CN114363105A - Intelligent equipment control method based on voice recognition - Google Patents
Intelligent equipment control method based on voice recognition Download PDFInfo
- Publication number
- CN114363105A CN114363105A CN202210049688.7A CN202210049688A CN114363105A CN 114363105 A CN114363105 A CN 114363105A CN 202210049688 A CN202210049688 A CN 202210049688A CN 114363105 A CN114363105 A CN 114363105A
- Authority
- CN
- China
- Prior art keywords
- sensor
- waveform
- person
- voice recognition
- detected
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001514 detection method Methods 0.000 claims description 18
- 238000009434 installation Methods 0.000 claims description 8
- 230000007613 environmental effect Effects 0.000 claims description 7
- 230000000087 stabilizing effect Effects 0.000 claims description 7
- 230000005587 bubbling Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000000977 initiatory effect Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 8
- 239000003990 capacitor Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000011084 recovery Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/2803—Home automation networks
- H04L12/2807—Exchanging configuration information on appliance services in a home automation network
- H04L12/2809—Exchanging configuration information on appliance services in a home automation network indicating that an appliance service is present in a home automation network
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/66—Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R1/00—Details of transducers, loudspeakers or microphones
- H04R1/08—Mouthpieces; Microphones; Attachments therefor
-
- 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)
- Signal Processing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Computer Networks & Wireless Communication (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
The application relates to an intelligent device control method based on voice recognition, which comprises the following steps: detecting whether a person exists in the area through a sensor of the corresponding area; if no person is detected, or the user is detected to set an interference-free state, or the current time is detected to be in an interference-free time period set by the user, controlling a voice module in the area to stop a voice recognition function; if a person is detected, and simultaneously, it is detected that the user does not set an interference-free state or the current time is not in the interference-free time period set by the user, enabling the local microphone to acquire a voice signal; and if the local microphone is in an enabling state, continuously acquiring microphone signals, and performing voice recognition to wake up and control the intelligent equipment in the area where the person is located. The method and the device can effectively avoid the problem that the intelligent devices in different rooms are awakened at the same time, and can realize that the short control words are used for controlling the intelligent devices in the corresponding areas.
Description
Technical Field
The application relates to the technical field of equipment control, in particular to an intelligent equipment control method based on voice recognition.
Background
With the popularization and popularity of smart homes, applications of controlling electronic devices in a voice recognition mode are increasing, such as controlling electric lamps, curtains, televisions, air conditioners and the like in rooms in a voice mode.
However, during the application process, the following problems still exist: for example, when a person is in a certain room, the device in the room needs to be controlled to be turned on, such as turning on an air conditioner or an electric lamp, but when the person turns on the air conditioner or turns on the electric lamp, the air conditioners or the electric lamps in other rooms can be turned on at the same time, and accurate voice control cannot be realized according to the requirements of users; in addition, the existing smart home controlled by a voice mode cannot avoid the situation that nobody is woken up by mistake.
Disclosure of Invention
In order to effectively avoid the problem that devices in different rooms are awakened at the same time, the application provides an intelligent device control method based on voice recognition.
The intelligent equipment control method based on voice recognition adopts the following technical scheme:
detecting whether a person exists in the area through a sensor of the corresponding area;
if no person is detected, or the user is detected to set an interference-free state, or the current time is detected to be in an interference-free time period set by the user, controlling a voice module in the area to stop a voice recognition function;
if a person is detected, and simultaneously, it is detected that the user does not set an interference-free state or the current time is not in the interference-free time period set by the user, enabling the local microphone to acquire a voice signal;
and if the local microphone is in an enabling state, continuously acquiring microphone signals, and performing voice recognition to wake up and control the intelligent equipment in the area where the person is located.
By adopting the above technology, when someone is detected, and simultaneously, when the user is detected not to be set with the disturbance-free state, the voice signal is detected, so that when the user is in a certain room or area, the voice signal can only be detected by the sensor in the room or area, and other rooms or areas can not be detected, so that only the intelligent device in the area can be awakened by voice, the problem that devices in different rooms or areas are awakened at the same time is effectively avoided, and meanwhile, the situation that the devices are awakened by mistake when no one is provided is also avoided. In addition, by adopting the technical scheme of the application, the off-line voice recognition training set and the quick and accurate recognition result can be stored in the local chip, the network is not required to be relied on, and even if the network does not exist, the voice control equipment can be executed and the intelligent home scene mode can be executed. Generally speaking, if online voice recognition is carried out at a cloud end for recognition and control, when a network is not good, recognition and control are usually slow, an offline recognition mode can be adopted in the method, the local linkage of the intelligent home gateway is matched, the device does not depend on any network bandwidth, for example, the lamp is turned on, the recognition result is sent to the intelligent home gateway by the device, the gateway searches a room area, the lamp of a corresponding room is automatically turned on, for example, a movie theatre mode, the intelligent home gateway automatically turns on a television and a set top box, a main lamp is turned off, an atmosphere lamp is turned on and the like, and quick response is achieved.
Preferably, whether a person is in a certain area is detected by the following method:
acquiring an output waveform of a sensor;
judging whether the walking waveform of the person exists according to the output waveform of the sensor, and further judging whether the person exists: if the voltage value of the first waveform exceeds the voltage setting threshold value, the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than twice of the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold value, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than half of the preset time, judging that the person is present;
alternatively, the first and second electrodes may be,
and if the voltage value of the first waveform exceeds the voltage setting threshold value, the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is longer than the preset time and is less than twice of the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold value, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is longer than the preset time, determining that the person is present.
Through adopting above technique, detect the duration of the voltage value condition of two wave forms and the wave form that the voltage value surpassed the voltage setting threshold value to the condition that can be more accurate detection have nobody, prevent the wrong report, the condition of failing to report takes place, for example, adopt the above-mentioned technique of this application, then intelligent equipment can not awaken up because of external interference noise, guaranteed the accurate control of equipment.
Preferably, after the presence of the person is detected, if it is detected that the voltage value of any one waveform reaches 2/3 of the voltage setting threshold value and the duration of the waveform of which the voltage value reaches 2/3 is greater than the preset time within the time delay threshold value from the presence of the person to the absence of the person, the presence of the person is judged, and a time delay threshold value counter from the presence of the person to the absence of the person is reset;
and when the waveform meeting the condition is not detected continuously within the time delay threshold from the person to the nobody, judging that the waveform is nobody, and restarting to detect the walking waveform of the person.
Through adopting above technique to can simple effectual realization to the slight action of human body (for example someone reads in the room, plays cell-phone etc.) carry out accuracy, effectual detection, specific saying so, if the people slightly lifts the hand, waves the hand in detection range, even only a brief wave form this moment, also can be detected the human and persist, thereby can further prevent that the phenomenon of sensor misreport, missing report from taking place.
Preferably, the voltage setting threshold is not more than 0.58V or not less than 1.75V; the preset time is 40-160 s. Therefore, the condition of the existence of the person can be more accurately detected, and the occurrence of the conditions of false alarm and missed alarm is reduced.
Preferably, the method further comprises:
acquiring an output waveform of a sensor;
when the sensor does not detect people, if the duration of a waveform exceeding a voltage set threshold in the detection waveform of the sensor is less than an effective human body trigger pulse width value, judging that an interference signal exists, and accumulating an interference count;
if the interference signals are continuously acquired for more than 7-15 times within 6 seconds, a prompt signal is sent to the intelligent gateway, and the counting of the local suspected fault counter of the equipment is increased by one;
and if the 'suspected fault counter' counts more than 3-10 times in 24 hours, reporting a fault signal to the intelligent gateway, and prompting a user to check the installation environment or replace the installation position.
By adopting the technology, the detection precision of the sensor can be further improved, the serious radiation interference possibly existing in the installation environment is eliminated, and the situations of false alarm and missed alarm are prevented.
Preferably, the output signal of the sensor is processed by the following method to obtain the output waveform of the sensor:
filtering the output signal of the sensor to filter out high-frequency interference signals;
amplifying the filtered signals by using an amplifying circuit in sequence, and then stabilizing the voltage by using a voltage stabilizing circuit;
and outputting the processed signal to obtain the output waveform of the sensor.
By adopting the technology, the signal of the sensor is filtered, processed and stabilized by hardware, so that more accurate and stable waveforms can be output, and the subsequent detection of the existence of people is more accurate.
Preferably, the sensor is a pyroelectric infrared sensor; the high-frequency interference signal is filtered by respectively connecting a 10-100p capacitor to the ground to a signal output pin and a power supply pin of the pyroelectric infrared sensor; and voltage stabilization processing is carried out on the amplified signal through a capacitor of 0.1-1 uF.
By adopting the technical means, the capacitor is adopted for filtering and stabilizing voltage, so that the finally obtained sensor detection waveform is more accurate for judging the existence condition of people.
Preferably, after acquiring the output waveform of the sensor, the output waveform is corrected by:
recording a drift curve of a hardware output signal obtained due to the change of the environmental temperature so as to obtain compensation values corresponding to different temperature differences;
and if the environmental temperature changes, correcting the obtained output waveform of the sensor by using the compensation value corresponding to the corresponding temperature difference.
By adopting the technology, when the temperature changes, the measured value of the hardware can be compensated by using the drift curve, so that the obtained output waveform is more accurate for judging whether the condition is existed or not, and the occurrence of the conditions of false alarm and missing alarm is reduced.
Preferably, the ambient temperature is calculated by:
acquiring adc values of the thermistors for multiple times;
sequencing the acquired adc values of the thermistors through a bubbling algorithm;
taking the 5 middle values to add and average to obtain an average value;
and obtaining the temperature corresponding to the average value by searching a temperature table.
By adopting the technology, particularly sequencing the acquired adc values of the thermistors through a bubbling algorithm; taking the 5 middle values to add and average to obtain an average value; and obtaining the temperature corresponding to the average value by searching a temperature table, so that the obtained environmental temperature data is more stable and cannot drift. According to the measurement of the environment temperature, the obtained detection waveform is more accurate after the obtained output waveform of the sensor is corrected by using the compensation value corresponding to the environment temperature difference, and the accuracy of the detection of the existence of people is further improved.
Preferably, the method further comprises:
when the sensor detects that the equipment state of the user changes, the equipment state of the sensor is sent to the intelligent gateway; when sending data, the sensor equipment judges the state of a link with the intelligent gateway, and if the link is disconnected, the sensor equipment inquires a connectable transit route again by adopting a dynamic period; the time intervals of the dynamic period become larger in sequence; if the sensor fails to be connected with the transit route all the time, the sensor is reconnected with any transit route or intelligent gateway all the time at the maximum interval period;
if the equipment state of the sensor is still not sent to the intelligent gateway after the timeout time of the heartbeat cycle reported by the sensor plus 1 minute exceeds 3 times, the intelligent gateway sets the corresponding sensor to be in an off-line state.
By adopting the technology, the problem of short-term power failure (a user may only temporarily close the relay router or temporarily cut off the power) or long-term power failure of the relay router can be solved, the purposes of saving power and reducing control data are achieved (after the relay is powered on again, the relay can be reconnected back to the network within the connection interval time), and finally stable connection between the sensor and the intelligent gateway is realized, so that the stability of voice control is ensured.
Preferably, when the intelligent device in the area where the person is located is awakened and controlled by using the voice signal, if the retransmission command of the intelligent gateway fails for more than 3 times, the application layer is used for initiating a new path request, and then the first 10 optimal routes are stored in the cache; and sending an instruction by using any one of the first 10 optimal routes, marking the route as NG if a certain route is not communicated, switching to the next optimal route for sending until all routes are tried, and discarding data.
By adopting the technology, when a certain relay routing software memory overflows or hardware has problems, other routes can be searched again to send data to the intelligent gateway central node, and stable connection between the intelligent gateway and the intelligent equipment is guaranteed.
In summary, the present application includes at least one of the following beneficial technical effects:
this application is detecting someone, detect the user simultaneously and when not setting up the state of exempting from to disturb, just detect speech signal, thereby when the user in a certain room or region, only can be detected by the sensor in this room or region, other rooms or region can not detect, thereby only this regional smart machine can be awakened up by pronunciation, effectively avoid the problem that different rooms or regional equipment are awakened up simultaneously, also avoided unmanned the condition emergence of awakening up by the mistake simultaneously. In addition, the implementation of the technology is very simple, the intelligent control system can be operated in a home local intelligent gateway, and is suitable for controlling home intelligent equipment simply and quickly by using the local intelligent gateway in a home.
This application can be simply effectual the realization to human slight action (for example someone reads in the room, plays cell-phone etc.) carry out accuracy, effectual detection, specific saying so, if the people slightly lifts the hand, waves the hand in detection range, even only a transient waveform this moment, also can be detected the people and persist to can further prevent that sensor misreport, the phenomenon of failing to report from taking place.
Drawings
FIG. 1 is a flow chart of a method of an embodiment of the present application.
Detailed Description
The present application is described in further detail below with reference to fig. 1.
The embodiment of the application discloses an intelligent device control method based on voice recognition. Referring to fig. 1, a method for controlling an intelligent device based on voice recognition includes the following steps:
s1, detecting whether a person exists in the area through the sensor of the corresponding area;
s2, if no person is detected or the user is detected to set the disturbance-free state or the current time is detected to be in the disturbance-free time period set by the user, controlling the voice module in the area to stop the voice recognition function;
s3, if a person is detected, and a user is detected not to set an interference-free state or the current time is not in the interference-free time period set by the user, enabling a local microphone to collect voice signals;
and S4, if the local microphone is in an enabled state, continuously acquiring microphone signals, and performing voice recognition to wake up and control the intelligent equipment in the area where the person is located.
After the scheme of the application is adopted, for example, after a certain person comes to the living room and says 'turn on light', because only the sensor of the living room detects someone (the sensor of other rooms can not detect someone), the intelligent gateway only controls the light of the living room to be turned on, and the lights in other rooms can not be turned on, so that the problem that a plurality of devices are simultaneously awakened is solved. In addition, in the application, the existence of the person is required to be judged, and the intelligent equipment is simultaneously controlled according to the state and the voice of the existence of the person, so that the condition that the equipment is mistakenly awakened by various interference noises (such as alarm sound, bird calling sound and outdoor human speaking sound) is avoided.
Furthermore, when a person in a plurality of rooms or areas is speaking at the same time and only one person wants to control the smart device in the room where the person is, in order to avoid the smart device being woken up by mistake, an adaptive room name wake-up word may be used to wake up the smart device, for example, the wake-up word may be: "you are good, kitchen", then when someone says "turn on the light", just the lamp in kitchen can be opened, even if other someone's room has gathered the sound of "turning on the light", can not control the lamp to open yet. And further, the accurate control of the intelligent equipment is realized. The room name in the awakening word corresponds to the area name when the intelligent device is divided into areas.
In this embodiment, whether a person exists in a certain area may be specifically detected by the following method:
s11, acquiring the output waveform of the sensor;
s12, judging whether the walking waveform of the person exists according to the output waveform of the sensor, and further judging whether the person exists: if the voltage value of the first waveform exceeds the voltage setting threshold value, the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than twice of the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold value, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than half of the preset time, judging that the person is present; or if the voltage value of the first waveform exceeds the voltage setting threshold, the duration of the waveform of which the voltage value exceeds the voltage setting threshold is greater than the preset time and less than twice the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold is greater than the preset time, determining that the person is present.
In other embodiments, the judgment of whether a person is present can also be directly made according to the comparison between the magnitude of the output signal of the sensor and the threshold value.
In order to accurately and effectively detect the slight movement of the human body (for example, someone reads in a room, plays a mobile phone, etc.), so as to further prevent the occurrence of the phenomena of false alarm and false alarm of the sensor, in this embodiment, the method further includes:
s13, when a person is detected, if the voltage value of any one waveform reaches 2/3 of the voltage setting threshold value and the duration time of the waveform of which the voltage value reaches 2/3 is longer than the preset time within the time delay threshold value from the person to the nobody, judging that the person is present, and resetting a counter of the time delay threshold value from the person to the nobody; the 'person-to-no-person delay threshold' can be set according to the use habits of users, for example, if the user is set to be no person for 30 minutes, the intelligent equipment is controlled to act, if the user is detected to be a person again when the time is counted down for 20 minutes, the time is counted down again for 30 minutes, and the intelligent equipment is controlled to act until no person is present for 30 minutes;
and S14, when the waveform meeting the condition is not detected continuously in the 'delay threshold from human to no human', judging that no human exists, and restarting to detect the walking waveform of the human.
In this embodiment, the voltage setting threshold is equal to or less than 0.58V or equal to or greater than 1.75V, specifically, for example, the downward threshold may be 0.58V, 0.48V, 0.4V, 0.35V, 0.28V, 0.22V, 0.12V, or 0.05V, and the upward threshold may be 1.75V, 1.85V, 1.95V, 2.05V, 2.15V, 2.25V, 2.35V, or 2.45V, and the specific threshold may be selected according to the actual sensitivity requirement.
The preset time may be 40-160s, and in a specific embodiment, the preset time may be set to 40ms, 80ms, 100ms, or 160ms, for example, and the specific preset time is selected according to an actual sensitivity requirement.
In order to further improve the detection accuracy of the sensor, eliminate significant radiation interference possibly existing in the installation environment, and prevent the occurrence of false alarm and false alarm, in this embodiment, the method further includes:
s101, acquiring an output waveform of a sensor;
s102, when the sensor does not detect a person, if the duration of a waveform exceeding a voltage set threshold in the detection waveforms of the sensor is less than an effective human body trigger pulse width value, judging that an interference signal exists, and accumulating an interference count;
s103, if interference signals are continuously acquired for more than 7-15 times (preferably 10 times) within 6 seconds, a prompt signal is sent to the intelligent gateway, and the number of the equipment counted by a local suspected fault counter is increased by one;
and S104, if the 'suspected fault counter' counts more than 3-10 times (preferably 5 times) in 24 hours, reporting a fault signal to the intelligent gateway, and prompting a user to check the installation environment or replace the installation position.
In order to further improve the detection accuracy of the presence or absence of people and prevent the occurrence of false alarm and false alarm, in the embodiment, the output signal of the sensor is processed by the following method to obtain the output waveform of the sensor:
filtering the output signal of the sensor to filter out high-frequency interference signals;
amplifying the filtered signals by using an amplifying circuit in sequence, and then stabilizing the voltage by using a voltage stabilizing circuit;
and outputting the processed signal to obtain the output waveform of the sensor.
The sensor can adopt a pyroelectric infrared sensor; the high-frequency interference signal can be filtered by respectively connecting a 10-100p capacitor to the ground to a signal output pin and a power supply pin of the pyroelectric infrared sensor; the voltage stabilization of the amplified signal can be realized by a capacitor of 0.1-1 uF.
In order to eliminate the influence of the change of the ambient temperature on the detection waveform of the sensor and further improve the detection accuracy of the sensor, in the embodiment, after the output waveform of the sensor is acquired, the output waveform is corrected by the following method:
recording a drift curve of a hardware output signal obtained due to environmental temperature change during an experiment so as to obtain compensation values corresponding to different temperature differences;
and if the environmental temperature changes, correcting the obtained output waveform of the sensor by using the compensation value corresponding to the corresponding temperature difference. For example, after actual test, the temperature is compensated by 0.08V upwards at-15 ℃, 0.06V upwards at 0 ℃, 0.03V upwards at 10 ℃ and the like.
In order to make the obtained ambient temperature data more stable and not drift, in this embodiment, the ambient temperature is calculated by the following method:
acquiring adc values of the thermistors for multiple times; at least 10 times;
sequencing the acquired adc values of the thermistors through a bubbling algorithm;
taking the 5 middle values to add and average to obtain an average value;
and obtaining the temperature corresponding to the average value by searching a temperature table.
In other embodiments, a conventional thermometer may also be employed to obtain the ambient temperature.
In specific implementation, when the output waveform of the sensor is obtained, the ADC can be configured to be in a scanning mode by using software, the original address of the DMA is configured to be the output address of the ADC, the destination address of the DMA is configured to be buffer, the acquisition is interrupted abnormally after the acquisition is completed, meanwhile, a hardware timer can be configured to be in a 15ms period, and the hardware automatically triggers the DMA to start the next round of ADC acquisition when the hardware overflows.
After the configuration of the thermistor and the acquisition input end of the sensor is finished, the MCU periodically enters an interrupt service function, copies the data acquired by the DMA to a new memory space at the moment, activates a task with the highest priority, executes the task after the interrupt exits, and starts to calculate the current temperature and the current sensor waveform.
In order to solve the problem of short-term power failure (a user may only temporarily turn off the relay router or short-term power failure) or long-term power failure of the relay router, and achieve the purposes of saving power and reducing control data (after the relay is powered on again, the relay router can be reconnected back to the network within the connection interval time), and finally realize stable connection between the sensor and the intelligent gateway to ensure the stability of voice control, in this embodiment, the method further includes:
when the sensor detects that the equipment state of the user changes, the equipment state of the sensor is sent to the intelligent gateway; when data is sent, the sensor equipment judges the link state of the intelligent gateway, if the link is disconnected, the sensor equipment inquires the connectable relay route again by adopting a dynamic period (for example, the equipment inquires the connectable relay route at intervals of 1 minute, 2 minutes, 3 minutes, 5 minutes, 15 minutes, 45 minutes, 90 minutes, 6 hours, 12 hours and 18 hours); the time intervals of the dynamic period become larger in sequence; if the sensor fails to be always connected with the transit route, the sensor is always reconnected with any transit route or intelligent gateway at the maximum interval period (such as 18 hours);
if the equipment state of the sensor is still not sent to the intelligent gateway after the timeout time of 3 times plus 1 minute exceeds the heartbeat cycle reported by the sensor (set according to the multiple of the heartbeat cycle of the equipment), the intelligent gateway sets the corresponding sensor to be in an offline state.
The intelligent gateway actively inquires and detects the connection state by detecting the heartbeat of the intelligent equipment and offline, so that the intelligent equipment is ensured to stably operate in the system. For example, the long power supply device reports data to the intelligent gateway at intervals of every 2min, if the intelligent gateway does not receive the reported data in a period exceeding 3 times, the intelligent gateway actively issues a command to read for 3 times, wherein the intervals are 1min, 2min and 3min respectively, and if the information of the device is not obtained, the corresponding device is set to be in an off-line state. If the intelligent gateway is connected with the network, the intelligent gateway is pushed to the user through the app, and if the intelligent gateway is not connected with the network, the intelligent gateway is pushed to the user through the local area network or the Bluetooth. In addition, when a sensor (such as a human body infrared sensor or a radar detector) is offline, the local voice recognition function is suspended to prevent disturbance caused by false awakening, specifically, through detection of heartbeat, data packets in a network can be reduced, the offline state of equipment can be detected, and the offline state is used for assisting the state enabling of local voice recognition.
In order to enable a relay routing software memory to overflow or hardware to have a problem, other routes can be searched again to send data to the intelligent gateway central node, in the embodiment, when the intelligent device in the area where a person is located is awakened and controlled by using a voice signal, if the intelligent gateway resending command fails for more than 3 times, an application layer is used to initiate a new path request, and then the first 10 optimal routes are stored in a cache; and sending an instruction by using any one of the first 10 optimal routes, marking the route as NG if a certain route is not communicated, switching to the next optimal route for sending until all routes are tried, and discarding data.
The route marked as NG can be maintained regularly, when the marked time exceeds 6 hours, the route is recovered to be used, and the link is adopted again when the route is planned again; for example, the more times of failures of a line marked NG, the longer the recovery time, for example, the first 6-hour recovery, the second 12-hour recovery, and the third 18-hour recovery.
The above embodiments are preferred embodiments of the present application, and the protection scope of the present application is not limited by the above embodiments, so: all equivalent variations made according to the methods and principles of the present application should be covered by the protection scope of the present application.
Claims (10)
1. An intelligent device control method based on voice recognition is characterized by comprising the following steps:
detecting whether a person exists in the area through a sensor of the corresponding area;
if no person is detected, or the user is detected to set an interference-free state, or the current time is detected to be in an interference-free time period set by the user, controlling a voice module in the area to stop a voice recognition function;
if a person is detected, and simultaneously, it is detected that the user does not set an interference-free state or the current time is not in the interference-free time period set by the user, enabling the local microphone to acquire a voice signal;
and if the local microphone is in an enabling state, continuously acquiring microphone signals, and performing voice recognition to wake up and control the intelligent equipment in the area where the person is located.
2. The intelligent device control method based on voice recognition according to claim 1, characterized in that whether a certain area is occupied is detected specifically by the following method:
acquiring an output waveform of a sensor;
judging whether the walking waveform of the person exists according to the output waveform of the sensor, and further judging whether the person exists: if the voltage value of the first waveform exceeds the voltage setting threshold value, the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than twice of the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold value, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is more than half of the preset time, judging that the person is present;
alternatively, the first and second electrodes may be,
and if the voltage value of the first waveform exceeds the voltage setting threshold value, the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is longer than the preset time and is less than twice of the preset time, and the voltage value of the second waveform exceeds the voltage setting threshold value, and the duration of the waveform of which the voltage value exceeds the voltage setting threshold value is longer than the preset time, determining that the person is present.
3. The intelligent device control method based on voice recognition according to claim 2, characterized in that: when a person is detected, if the voltage value of any waveform reaches 2/3 of the voltage setting threshold value and the duration time of the waveform of which the voltage value reaches 2/3 is longer than the preset time is detected within the person-to-no person delay threshold value, the person is judged to be present, and a person-to-no person delay threshold value counter is reset;
and when the waveform meeting the condition is not detected continuously within the time delay threshold from the person to the nobody, judging that the waveform is nobody, and restarting to detect the walking waveform of the person.
4. The intelligent device control method based on voice recognition according to claim 2 or 3, wherein: the voltage setting threshold is less than or equal to 0.58V or more than or equal to 1.75V; the preset time is 40-160 s.
5. The intelligent device control method based on voice recognition according to claim 3, wherein the method further comprises:
acquiring an output waveform of a sensor;
when the sensor does not detect people, if the duration of a waveform exceeding a voltage set threshold in the detection waveform of the sensor is less than an effective human body trigger pulse width value, judging that an interference signal exists, and accumulating an interference count;
if the interference signals are continuously acquired for more than 7-15 times within 6 seconds, a prompt signal is sent to the intelligent gateway, and meanwhile, the counting of a local suspected fault counter of the equipment is increased by one;
and if the 'suspected fault counter' counts more than 3-10 times in 24 hours, reporting a fault signal to the intelligent gateway, and prompting a user to check the installation environment or replace the installation position.
6. The intelligent device control method based on voice recognition according to claim 2 or 5, characterized in that the output waveform of the sensor is obtained by processing the output signal of the sensor by the following method:
filtering the output signal of the sensor to filter out high-frequency interference signals;
amplifying the filtered signals by using an amplifying circuit in sequence, and then stabilizing the voltage by using a voltage stabilizing circuit;
and outputting the processed signal to obtain the output waveform of the sensor.
7. The intelligent device control method based on voice recognition according to claim 6, wherein: after acquiring the output waveform of the sensor, correcting the output waveform by the following method:
recording a drift curve of a hardware output signal obtained due to the change of the environmental temperature so as to obtain compensation values corresponding to different temperature differences;
and if the environmental temperature changes, correcting the obtained output waveform of the sensor by using the compensation value corresponding to the corresponding temperature difference.
8. The intelligent device control method based on voice recognition according to claim 7, wherein the ambient temperature is calculated by:
acquiring adc values of the thermistors for multiple times;
sequencing the acquired adc values of the thermistors through a bubbling algorithm;
taking the 5 middle values to add and average to obtain an average value;
and obtaining the temperature corresponding to the average value by searching a temperature table.
9. The intelligent device control method based on voice recognition according to claim 1, wherein the method further comprises:
when the sensor detects that the equipment state of the user changes, the equipment state of the sensor is sent to the intelligent gateway; when sending data, the sensor equipment judges the state of a link with the intelligent gateway, and if the link is disconnected, the sensor equipment inquires a connectable transit route again by adopting a dynamic period; the time intervals of the dynamic period become larger in sequence; if the sensor fails to be connected with the transit route all the time, the sensor is reconnected with any transit route or intelligent gateway all the time at the maximum interval period;
and when the overtime time of 3 times plus 1 minute of the heartbeat cycle reported by the sensor still does not send the equipment state of the sensor to the intelligent gateway, the intelligent gateway sets the corresponding sensor to be in an off-line state.
10. The intelligent device control method based on voice recognition according to claim 1, characterized in that: when the intelligent equipment in the area where the person is located is awakened and controlled by using the voice signal, if the intelligent gateway resending command fails for more than 3 times, initiating a new path request by using an application layer, and then storing the first 10 optimal routes in a cache; and sending an instruction by using any one of the first 10 optimal routes, marking the route as NG if a certain route is not communicated, switching to the next optimal route for sending until all routes are tried, and discarding data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210049688.7A CN114363105A (en) | 2022-01-17 | 2022-01-17 | Intelligent equipment control method based on voice recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210049688.7A CN114363105A (en) | 2022-01-17 | 2022-01-17 | Intelligent equipment control method based on voice recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114363105A true CN114363105A (en) | 2022-04-15 |
Family
ID=81090751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210049688.7A Pending CN114363105A (en) | 2022-01-17 | 2022-01-17 | Intelligent equipment control method based on voice recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114363105A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933816A (en) * | 2014-03-17 | 2015-09-23 | 南充鑫源通讯技术有限公司 | Induction distance setting method and device for automatic induction security and protection system |
| CN107195304A (en) * | 2017-06-30 | 2017-09-22 | 珠海格力电器股份有限公司 | The voice control circuit and method of a kind of electric equipment |
| CN108320742A (en) * | 2018-01-31 | 2018-07-24 | 广东美的制冷设备有限公司 | Voice interactive method, smart machine and storage medium |
| CN109003604A (en) * | 2018-06-20 | 2018-12-14 | 恒玄科技（上海）有限公司 | A kind of audio recognition method that realizing low-power consumption standby and system |
| CN110677962A (en) * | 2019-09-20 | 2020-01-10 | 深圳市海洋王照明工程有限公司 | Illumination control method, illumination control device, computer equipment and storage medium |
| CN111503831A (en) * | 2020-04-29 | 2020-08-07 | 四川虹美智能科技有限公司 | Control method and intelligent air conditioner |
| CN111930017A (en) * | 2020-07-17 | 2020-11-13 | 润海智能科技扬州有限公司 | Method and equipment for controlling smart home |
| CN112071004A (en) * | 2019-06-28 | 2020-12-11 | 超级智慧家(上海)物联网科技有限公司 | Infrared human body detection method and infrared human body detector |
| CN112634894A (en) * | 2020-12-25 | 2021-04-09 | 苏州思必驰信息科技有限公司 | Awakening method, awakening device and terminal |
-
2022
- 2022-01-17 CN CN202210049688.7A patent/CN114363105A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104933816A (en) * | 2014-03-17 | 2015-09-23 | 南充鑫源通讯技术有限公司 | Induction distance setting method and device for automatic induction security and protection system |
| CN107195304A (en) * | 2017-06-30 | 2017-09-22 | 珠海格力电器股份有限公司 | The voice control circuit and method of a kind of electric equipment |
| CN108320742A (en) * | 2018-01-31 | 2018-07-24 | 广东美的制冷设备有限公司 | Voice interactive method, smart machine and storage medium |
| CN109003604A (en) * | 2018-06-20 | 2018-12-14 | 恒玄科技（上海）有限公司 | A kind of audio recognition method that realizing low-power consumption standby and system |
| CN112071004A (en) * | 2019-06-28 | 2020-12-11 | 超级智慧家(上海)物联网科技有限公司 | Infrared human body detection method and infrared human body detector |
| CN110677962A (en) * | 2019-09-20 | 2020-01-10 | 深圳市海洋王照明工程有限公司 | Illumination control method, illumination control device, computer equipment and storage medium |
| CN111503831A (en) * | 2020-04-29 | 2020-08-07 | 四川虹美智能科技有限公司 | Control method and intelligent air conditioner |
| CN111930017A (en) * | 2020-07-17 | 2020-11-13 | 润海智能科技扬州有限公司 | Method and equipment for controlling smart home |
| CN112634894A (en) * | 2020-12-25 | 2021-04-09 | 苏州思必驰信息科技有限公司 | Awakening method, awakening device and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN208398939U (en) | A kind of room air detection device | |
| US20170146962A1 (en) | Intelligent home system and control method thereof | |
| US20180293872A1 (en) | Method and Apparatus for Reminding Forgotten Mobile Terminal | |
| CN110500721B (en) | Air conditioner voice control method and device and air conditioner | |
| CN106989491B (en) | A kind of control system for air cleaning facility start and stop | |
| CN105302304A (en) | Method and apparatus for automatically starting illumination apparatus and intelligent watch | |
| CN105699936B (en) | Intelligent home indoor positioning method | |
| CN107548216B (en) | Illumination control method, device, equipment and storage medium | |
| CN114363105A (en) | Intelligent equipment control method based on voice recognition | |
| WO2016045525A1 (en) | Method and apparatus for processing state of terminal | |
| CN104270858B (en) | Method and device for intelligently controlling lighting system through voice | |
| WO2021219037A1 (en) | Oxygen-enriched air conditioner and control method for same | |
| CN109152167B (en) | Light control method and system | |
| KR101747494B1 (en) | Communication apparatus, communication method, and storage medium storing program | |
| CN113719980A (en) | Oxygen-enriched air conditioner and control method for oxygen-enriched air conditioner | |
| CN207993202U (en) | A kind of child room intelligent monitoring and controlling device | |
| CN106092313A (en) | The intelligent reminding system of SCM Based automatic identification noise | |
| KR102465304B1 (en) | Situation based ai smart home system using ai switch and ai living information device | |
| CN213780776U (en) | Environment detection control device | |
| CN108803369A (en) | Wireless intelligent house system | |
| CN114815635A (en) | Computer readable storage medium, intelligent panel and voice interaction method thereof | |
| CN205862788U (en) | The speech control system of household electrical appliance | |
| JP3563847B2 (en) | Disaster prevention monitoring control device and control method of disaster prevention monitoring control device | |
| CN211233269U (en) | Air conditioner capable of automatically adjusting sleep | |
| CN108494856B (en) | Space use condition management method and 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 |