit to interact with an online server in an initial configuration stage, and the online server recognizes a control instruction of a user according to user audio data; the method only needs an initial configuration stage to identify a control instruction of a user according to an online server neural network model, then a mapping table between the audio electric signal characteristics of the user and the control instruction of the user is established, the mapping table is established based on accurate identification, and the mapping table is stored in a local storage circuit; in a non-initial configuration stage, namely a use stage, a mapping table between the audio electric signal characteristics of the user and the control instructions of the user is searched in a local storage circuit, so that the control instructions of the user corresponding to the audio electric signal characteristics to be identified can be determined, and interaction with a server is not needed.
Description
Technical Field
The invention belongs to the field of switches, and particularly relates to an intelligent voice recognition electric switch for curtain control.
Background
The intelligent switch for controlling the curtain is generally divided into 3 types, and the first type is that a high-performance processing chip is arranged on one local side of the switch, so that an AI algorithm can be realized. The second type is to configure a simple recording recognition and control circuit locally at the intelligent switch. The third category is to configure a simple recording recognition and control circuit locally at the intelligent switch and connect with an online server through a network, while the second and third categories are more common on the market, while the first category is rare and almost the same in research.
These classes all have certain technical drawbacks or practical problems.
For example, if an AI chip is set on the local side of the switch, this cost is relatively high, the requirement of setting the AI chip is based on that in the voice recognition process, the identification is performed by the MCU with low local performance, the judgment error is easy to occur, the voice recognition accuracy is not high enough, so some chips with relatively strong processing capability are needed to realize software control and even data processing identification, that is, according to the audio data of the user, the voice command of the user is identified by a software mode and even by an artificial intelligence mode based on a training model, and the identification is relatively accurate. However, if the AI chip is provided on one side of the electrical switch, the configuration cost is particularly high, even far beyond that of a general mobile terminal. So the first category is very rarely popularized because of higher cost and better performance.
For the second class, the switch of the second class is only provided with a simple chip at the local part of the equipment, the simple chip is a singlechip with lower data capacity, and the control cannot realize high-precision recognition algorithms such as an artificial intelligence algorithm and the like because the control cannot realize the high-precision recognition algorithm through the complex processing of voice data, so that the recognition accuracy is lower.
There is another product on the market, the third category mentioned above, that interacts locally with a remote server that implements an artificial intelligence algorithm through network communications for identification. The third category is to locally acquire the voice audio of the user before sending it to the remote server. After the identification is performed by the server, the result is returned to the local area for further processing, and the mode is equivalent to the expansion of the local operation capability by the online server. The third product still has some problems in practical application, firstly, if the users have more requests to the remote server, the remote server has limited computing resource capacity, so that the delay degree of the request return result of the users is higher, the computing requirement to the remote server is higher, and secondly, each time the users need to interact with the server when performing voice recognition, and the method cannot be used under the condition of no network. Overall, the application was proposed to solve the above problems, especially the third category of products, because of the poor application effect due to the need for high frequency remote interaction with the server.
Disclosure of Invention
The present invention is directed to an intelligent voice recognition electric switch for controlling curtains, so as to solve the problems set forth in the background art.
In order to solve the technical problems, the invention provides the following technical scheme:
the invention provides an intelligent voice recognition electric switch for curtain control, which comprises a local MCU, a local storage circuit and a network communication circuit which are electrically connected, wherein the local MCU is also electrically connected with an audio coding and decoding circuit, the audio coding and decoding circuit is electrically connected with an audio amplifying circuit, the audio amplifying circuit is electrically connected with a microphone and a loudspeaker, and the microphone is used for receiving audio corresponding to a voice instruction sent by a user and converting the voice audio into an instruction electric signal; the audio amplifying circuit is used for amplifying the instruction electric signal; the audio encoding and decoding circuit is used for converting the amplified instruction electric signals of the user into user audio data and sending the user audio data to the local MCU; the local MCU is used for identifying a control instruction of a user according to the user audio data; in the initial configuration stage, the local MCU is also used for interacting the network communication circuit with the online server, the online server identifies the control instruction of the user according to the user audio data, and the mapping table between the audio electric signal characteristics of the user and the control instruction of the user is established by analyzing the audio electric signal characteristics corresponding to the control instruction of the user through the online server, and the local MCU is also used for storing the mapping table between the audio electric signal characteristics of the user and the control instruction of the user in the local storage circuit;
in a non-initial configuration stage, the local MCU is further used for identifying user audio electric signal characteristics according to user audio data and defining the user audio electric signal characteristics as audio electric signal characteristics to be identified, and determining control instructions of a user corresponding to the audio electric signal characteristics to be identified according to a mapping table between the user audio electric signal characteristics and the control instructions of the user.
Further, the local MCU is also electrically connected with a switch driving control circuit, and the switch driving control circuit is electrically connected with an electric push rod motor and an electric scroll motor; the local MCU is used for identifying a control command of a user according to the user audio data and sending a control command to the switch driving control circuit according to the control command of the user, and the electric push rod motor and the electric scroll motor are used for executing the control command sent by the switch driving control circuit.
Further, the local MCU is further used for identifying a control instruction of a user according to the audio data of the user, generating a reply instruction according to the control instruction of the user and returning the reply instruction to the audio codec circuit, the audio codec circuit is further used for converting the reply instruction into a reply audio electric signal, and the audio amplifying circuit is further used for amplifying the reply audio electric signal; the speaker is used for playing the amplified reply audio electric signal.
Further, the system also comprises a display screen, wherein the display screen is used for outputting characters or patterns according to the local MCU control command.
Further, the remote control system also comprises a touch screen circuit, wherein the touch screen circuit is used for collecting touch screen control commands of a user and sending the touch screen control commands to the local MCU.
Further, the on-line server analyzes the audio electric signal characteristics corresponding to the control instruction of the user, specifically, performs time domain analysis on the audio electric signal corresponding to the control instruction of the user, establishes a multi-dimensional characteristic vector, and takes the multi-dimensional characteristic vector as the audio electric signal characteristics corresponding to the control instruction of the user.
Further, performing time domain analysis on an audio electric signal corresponding to a control instruction of a user to establish a multi-dimensional feature vector, which specifically comprises the steps of firstly obtaining the audio electric signal corresponding to the control instruction of the user, then processing the audio electric signal corresponding to the control instruction of the user into a time sequence, then collecting integral quantity, variation quantity and second derivative function quantity of the time sequence in a certain period, and taking the integral quantity, variation quantity and second derivative function quantity of the time sequence in the certain period as one component of the multi-dimensional feature vector to establish the multi-dimensional feature vector.
Further, establishing a mapping table between the audio electric signal characteristics of the user and the control instructions of the user specifically means that the on-line server firstly identifies the control instructions through a neural network model, then characterizes the identified control instructions with target storage variables, determines multi-dimensional characteristic vectors after the time domain analysis of the audio electric signal characteristics, and establishes the mapping table between the audio electric signal characteristics of the user and the control instructions of the user by pointing the multi-dimensional characteristic vectors to the target storage variables.
Further, the on-line server firstly identifies the specific instruction of the control instruction through the neural network model, and firstly carries out denoising, speed reduction and enhancement treatment on the user audio data so as to improve the accuracy of voice identification; converting the processed audio data into feature vectors, wherein the feature vectors need to capture tone, tone color and tone intensity in the audio data; constructing a cyclic neural network, and training a neural network model by using a voice data set so that the neural network model can identify user audio data; gradient descent and regularization are needed in the training process to improve the generalization capability of the model; and evaluating the performance of the model on the test set, adjusting and optimizing according to the evaluation result, adjusting the network structure and optimizing parameters, and applying the model to the recognition control instruction after model training is completed.
The beneficial effects are that: the method only needs an initial configuration stage to identify a control instruction of a user according to an online server neural network model, then a mapping table between the audio electric signal characteristics of the user and the control instruction of the user is established, the mapping table is established based on accurate identification, and the mapping table is stored in a local storage circuit; in a non-initial configuration stage, namely a use stage, a mapping table between the audio electric signal characteristics of a user and the control instructions of the user is searched in a local storage circuit, so that the control instructions of the user corresponding to the audio electric signal characteristics to be identified can be determined, interaction with a server is not needed, the problem that the prior problem is solved is that if more users exist, the users send requests to a remote server at the same time, the request return result delay degree of the user is higher due to limited computing resource capacity of the remote server, and the computing requirement of the remote server is higher; every time a user performs voice recognition, the user needs to interact with the server, and the user has no way to use the voice recognition system under the condition of no network, so that the recognition accuracy is ensured.
Drawings
Fig. 1 is a block diagram of the circuit components of the present application.
Detailed Description
The technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention discloses an intelligent voice recognition electric switch for curtain control, which is an intelligent voice recognition electric switch for curtain control based on the background technology, wherein the intelligent voice recognition electric switch for curtain control is locally and interactively recognized by a remote server capable of realizing an artificial intelligent algorithm through network communication, and comprises a local MCU (micro control unit), a local storage circuit and a network communication circuit which are electrically connected, wherein the local MCU is also electrically connected with an audio encoding and decoding circuit, the audio encoding and decoding circuit is electrically connected with an audio amplifying circuit, the audio amplifying circuit is electrically connected with a microphone and a loudspeaker, the local MCU is also electrically connected with a switch driving control circuit, and the switch driving control circuit is electrically connected with an electric push rod motor and an electric scroll motor;
the microphone is used for receiving audio corresponding to voice instructions sent by a user and converting the voice audio into instruction electric signals; the audio amplifying circuit is used for amplifying the instruction electric signal;
the audio encoding and decoding circuit is used for converting the amplified instruction electric signals of the user into user audio data and sending the user audio data to the local MCU, the local MCU is used for identifying control instructions of the user according to the user audio data and sending control commands to the switch driving control circuit according to the control instructions of the user, and the electric push rod motor and the electric scroll motor are used for executing the control commands sent by the switch driving control circuit; the local MCU is also used for identifying a control instruction of a user according to the user audio data, generating a reply instruction according to the control instruction of the user and returning the reply instruction to the audio coding and decoding circuit, the audio coding and decoding circuit is also used for converting the reply instruction into a reply audio electric signal, and the audio amplifying circuit is also used for amplifying the reply audio electric signal; the loudspeaker is used for playing the amplified reply audio electric signal; the display screen is used for outputting characters or patterns according to the local MCU control command; the touch screen circuit is used for collecting touch screen control commands of a user and sending the touch screen control commands to the local MCU;
the control instruction for identifying the user according to the user audio data specifically comprises the following steps:
in the initial configuration stage, the local MCU is also used for the interaction of the network communication circuit and the online server, the online server recognizes the control instruction of the user according to the user audio data, the online server firstly recognizes the control instruction through a neural network model, and firstly carries out denoising, speed reduction and enhancement treatment on the user audio data so as to improve the accuracy of voice recognition; converting the processed audio data into feature vectors, wherein the feature vectors need to capture tone, tone color and tone intensity in the audio data; the construction of a Recurrent Neural Network (RNN), which is a special neural network capable of processing sequence data such as audio signals, has found widespread use in audio recognition. In audio recognition, the RNN may capture time dependencies in the audio signal, thereby enabling classification and recognition of audio. Specifically, RNN can capture time-dependent information in input data by simulating characteristics of human brain neurons. In audio recognition, the RNN may capture features in the audio signal and pass these features as inputs to the model for training and prediction. Through the trained RNN model, the input audio signals can be classified and identified, so that the audio identification task is realized. In addition, RNNs may also be applied to the generation and repair of audio signals. By modeling and predicting the audio signal using the RNN, an audio signal similar to the original audio signal may be generated, or noise and distortion in the audio signal may be repaired. Training the neural network model using the speech dataset to enable it to identify user audio data; gradient descent and regularization are needed in the training process to improve the generalization capability of the model; evaluating the performance of the model on the test set, adjusting and optimizing according to the evaluation result, adjusting the network structure and optimizing parameters, and applying the model to the recognition control instruction after model training is completed;
analyzing the audio electric signal characteristics corresponding to the control instruction of the user through the online server, performing time domain analysis on the audio electric signal corresponding to the control instruction of the user to obtain the audio electric signal corresponding to the control instruction of the user, processing the audio electric signal corresponding to the control instruction of the user into a time sequence, collecting the integral quantity, the variation quantity and the second derivative function quantity of the time sequence in a certain period, respectively taking the integral quantity, the variation quantity and the second derivative function quantity of the time sequence in the certain period as one component of the multi-dimensional characteristic vector to establish the multi-dimensional characteristic vector, and taking the multi-dimensional characteristic vector as the audio electric signal characteristics corresponding to the control instruction of the user;
establishing a mapping table between the audio electric signal characteristics of the user and the control instructions of the user, firstly identifying the control instructions by an online server through a neural network model (namely a cyclic neural network (RNN) mentioned above), then characterizing the identified control instructions by target storage variables, determining multi-dimensional characteristic vectors after the time domain analysis of the audio electric signal characteristics, and establishing the mapping table between the audio electric signal characteristics of the user and the control instructions of the user by pointing the multi-dimensional characteristic vectors to the target storage variables;
the local MCU is also used for storing a mapping table between the audio electric signal characteristics of the user and the control instructions of the user in the local storage circuit;
in a non-initial configuration stage, the local MCU is further used for identifying user audio electric signal characteristics according to user audio data and defining the user audio electric signal characteristics as audio electric signal characteristics to be identified, and determining a user control instruction corresponding to the audio electric signal characteristics to be identified according to a mapping table between the user audio electric signal characteristics and the user control instruction, namely searching the mapping table between the user audio electric signal characteristics and the user control instruction in the local storage circuit, and interaction with a server is not needed.
It can be seen that the present application only needs an initial configuration stage to identify user audio data according to an online server neural network model (and also has super-strong computing resources), namely, identify a control instruction of a user, and then establish a mapping table between audio electric signal characteristics of the user and the control instruction of the user, where the mapping table is established based on accurate identification, and the mapping table is stored in a local storage circuit; in a non-initial configuration stage, namely a use stage, a mapping table between audio electric signal characteristics of a user and control instructions of the user is searched in a local storage circuit, so that the control instructions of the user corresponding to the audio electric signal characteristics to be identified can be determined, interaction with a server is not needed, the problem that the existing problem is solved is that if more users exist, the users send out requests to a remote server at the same time, the delay degree of the request return result of the user is higher due to the limited computing resource capacity of the remote server, and the computing requirement on the remote server is higher (because interaction with the server is not needed in the conventional use of the application, only the configuration stage interaction is more, and even if network delay exists, the network delay can be accepted by the user); every time a user needs to interact with a server when performing voice recognition, no method is available without a network (because no interaction with the server is needed or the network is needed in the conventional use of the application), and recognition accuracy is ensured (because the mapping table is established based on the accurate recognition of the server).
Embodiments of the present application that require protection include:
the intelligent voice recognition electric switch for controlling the curtain comprises a local MCU, a local storage circuit and a network communication circuit which are electrically connected, wherein the local MCU is also electrically connected with an audio coding and decoding circuit, the audio coding and decoding circuit is electrically connected with an audio amplifying circuit, the audio amplifying circuit is electrically connected with a microphone and a loudspeaker, and the microphone is used for receiving voice instructions sent by a user and corresponding to the voice and converting the voice audio into instruction electric signals; the audio amplifying circuit is used for amplifying the instruction electric signal; the audio encoding and decoding circuit is used for converting the amplified instruction electric signals of the user into user audio data and sending the user audio data to the local MCU; the local MCU is used for identifying a control instruction of a user according to the user audio data;
the control instruction for identifying the user according to the user audio data specifically comprises the following steps:
in the initial configuration stage, the local MCU is also used for interacting the network communication circuit with the online server, the online server identifies the control instruction of the user according to the user audio data, and the mapping table between the audio electric signal characteristics of the user and the control instruction of the user is established by analyzing the audio electric signal characteristics corresponding to the control instruction of the user through the online server, and the local MCU is also used for storing the mapping table between the audio electric signal characteristics of the user and the control instruction of the user in the local storage circuit;
in a non-initial configuration stage, the local MCU is further used for identifying user audio electric signal characteristics according to user audio data and defining the user audio electric signal characteristics as audio electric signal characteristics to be identified, and determining control instructions of a user corresponding to the audio electric signal characteristics to be identified according to a mapping table between the user audio electric signal characteristics and the control instructions of the user.
Preferably, the local MCU is also electrically connected with a switch driving control circuit, and the switch driving control circuit is electrically connected with an electric push rod motor and an electric scroll motor; the local MCU is used for identifying a control command of a user according to the user audio data and sending a control command to the switch driving control circuit according to the control command of the user, and the electric push rod motor and the electric scroll motor are used for executing the control command sent by the switch driving control circuit.
Preferably, the local MCU is further used for identifying a control instruction of a user according to the audio data of the user, generating a reply instruction according to the control instruction of the user and returning the reply instruction to the audio codec circuit, the audio codec circuit is further used for converting the reply instruction into a reply audio electric signal, and the audio amplifying circuit is further used for amplifying the reply audio electric signal; the speaker is used for playing the amplified reply audio electric signal.
Preferably, the system further comprises a display screen, wherein the display screen is used for outputting characters or patterns according to the local MCU control command.
Preferably, the remote control unit further comprises a touch screen circuit, wherein the touch screen circuit is used for collecting touch screen control commands of a user and sending the touch screen control commands to the local MCU.
Preferably, the on-line server analyzes the audio electric signal characteristics corresponding to the control instruction of the user, specifically, performs time domain analysis on the audio electric signal corresponding to the control instruction of the user, establishes a multi-dimensional characteristic vector, and uses the multi-dimensional characteristic vector as the audio electric signal characteristics corresponding to the control instruction of the user.
Preferably, time domain analysis is performed on an audio electric signal corresponding to a control instruction of a user to establish a multi-dimensional feature vector, and specifically includes firstly acquiring the audio electric signal corresponding to the control instruction of the user, then processing the audio electric signal corresponding to the control instruction of the user into a time sequence, then collecting an integral quantity, a variation quantity and a second derivative function quantity of the time sequence in a certain period, and taking the integral quantity, the variation quantity and the second derivative function quantity of the time sequence in the certain period as one component of the multi-dimensional feature vector to establish the multi-dimensional feature vector.
Preferably, the step of establishing a mapping table between the audio electric signal characteristics of the user and the control instructions of the user specifically means that the on-line server firstly identifies the control instructions through a neural network model, then characterizes the identified control instructions by a target storage variable, determines multi-dimensional characteristic vectors after the audio electric signal characteristic time domain analysis, and establishes the mapping table between the audio electric signal characteristics of the user and the control instructions of the user by pointing the multi-dimensional characteristic vectors to the target storage variable.
Preferably, the on-line server firstly identifies the control instruction by a neural network model, and firstly carries out denoising, speed reduction and enhancement treatment on the user audio data so as to improve the accuracy of voice identification; converting the processed audio data into feature vectors, wherein the feature vectors need to capture tone, tone color and tone intensity in the audio data; constructing a cyclic neural network, and training a neural network model by using a voice data set so that the neural network model can identify user audio data; gradient descent and regularization are needed in the training process to improve the generalization capability of the model; and evaluating the performance of the model on the test set, adjusting and optimizing according to the evaluation result, adjusting the network structure and optimizing parameters, and applying the model to the recognition control instruction after model training is completed.
The online servers referred to herein may vary widely in configuration or performance and may include one or more central processing units (Central Processing Units, CPU) and memory (e.g., one or more processors), one or more storage media (e.g., one or more mass storage devices) storing applications or data. The memory and storage medium may be transitory or persistent. The program stored on the storage medium may include one or more modules, each of which may include a series of instruction operations on the server. Still further, the central processor may be configured to communicate with a storage medium and execute a series of instruction operations on the storage medium on a server.
The server may also include one or more power supplies, one or more wired or wireless network interfaces, one or more input/output interfaces, and/or one or more operating systems, such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, etc.
In addition, the embodiment of the application also provides a storage medium for storing a computer program for executing the above embodiment.
The present embodiments also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the above embodiments.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above embodiments may be implemented by hardware associated with program instructions, where the above program may be stored in a computer readable storage medium, and when executed, the program performs steps including the above method embodiments; and the aforementioned storage medium may be at least one of the following media: read-only Memory (ROM), RAM, magnetic disk or optical disk, etc.
The application also discloses an electronic device comprising a computer readable storage medium for storing a computer program for executing the above embodiments.
It should be noted that, in the present specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment is mainly described in a different point from other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, with reference to the description of the method embodiments in part. The apparatus and system embodiments described above are merely illustrative, in which elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The foregoing is merely one specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. The intelligent voice recognition electric switch for controlling the curtain comprises a local MCU, a local storage circuit and a network communication circuit which are electrically connected, wherein the local MCU is also electrically connected with an audio coding and decoding circuit, the audio coding and decoding circuit is electrically connected with an audio amplifying circuit, the audio amplifying circuit is electrically connected with a microphone and a loudspeaker, and the microphone is used for receiving voice instructions sent by a user and corresponding to the voice and converting the voice audio into instruction electric signals; the audio amplifying circuit is used for amplifying the instruction electric signal; the audio encoding and decoding circuit is used for converting the amplified instruction electric signals of the user into user audio data and sending the user audio data to the local MCU; the local MCU is used for identifying a control instruction of a user according to the user audio data;
the method is characterized in that the control instruction for identifying the user according to the user audio data specifically comprises the following steps:
in the initial configuration stage, the local MCU is also used for interacting the network communication circuit with the online server, the online server identifies the control instruction of the user according to the user audio data, and the mapping table between the audio electric signal characteristics of the user and the control instruction of the user is established by analyzing the audio electric signal characteristics corresponding to the control instruction of the user through the online server, and the local MCU is also used for storing the mapping table between the audio electric signal characteristics of the user and the control instruction of the user in the local storage circuit;
in a non-initial configuration stage, the local MCU is also used for identifying user audio electric signal characteristics according to user audio data and defining the user audio electric signal characteristics as audio electric signal characteristics to be identified, and determining control instructions of a user corresponding to the audio electric signal characteristics to be identified according to a mapping table between the user audio electric signal characteristics and the control instructions of the user; the local MCU is also electrically connected with a switch driving control circuit, and the switch driving control circuit is electrically connected with an electric push rod motor and an electric scroll motor; the local MCU is used for identifying a control command of a user according to the user audio data and sending a control command to the switch driving control circuit according to the control command of the user, and the electric push rod motor and the electric scroll motor are used for executing the control command sent by the switch driving control circuit.
2. The intelligent voice recognition electrical switch for curtain control of claim 1, wherein the local MCU is further configured to recognize a control command of a user based on the user audio data and generate a reply command based on the control command of the user and return the reply command to the audio codec circuit, the audio codec circuit is further configured to convert the reply command into a reply audio electrical signal, and the audio amplifier circuit is further configured to amplify the reply audio electrical signal; the speaker is used for playing the amplified reply audio electric signal.
3. The intelligent voice-recognition electric switch for controlling curtains according to claim 2, further comprising a display screen for outputting characters or patterns according to a control command of the local MCU.
4. The intelligent voice-recognition electrical switch for controlling a window covering of claim 3, further comprising a touch screen circuit for capturing touch screen control commands of a user and transmitting to the local MCU.
5. The intelligent voice recognition electric switch for controlling curtain of claim 4, wherein the on-line server analyzes the audio signal characteristics corresponding to the control command of the user, specifically, performs time domain analysis on the audio signal corresponding to the control command of the user, establishes a multi-dimensional feature vector, and uses the multi-dimensional feature vector as the audio signal characteristics corresponding to the control command of the user.
6. The intelligent voice recognition electric switch for curtain control according to claim 5, wherein the time domain analysis is performed on the audio electric signal corresponding to the control command of the user to establish a multi-dimensional feature vector, specifically comprising the steps of firstly obtaining the audio electric signal corresponding to the control command of the user, then processing the audio electric signal corresponding to the control command of the user into a time sequence, then collecting the integral quantity, the variation quantity and the second derivative function quantity of the time sequence in a certain period, and respectively taking the integral quantity, the variation quantity and the second derivative function quantity of the time sequence in a certain period as one component of the multi-dimensional feature vector to establish the multi-dimensional feature vector.
7. The intelligent voice-recognition electric switch for curtain control of claim 6, wherein the step of establishing a mapping table between the audio signal characteristics of the user and the control instructions of the user is to identify the control instructions by the on-line server through a neural network model, then characterize the identified control instructions with target storage variables, determine multi-dimensional feature vectors after time domain analysis of the audio signal characteristics, and establish a mapping table between the audio signal characteristics of the user and the control instructions of the user by pointing the multi-dimensional feature vectors to the target storage variables.
8. The intelligent voice recognition electric switch for curtain control according to claim 7, wherein the on-line server firstly recognizes the specific instruction of the control instruction through a neural network model, and firstly performs denoising, speed reduction and enhancement treatment on the user audio data so as to improve the accuracy of voice recognition; converting the processed audio data into feature vectors, wherein the feature vectors need to capture tone, tone color and tone intensity in the audio data; constructing a cyclic neural network, and training a neural network model by using a voice data set so that the neural network model can identify user audio data; gradient descent and regularization are needed in the training process to improve the generalization capability of the model; and evaluating the performance of the model on the test set, adjusting and optimizing according to the evaluation result, adjusting the network structure and optimizing parameters, and applying the model to the recognition control instruction after model training is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410157967.4A CN117690435A (en) | 2024-02-04 | 2024-02-04 | Intelligent voice recognition electric switch for curtain control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410157967.4A CN117690435A (en) | 2024-02-04 | 2024-02-04 | Intelligent voice recognition electric switch for curtain control |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117690435A true CN117690435A (en) | 2024-03-12 |
Family
ID=90139487
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410157967.4A Pending CN117690435A (en) | 2024-02-04 | 2024-02-04 | Intelligent voice recognition electric switch for curtain control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117690435A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105913847A (en) * | 2016-06-01 | 2016-08-31 | 北京灵隆科技有限公司 | Voice control system, user end device, server and central control unit |
| CN107320974A (en) * | 2017-08-17 | 2017-11-07 | 湖州金软电子科技有限公司 | A kind of infant's interactive voice toy |
| CN110506452A (en) * | 2017-02-07 | 2019-11-26 | 路创技术有限责任公司 | Load control system based on audio |
| CN110534096A (en) * | 2019-08-22 | 2019-12-03 | 北京技德系统技术有限公司 | A kind of artificial intelligent voice recognition methods and system based on microcontroller |
| CN111009237A (en) * | 2019-12-12 | 2020-04-14 | 北京达佳互联信息技术有限公司 | Voice recognition method and device, electronic equipment and storage medium |
| CN111081231A (en) * | 2016-03-23 | 2020-04-28 | 谷歌有限责任公司 | Adaptive audio enhancement for multi-channel speech recognition |
| CN111091818A (en) * | 2019-12-24 | 2020-05-01 | 广东美的白色家电技术创新中心有限公司 | Voice recognition circuit, voice interaction equipment and household appliance |
| CN111667826A (en) * | 2020-05-28 | 2020-09-15 | 深圳创维－Rgb电子有限公司 | Radio and television monitor with AI voice control function and AI voice control method |
| CN111670471A (en) * | 2018-01-04 | 2020-09-15 | 谷歌有限责任公司 | Learning offline voice commands based on use of online voice commands |
| CN211879013U (en) * | 2019-12-03 | 2020-11-06 | 佛山市顺德区美的电热电器制造有限公司 | Voice recognition device, remote control terminal, operation control device and electrical equipment |
| CN112423065A (en) * | 2020-11-11 | 2021-02-26 | 深圳市康意数码科技有限公司 | Voice recognition device for television |
| CN116259312A (en) * | 2021-12-21 | 2023-06-13 | 上海理工大学 | Method for automatically editing task by aiming at voice and neural network model training method |
-
2024
- 2024-02-04 CN CN202410157967.4A patent/CN117690435A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111081231A (en) * | 2016-03-23 | 2020-04-28 | 谷歌有限责任公司 | Adaptive audio enhancement for multi-channel speech recognition |
| CN105913847A (en) * | 2016-06-01 | 2016-08-31 | 北京灵隆科技有限公司 | Voice control system, user end device, server and central control unit |
| CN110506452A (en) * | 2017-02-07 | 2019-11-26 | 路创技术有限责任公司 | Load control system based on audio |
| CN107320974A (en) * | 2017-08-17 | 2017-11-07 | 湖州金软电子科技有限公司 | A kind of infant's interactive voice toy |
| CN111670471A (en) * | 2018-01-04 | 2020-09-15 | 谷歌有限责任公司 | Learning offline voice commands based on use of online voice commands |
| CN110534096A (en) * | 2019-08-22 | 2019-12-03 | 北京技德系统技术有限公司 | A kind of artificial intelligent voice recognition methods and system based on microcontroller |
| CN211879013U (en) * | 2019-12-03 | 2020-11-06 | 佛山市顺德区美的电热电器制造有限公司 | Voice recognition device, remote control terminal, operation control device and electrical equipment |
| CN111009237A (en) * | 2019-12-12 | 2020-04-14 | 北京达佳互联信息技术有限公司 | Voice recognition method and device, electronic equipment and storage medium |
| CN111091818A (en) * | 2019-12-24 | 2020-05-01 | 广东美的白色家电技术创新中心有限公司 | Voice recognition circuit, voice interaction equipment and household appliance |
| CN111667826A (en) * | 2020-05-28 | 2020-09-15 | 深圳创维－Rgb电子有限公司 | Radio and television monitor with AI voice control function and AI voice control method |
| CN112423065A (en) * | 2020-11-11 | 2021-02-26 | 深圳市康意数码科技有限公司 | Voice recognition device for television |
| CN116259312A (en) * | 2021-12-21 | 2023-06-13 | 上海理工大学 | Method for automatically editing task by aiming at voice and neural network model training method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110600017B (en) | Training method of voice processing model, voice recognition method, system and device | |
| US10649060B2 (en) | Sound source localization confidence estimation using machine learning | |
| US20220036903A1 (en) | Reverberation compensation for far-field speaker recognition | |
| CN112599122B (en) | Voice recognition method and device based on self-attention mechanism and memory network | |
| WO2020014899A1 (en) | Voice control method, central control device, and storage medium | |
| CN111722696B (en) | Voice data processing method and device for low-power-consumption equipment | |
| CN111415653B (en) | Method and device for recognizing speech | |
| US20240046955A1 (en) | Voice extraction method and apparatus, and electronic device | |
| CN111341333B (en) | Noise detection method, noise detection device, medium, and electronic apparatus | |
| KR102194194B1 (en) | Method, apparatus for blind signal seperating and electronic device | |
| CN110648669B (en) | Multi-frequency shunt voiceprint recognition method, device and system and computer readable storage medium | |
| US10910000B2 (en) | Method and device for audio recognition using a voting matrix | |
| CN117690435A (en) | Intelligent voice recognition electric switch for curtain control | |
| US20230186943A1 (en) | Voice activity detection method and apparatus, and storage medium | |
| CN112489678A (en) | Scene recognition method and device based on channel characteristics | |
| CN113593603A (en) | Audio category determination method and device, storage medium and electronic device | |
| Li et al. | Dynamic attention based generative adversarial network with phase post-processing for speech enhancement | |
| CN113113048A (en) | Speech emotion recognition method and device, computer equipment and medium | |
| CN117636909B (en) | Data processing method, device, equipment and computer readable storage medium | |
| CN116705013B (en) | Voice wake-up word detection method and device, storage medium and electronic equipment | |
| KR102334580B1 (en) | Apparatus and method for recognizing emotion based on user voice and graph neural network | |
| CN114882890A (en) | Deep learning model training method, voiceprint recognition method, device and equipment | |
| CN116229966A (en) | Method and system for controlling intelligent electrical appliance based on voice | |
| CN112102830A (en) | Coarse granularity instruction identification method and device | |
| CN113870879A (en) | Sharing method of microphone of intelligent household appliance, intelligent household appliance and readable storage medium |
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 |