CN111261142A - Intelligent switch system of nonspecific voice recognition based on singlechip - Google Patents
Intelligent switch system of nonspecific voice recognition based on singlechip Download PDFInfo
- Publication number
- CN111261142A CN111261142A CN202010055134.9A CN202010055134A CN111261142A CN 111261142 A CN111261142 A CN 111261142A CN 202010055134 A CN202010055134 A CN 202010055134A CN 111261142 A CN111261142 A CN 111261142A
- Authority
- CN
- China
- Prior art keywords
- voice recognition
- module
- password
- chip microcomputer
- voice
- 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
- 238000000605 extraction Methods 0.000 claims abstract description 12
- 230000009471 action Effects 0.000 claims abstract description 9
- 238000010183 spectrum analysis Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000013507 mapping Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 12
- 238000012549 training Methods 0.000 abstract description 4
- 238000013461 design Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000001939 inductive effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012567 pattern recognition method Methods 0.000 description 1
- 230000004044 response Effects 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
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- 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
- G10L15/08—Speech classification or search
-
- 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
- G10L15/08—Speech classification or search
- G10L15/14—Speech classification or search using statistical models, e.g. Hidden Markov Models [HMMs]
- G10L15/142—Hidden Markov Models [HMMs]
-
- 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
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/183—Speech classification or search using natural language modelling using context dependencies, e.g. language models
-
- 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
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- 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
- G10L15/08—Speech classification or search
- G10L2015/086—Recognition of spelled words
-
- 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
- G10L15/08—Speech classification or search
- G10L2015/088—Word spotting
-
- 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)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Probability & Statistics with Applications (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Selective Calling Equipment (AREA)
Abstract
The utility model provides a non-specific voice recognition intelligent switch system based on a single chip microcomputer, which comprises a voice recognition module, a main control module and a relay switch module; the voice recognition module is used for carrying out spectrum analysis and voice feature extraction according to the received sound, and matching the extracted voice feature with the keywords in the keyword list to obtain a voice recognition result; the main control module sends a control instruction to the relay switch module according to the received voice recognition result, and the relay switch module performs switching action according to the received instruction; the method and the device realize accurate identification of the household appliance control command, and avoid complex manual training; through the mutual cooperation of the two instructions, the influence of other sound sources is reduced, the recognition rate is improved, and the control precision of the household appliance is greatly improved.
Description
Technical Field
The disclosure relates to the technical field of voice recognition, in particular to a non-specific voice recognition intelligent switching system based on a single chip microcomputer.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
The intelligent household is characterized in that a house is taken as a platform, a system, a structure and control are integrated, various devices related to household life are organically combined by utilizing an advanced electric power automation technology, a computer technology and a radio sensing technology, and household devices are comprehensively managed in a networked mode to create a living environment space with high quality, convenience, high efficiency, energy conservation, health and environmental protection.
The inventor of the present disclosure finds that the traditional household appliance control modes, including the switch key and the infrared remote control, mainly have the following problems: the potential safety hazard of strong current leakage exists because people need to directly touch the screen; the control is inconvenient; unsafe, sometimes unsafe for private wiring of the switch; the control places are few, and at most, only two places of control can be realized; the control distance is short, and the wall cannot be penetrated; when the remote controllers have too many models, signals are easy to be mixed up.
Disclosure of Invention
In order to overcome the defects of the prior art, the non-specific voice recognition intelligent switching system based on the single chip microcomputer is provided, so that the accurate recognition of the household appliance control command is realized, and the complicated manual training is avoided; through the mutual cooperation of the two instructions, the influence of other sound sources is reduced, the recognition rate is improved, and the control precision of the household appliance is improved.
In order to achieve the purpose, the following technical scheme is adopted in the disclosure:
a non-specific voice recognition intelligent switch system based on a single chip microcomputer comprises a voice recognition module, a main control module and a relay switch module;
the voice recognition module is used for carrying out spectrum analysis and voice feature extraction according to the received sound, and matching the extracted voice feature with the keywords in the keyword list to obtain a voice recognition result;
the main control module sends a control instruction to the relay switch module according to the received voice recognition result, and the relay switch module performs switching action according to the received instruction.
As some possible implementation manners, the voice recognition module matches the extracted voice features with keywords in the keyword list, and finds out the keyword with the highest score to be output as a recognition result.
As some possible implementation modes, the voice recognition module adopts a multi-password voice recognition control mode, at least comprises a first password and a second password, and when the first password is correct, the recognition of the second password is carried out.
As a further limitation, the switch system further comprises a plurality of indicator lights at least comprising a password indicator light and a relay switch action indicator light, when the main control module identifies the first password and the first password is correct, the password indicator light is controlled to be on, when the main control module identifies the second password and the second password is correct, the password indicator light is controlled to be gradually turned off, at the moment, the relay performs the switch action, and the relay switch action indicator light is turned on.
As a further limitation, a list of key terms to be recognized is preset in the main control module, and the key terms are dynamically transmitted to the speech recognition module in the form of characters for recognizing the key terms.
As a further limitation, the keyword list is provided with a plurality of candidate recognition sentences, each recognition sentence is a single word, a phrase or a short sentence, and the short sentence is a pinyin string with the length not exceeding the preset number of words or the preset number of letters.
As some possible implementations, the speech recognition module includes a signal processing and feature extraction module, a pronunciation dictionary, a statistical acoustic model, a language model, and a decoder for finding a word string capable of outputting the signal with the highest probability according to the statistical acoustic model, the language model, and the pronunciation dictionary based on the extracted input signal features.
As a further limitation, the statistical acoustic model utilizes acoustic corpora and adopts a first-order hidden markov model for modeling.
By way of further limitation, the pronunciation dictionary contains a collection of words that the system can process and their pronunciations, and provides a mapping between the acoustic modeling unit and the language model modeling unit.
By way of further limitation, the language model is used to model the language for which the system is intended, including canonical languages, employing statistical-based N-grams and variants thereof.
Compared with the prior art, the beneficial effect of this disclosure is:
1. the method realizes efficient and accurate identification of the control command of the household appliance, can be easily used without manual training, realizes communication and control between voice and the household appliance, has the characteristics of strong anti-interference capability, multiple relations, safety, convenience (for the old and the disabled), flexibility and the like, and has wide application prospect.
2. The method adopts a mode that a non-specific ASR voice recognition and keyword extraction technology is combined with an STC12LE5A60S2 type single chip microcomputer, an ASR technology voice recognition chip performs spectrum analysis and voice feature extraction on sound input by MIC, the voice feature extraction is matched with keywords in a keyword list, and the single chip microcomputer processes commands to corresponding operation instructions of household appliances, so that control operation of a household electrical switch system is realized, and the recognition accuracy is increased.
3. The speech recognition module presets a key word list to be recognized and dynamically transmits the key words to the inside of the chip in a character form, so that the key words spoken by a user can be recognized, the unspecific speech recognition module for the user to train and record in advance is not needed, a system can support multiple scenes, and the pinyin combination of dialects can be added properly according to some accents in the local, so that the local dialects can be recognized, individuation is increased, and the practicability and the popularization are high.
4. According to the method, the household appliance is controlled by adopting the two-stage password, the influence of other sound sources is greatly reduced, the recognition rate is improved, the recognition accuracy is up to 95%, the keywords of the recognition list can be dynamically edited, and the method has obvious advantages compared with other voice recognition systems, and is convenient for wide popularization and application.
5. The method greatly improves the accuracy of voice recognition and effectively avoids error recognition by the design of a specific voice recognition system and the cooperation of a signal processing and feature extraction module, a pronunciation dictionary, a statistical acoustic model, a language model and a decoder and the combination of an editable recognition list, and can recognize various dialects by a pinyin editing mode so as to further improve the recognition accuracy.
Drawings
Fig. 1 is a schematic diagram of an overall control flow of a non-specific voice recognition intelligent switch system based on a single chip microcomputer according to embodiment 1 of the present disclosure.
Fig. 2 is a schematic overall flow chart of the load circuit provided in embodiment 1 of the present disclosure.
Fig. 3 is a schematic view of a work flow of a speech recognition module provided in embodiment 1 of the present disclosure.
Fig. 4 is a schematic diagram of a statistical pattern recognition process provided in embodiment 1 of the present disclosure.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The embodiments and features of the embodiments in the present disclosure may be combined with each other without conflict.
Example 1:
the embodiment 1 of the present disclosure provides a non-specific voice recognition intelligent switch system based on a single chip, which utilizes an LD3320 voice chip to design interface circuits such as a microphone bias, a single chip controller, an analog power supply assist, and the like, writes a driving program, and designs a "voice recognition module" to realize recognition of a "household appliance control command", and the voice recognition system is high in recognition rate, does not need manual training, and can be easily used. Realizes the communication and control of voice and household appliances, and is safe, convenient and flexible. The anti-interference device has the characteristics of strong anti-interference capability, multiple languages, convenience (for the old and the disabled), safety and the like, and has wide application prospect.
(1) Control strategy design
The specific control mode is as follows:
(1-1) combining the technology of nonspecific ASR speech recognition and keyword extraction with STC12LE5A60S2 type single-chip microcomputer
The main technical key of this embodiment is the collection, processing and judgment of the voice signal and the voice playing, and the voice recognition technology provided by the voice chip LD3320 is a recognition technology based on the keyword list. The work finished by the ASR technology speech recognition chip is as follows: the voice input by MIC is subjected to spectrum analysis, voice features are extracted and matched with the keywords in the keyword list, and the single chip microcomputer processes commands into corresponding 'operation instructions' of the household appliances, so that the control operation of the household electrical switch system is realized, and the identification accuracy is increased.
(1-2) double password voice recognition control mode (password mode + IO control mode) is adopted
Before using the speech recognition instruction, it must be spoken: the first-level password 'white swan' can be identified by the second-level password: turning on light and turning off light; turning on the fan and turning off the fan; playing music and stopping playing; opening the curtain and closing the curtain; all open and all closed. Therefore, the on-off of the LED lamp on the control circuit board, the on-off of the fan, the playing and stopping of music and the opening and closing of the curtain are controlled, and the on-off control of the intelligent household electrical appliance is simulated. (all passwords can be custom set).
(1-3) the multi-path relay load circuit generates a control signal to control the closing of the relay
As shown in fig. 2, when the operator sends a voice command, the system receives the voice command, processes and queries the voice command internally, and generates a response control signal, so as to control the on/off of the relay, thereby realizing the control of the on/off of the household appliance. When realizing the function for receiving voice command "white swan", 3 red green blue LED pilot lamps of development board are lighted, then receive second grade voice command "turn on the light" or "turn off the light", 3 red green blue LED pilot lamps extinguish gradually, and the load LED electric light of relay control is "lighted" or "is closed" this moment, goes to output through multichannel relay and can go to control the household electrical appliances in the family.
The Voice Recognition LD3320 Voice Recognition chip is a non-specific human Voice Recognition module which can recognize the key words spoken by the user only by setting a key word list to be recognized in a program of the main control MCU and dynamically transmitting the key words to the chip in a character form without the need of the user to train and record in advance. The module can set 50 candidate recognition sentences, each recognition sentence can be a single character, a phrase or a short sentence, the short sentence is a pinyin string with the length not more than 10 Chinese characters or 79 letters, a system can support various scenes, and the pinyin combination of dialects can be properly added according to some local accents, so that the local dialects can be recognized, and individuation is increased.
(2) Hardware design
In this embodiment, an STC12LE5a60S 21T enhanced single chip microcomputer is used, voice recognition is performed using an LD3320 voice recognition module, the system can communicate with the single chip microcomputer in parallel through the voice module, a hardware circuit of the whole system mainly includes a core main control part and a voice recognition part, the main control part is an STC12LE5a60S2 single chip microcomputer, the voice recognition part is an LD3320 voice recognition chip system, after a voice of a user enters the voice recognition part, the LD3320 transmits processed data to the main control unit in parallel, and after the voice is processed by the main control unit, command data is sent to an extensible peripheral serial device to realize control operation.
(2-1) design of a non-person-specific voice recognition module, as shown in FIG. 3.
(2-1-1) the speech recognition is divided into three steps: spectrum analysis, feature extraction and matching identification. The LD3320 integrates hardware of each part on a single chip, and mainly carries out secondary development on the LD 3320;
(2-1-2) the LD3320 chip is directly connected with the STC12 type single chip microcomputer in a parallel mode, 10K resistors are adopted for pulling up, and A0 is used for judging whether the data section or the address section control signal RDBWRBCSBRSTB (reset signal) and the interrupt return signal INTB are directly connected with the STC12 type single chip microcomputer;
(2-1-3) design step: determining a register read-write operation mode (parallel or serial SPI), familiarizing with the functions of the register, determining the address of a required register, and writing a driver (initialization → writing an identification list → starting identification → responding to interruption);
(2-1-4) in order to assist the system to work stably, a 10k resistor pull-up LD3320 chip and a single chip microcomputer are used for adopting the same external 24MHz clock;
(2-1-5) MBS is used as microphone bias and is additionally connected with an RC circuit to ensure that a floating voltage can be output to the microphone.
(2-2) STC12LE5A60S2 type single-chip microcomputer controller
The system described in this embodiment selects the main controller STC12LE5a60S2 single chip microcomputer with a working voltage of 3.3V, has 8K FLASH, 512 byte SRAM, contains asynchronous serial interfaces, 40I/O interfaces, and thus is sufficient to meet the requirements of the main control system for resources.
And (3) performing spectrum analysis on the sound input by the MIC- > extracting the voice characteristics- > and performing comparison matching on the key words in the key word list- > finding out the key words with the highest score to be output as a recognition result.
Determining a register read-write operation mode (parallel or serial SPI), familiarizing with the functions of the register, determining the address of a required register, and writing a driver (initialization → writing an identification list → starting identification → responding to interruption); and finally, the main singlechip completes the control of the intelligent switch.
(2-3) household switch controlled by multi-path load circuit
L9110 is a two-channel push-pull power amplification special integrated circuit device designed for controlling and driving a motor, and a discrete circuit is integrated in a single chip IC, so that the cost of peripheral devices is reduced, and the reliability of the whole machine is improved. The chip has two inputs of TTL/CMOS compatible level, and has good anti-interference performance; the two output ends can directly drive the motor to move forward and backward, the motor has larger current driving capability, each channel can pass 800mA continuous current, and the peak current capability can reach 1.5A; meanwhile, the output saturation voltage drop is low; the built-in clamping diode can release reverse impact current of inductive load, so that the use of the inductive load in driving a relay, a direct current motor, a stepping motor or a switching power tube is safe and reliable.
In order to realize control over household appliances, the LED lamp and the relay are used to simulate a real object lamp or household appliances in this embodiment, and the switches of the lamp and the household appliances are controlled through voice recognition, so as to reduce the influence of other sound sources and improve the recognition rate, the system adopts two-stage instructions to complete control over the system.
When keywords are added to the LD module, a two-dimensional array is defined by an editing program to set a first-level voice instruction as 'white swan' and a second-level voice instruction as 'light on' and 'light off', when the instructions are set, an input mode of pinyin is added as the keyword array, for example, the 'white swan' command is added, the 'bai tie' is written, and the pinyin among the Chinese characters is separated by spaces. The implementation function is that when receiving voice command "white swan", 3 red green blue LED pilot lamps of development board are lighted, then receive second grade voice command "turn on the light" or "turn off the light", 3 red green blue LED pilot lamps are extinguished gradually, and the load LED electric light of relay control is "lighted" or "is closed" this moment. The output can be controlled through a 4-way relay to control household appliances, such as a fan, a television, strong electric control of sound equipment and the like, so that: the system comprises a fan, a music player, a computer and the like, wherein the fan is turned on, the fan is turned off, or music is turned on, the music is turned off, and the like, and a series of secondary.
(2-4) typical Speech recognition System based on statistical Pattern recognition method
(2-4-1) Signal processing and feature extraction Module
The main task of this module is to extract features from the input signal for acoustic model processing. Meanwhile, it generally includes some signal processing techniques to reduce the influence of environmental noise, channels, speakers, etc. on the characteristics as much as possible.
(2-4-2) statistical Acoustic model
The typical system adopts a first-order hidden Markov model for modeling.
(2-4-3) pronunciation dictionary
The pronunciation dictionary contains the vocabulary and its pronunciations that the system can handle. The pronunciation dictionary actually provides a mapping between the acoustic modeling unit and the language model modeling unit.
(2-4-4) language model
The speech model is modeled aiming at the language which needs to be recognized by the system, and theoretically, various language models including a regular language and a context-free grammar can be used as the language models. But currently, statistical-based N-grams and variations thereof are commonly used by various systems.
(2-4-5) decoder
The decoder is one of the cores of a speech recognition system, and its task is to find a word string that can output an input signal with the maximum probability based on acoustics, language models, and dictionaries.
The embodiment realizes the remote control of the household appliances by unspecified human voice, the significance of the remote control is not reflected on intelligent home, and meanwhile, the remote control system can also be applied to various fields such as public lighting systems, sanitation systems, building conversations, navigators and the like. According to the intelligent identification electric appliance control system based on the LD3320, the household voice intelligent switching system is used for non-specific voice identification, the structure is simple, the identification accuracy is up to 95%, the cost performance is high, the keywords of the identification list can be dynamically edited, and the intelligent identification electric appliance control system has obvious advantages compared with other voice identification systems and is convenient to widely popularize and apply.
A specific voice recognition instruction is spoken through an MIC microphone, so that the on and off of household LED lamplight illumination are controlled, and the on and off of a fan and the like are controlled to simulate the control of the intelligent household.
The hardware module of the embodiment is reasonable in type selection, perfect in program function design, high in system recognition rate, good in stability and convenient to integrate, ideally realizes the function of controlling the switch of the household appliance system through voice commands by unspecified persons, and has wide market prospects in the fields of industrial automatic control, medical systems, smart home and the like.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Although the present disclosure has been described with reference to specific embodiments, it should be understood that the scope of the present disclosure is not limited thereto, and those skilled in the art will appreciate that various modifications and changes can be made without departing from the spirit and scope of the present disclosure.
Claims (10)
1. A non-specific voice recognition intelligent switch system based on a single chip microcomputer is characterized by comprising a voice recognition module, a main control module and a relay switch module;
the voice recognition module is used for carrying out spectrum analysis and voice feature extraction according to the received sound, and matching the extracted voice feature with the keywords in the keyword list to obtain a voice recognition result;
the main control module sends a control instruction to the relay switch module according to the received voice recognition result, and the relay switch module performs switching action according to the received instruction.
2. The intelligent switch system for nonspecific speech recognition based on a single-chip microcomputer according to claim 1, wherein the speech recognition module matches the extracted speech features with keywords in the keyword list to find out the keyword with the highest score as the recognition result and output the keyword as the recognition result.
3. The intelligent switch system for nonspecific voice recognition based on a single-chip microcomputer as claimed in claim 1, wherein the voice recognition module adopts a multiple password voice recognition control mode, including at least a first password and a second password, and when the first password is correct, the second password is recognized.
4. The intelligent switch system for nonspecific voice recognition based on a single-chip microcomputer according to claim 3, wherein the switch system further comprises a plurality of indicator lights at least comprising a password indicator light and a relay switch action indicator light, when the main control module recognizes the first password and the first password is correct, the password indicator light is controlled to be on, when the main control module recognizes the second password and the second password is correct, the password indicator light is controlled to be gradually turned off, and at the moment, the relay performs a switch action and the relay switch action indicator light is on.
5. The intelligent switch system for nonspecific voice recognition based on a single-chip microcomputer according to claim 3, wherein a list of key words to be recognized is preset in the main control module, and the key words are dynamically transmitted to the voice recognition module in the form of characters for recognition of the key words.
6. The intelligent switch system for nonspecific voice recognition based on a single-chip microcomputer according to claim 5, wherein the keyword list is provided with a plurality of candidate recognition sentences, each recognition sentence is a single word, a phrase or a short sentence, and the short sentence is a pinyin string with a length not exceeding a preset number of words or a preset number of letters.
7. The intelligent switch system for unspecified voice recognition based on one-chip microcomputer according to claim 1, wherein said voice recognition module comprises a signal processing and feature extraction module, a pronunciation dictionary, a statistical acoustic model, a language model and a decoder for finding a word string capable of outputting the signal with the maximum probability based on the statistical acoustic model, the language model and the pronunciation dictionary based on the extracted input signal features.
8. The intelligent switch system for unspecified voice recognition based on single chip microcomputer according to claim 7, wherein said statistical acoustic model utilizes acoustic corpus and is modeled by using a first-order hidden markov model.
9. The single-chip microcomputer based intelligent switch system for unspecified speech recognition according to claim 7, wherein said pronunciation dictionary comprises vocabulary sets and pronunciations thereof that can be processed by the system, said pronunciation dictionary providing mapping between the acoustic modeling unit and the language modeling unit.
10. The single-chip microcomputer-based unspecific voice recognition intelligent switching system according to claim 7, wherein said language model is used for modeling a language for which the system is directed, including a regular language, using a statistical-based N-gram and its variants.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010055134.9A CN111261142A (en) | 2020-01-17 | 2020-01-17 | Intelligent switch system of nonspecific voice recognition based on singlechip |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010055134.9A CN111261142A (en) | 2020-01-17 | 2020-01-17 | Intelligent switch system of nonspecific voice recognition based on singlechip |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111261142A true CN111261142A (en) | 2020-06-09 |
Family
ID=70950813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010055134.9A Pending CN111261142A (en) | 2020-01-17 | 2020-01-17 | Intelligent switch system of nonspecific voice recognition based on singlechip |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111261142A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112687269A (en) * | 2020-12-18 | 2021-04-20 | 山东盛帆蓝海电气有限公司 | Building management robot voice automatic identification method and system |
| CN113160810A (en) * | 2021-01-13 | 2021-07-23 | 安徽师范大学 | LD 3320-based voice recognition interaction method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102889751A (en) * | 2012-10-31 | 2013-01-23 | 昆山黄昆纺织机械有限公司 | Speech recognition refrigerator |
| CN104424942A (en) * | 2013-09-02 | 2015-03-18 | 张凯 | Method for improving character speed input accuracy |
| CN106713111A (en) * | 2015-11-17 | 2017-05-24 | 腾讯科技（深圳）有限公司 | Processing method for adding friends, terminal and server |
| CN107545891A (en) * | 2016-06-28 | 2018-01-05 | 广州零号软件科技有限公司 | A kind of pet cry recognition methods applied on service robot and system |
-
2020
- 2020-01-17 CN CN202010055134.9A patent/CN111261142A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102889751A (en) * | 2012-10-31 | 2013-01-23 | 昆山黄昆纺织机械有限公司 | Speech recognition refrigerator |
| CN104424942A (en) * | 2013-09-02 | 2015-03-18 | 张凯 | Method for improving character speed input accuracy |
| CN106713111A (en) * | 2015-11-17 | 2017-05-24 | 腾讯科技（深圳）有限公司 | Processing method for adding friends, terminal and server |
| CN107545891A (en) * | 2016-06-28 | 2018-01-05 | 广州零号软件科技有限公司 | A kind of pet cry recognition methods applied on service robot and system |
Non-Patent Citations (5)
| Title |
|---|
| 彭为主编: "《创客行》", 30 April 2017, pages: 67 - 69 * |
| 王中心等: "基于LD3320语音识别模块的智能家居声控***", 《物联网技术》, no. 11, 20 November 2016 (2016-11-20), pages 19 - 21 * |
| 许文卓等: "基于LD3320语音识别的智能家居控制***", 《科学技术创新》 * |
| 许文卓等: "基于LD3320语音识别的智能家居控制***", 《科学技术创新》, no. 27, 12 October 2018 (2018-10-12), pages 90 - 91 * |
| 陈超: "《导盲机器人定位与路径规划技术》", 国防工业出版社, pages: 25 - 29 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112687269A (en) * | 2020-12-18 | 2021-04-20 | 山东盛帆蓝海电气有限公司 | Building management robot voice automatic identification method and system |
| CN113160810A (en) * | 2021-01-13 | 2021-07-23 | 安徽师范大学 | LD 3320-based voice recognition interaction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3480817B1 (en) | Voice interactive device and method for controlling voice interactive device | |
| CN111261142A (en) | Intelligent switch system of nonspecific voice recognition based on singlechip | |
| US7502731B2 (en) | System and method for performing speech recognition by utilizing a multi-language dictionary | |
| CN2429801Y (en) | Intelleclural refrigerator with speech control | |
| CN110265012A (en) | It can interactive intelligence voice home control device and control method based on open source hardware | |
| CA2308946A1 (en) | Method and apparatus for controlling voice controlled devices | |
| CN101383150B (en) | Control method of speech soft switch and its application in geographic information system | |
| CN102760432A (en) | Voice-activated remote controller for household appliance and control method thereof | |
| CN103151038A (en) | Method of achieving voice recognition control in electronic products | |
| CN104239007B (en) | The method of Chinese natural human speech voice remote control electric light | |
| CN100578613C (en) | System and method for speech recognition utilizing a merged dictionary | |
| KR20130105219A (en) | Electronic device and method for controlling power using voice recognition thereof | |
| JP7122667B2 (en) | Lighting control system and device | |
| CN215007496U (en) | Off-line voice air conditioner remote controller | |
| CN201408365Y (en) | Intelligent house sound control device | |
| US7353173B2 (en) | System and method for Mandarin Chinese speech recognition using an optimized phone set | |
| Fezari et al. | New speech processor and ultrasonic sensors based embedded system to improve the control of a motorised wheelchair | |
| US20040193417A1 (en) | System and method for effectively implementing a mandarin chinese speech recognition dictionary | |
| WO2020175293A1 (en) | Apparatus control system, apparatus control method, and program | |
| CN206908920U (en) | A kind of light fixture of voice interface control | |
| CN109587900A (en) | A kind of control method of headlamp, system, device and medium | |
| CN210466389U (en) | Intelligent Chinese-English speech translation machine | |
| CN204315226U (en) | Based on the intelligent LED car lamp control system of signer-independent sign language recognition | |
| CN117784632B (en) | Intelligent household control system based on offline voice recognition | |
| CN106455244A (en) | Light control system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200609 |