CN108536413A - A kind of audio data collecting method and apparatus - Google Patents
A kind of audio data collecting method and apparatus Download PDFInfo
- Publication number
- CN108536413A CN108536413A CN201710127263.2A CN201710127263A CN108536413A CN 108536413 A CN108536413 A CN 108536413A CN 201710127263 A CN201710127263 A CN 201710127263A CN 108536413 A CN108536413 A CN 108536413A
- Authority
- CN
- China
- Prior art keywords
- trigger
- processor
- threshold
- triggered
- data
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/16—Sound input; Sound output
- G06F3/162—Interface to dedicated audio devices, e.g. audio drivers, interface to CODECs
Abstract
The present invention relates to a kind of audio data collecting method and apparatus, the audio data for reducing leakage acquisition improves the integrality of data.Wherein audio data collecting method includes：During detecting audio signal by the first trigger, the second trigger and third trigger, determine that the energy of the audio signal detected by the first trigger is greater than or equal to the trigger threshold of first trigger, it is determined that first trigger is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；The data channel of enabled memory, to be input to the audio data of the audio input system by memory acquisition.
Description
Technical field
The present invention relates to audio signal processing technique field, more particularly to a kind of audio data collecting method and apparatus.
Background technology
Currently, many equipment support voice input, thus it is general all provided with audio input system in equipment.In order to reduce
The power consumption of system, audio input system are at dormant state before entering normal work or carrying out speech recognition, need
It is waken up by voice.For example, if the energy of the voice of input meets touching for the trigger being arranged for audio input system
Thresholding is sent out, then the trigger is triggered, and opens the number of first in, first out (First Input First Output, FIFO) memory
According to access, the audio data of input is acquired, and wake up central processing unit (CPU), be equivalent to and trigger audio input system.
That is, whether audio input system can be triggered, the trigger threshold of the trigger is depended on.It is general at present
By the higher of the trigger threshold setting of the trigger, then, before the trigger is triggered, even if to the audio input system
Audio data more than input again can not all be acquired since the audio input system is not triggered.
As it can be seen that current scheme can lead to leakage acquisition audio data, cause the data of acquisition sufficiently complete.
Invention content
The embodiment of the present invention provides a kind of audio data collecting method and apparatus, the audio data for reducing leakage acquisition,
Improve the integrality of data.
In a first aspect, a kind of audio data collecting method is provided, including：
During detecting audio signal by the first trigger, the second trigger and third trigger, determination passes through
The energy of the audio signal of first trigger detection is greater than or equal to the trigger threshold of first trigger, it is determined that described the
One trigger is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；
The data channel of enabled memory, to be input to the audio of the audio input system by memory acquisition
Data.
Optionally, the method further includes：
In the first preset duration after first trigger is triggered, determine through second trigger detection
The energy of audio signal is greater than or equal to the trigger threshold of second trigger, it is determined that second trigger is triggered；
The trigger threshold of second trigger is more than the second predetermined threshold value；
Whether the data volume for detecting the memory storage reaches the maximum amount of data that the memory may store；
If it is determined that the data volume of the memory storage has reached the maximum amount of data that the memory may store,
Then wake-up processor；
The audio data stored in the memory is read by the processor.
Optionally, after reading the audio data stored in the memory by the processor, further include：
Determine that the audio data read is not enough to carry out speech recognition by the processor；
Indicate that the processor enters dormant state；Or
Determine that the audio data read is sufficient for speech recognition by the processor；
Speech recognition is carried out according to the audio data of reading by the processor.
Optionally, the method further includes：
After first trigger is triggered, determine the energy of audio signal detected by third trigger be more than or
Equal to the trigger threshold of the third trigger, it is determined that the third trigger is triggered；
In the second preset duration after the third trigger is triggered, if the third trigger is not triggered,
Then whether in a dormant state to determine the processor；
If the processor is in the dormant state, the processor is waken up；
The trigger threshold of first trigger and/or second trigger is reset by the processor.
Optionally, touching for first trigger and/or second trigger is being reset by the processor
Before sending out thresholding, further include：
If the trigger threshold of first trigger need to be reset, it is determined that the trigger threshold of first trigger is big
In or equal to first trigger the legal trigger threshold of minimum；Or
If the trigger threshold of second trigger need to be reset, it is determined that the trigger threshold of second trigger is big
In or equal to second trigger the legal trigger threshold of minimum.
Optionally, the method further includes：
In the first preset duration after first trigger is triggered, determine through second trigger detection
The energy of audio signal is respectively less than the trigger threshold of second trigger, it is determined that second trigger is not triggered；Institute
The trigger threshold for stating the second trigger is more than the second predetermined threshold value；
Close the data channel of the memory.
Optionally, the method further includes：
Determine the triggering frequency of second trigger higher than default triggering frequency by the processor；
The trigger threshold of second trigger is reset by the processor.
Second aspect provides a kind of audio data collecting equipment, including：
Data operation module, for detecting audio signal by the first trigger, the second trigger and third trigger
During, determine that the energy of the audio signal detected by the first trigger is greater than or equal to the triggering of first trigger
Thresholding, it is determined that first trigger is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；
Data path control module, the data channel for enabling memory, to be input to by memory acquisition
The audio data of the audio input system.
Optionally, the equipment further includes detection module, wake-up module and operation module；
The data operation module is additionally operable to：In the first preset duration after first trigger is triggered, determine
The energy of the audio signal detected by second trigger is greater than or equal to the trigger threshold of second trigger, then really
Fixed second trigger is triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
The detection module is used for：Whether the data volume for detecting memory storage reaches the memory and may store
Maximum amount of data；
The wake-up module is used for：If it is determined that the data volume of the memory storage has reached the memory and may store
The maximum amount of data, then wake-up processor；
The operation module is used for：The audio data stored in the memory is read by the processor.
Optionally, the equipment further includes system suspend mode control module and identification module；
The operation module is additionally operable to：Determine that the audio data read is not enough to carry out voice knowledge by the processor
Not；
The system suspend mode control module is used for：Indicate that the processor enters dormant state；Or
The operation module is additionally operable to：Determine that the audio data read is sufficient for speech recognition by the processor；
The identification module is used for：Speech recognition is carried out according to the audio data of reading by the processor.
Optionally, the equipment further includes setup module；
The data operation module is additionally operable to：After first trigger is triggered, determination is examined by third trigger
The energy of the audio signal of survey is greater than or equal to the trigger threshold of the third trigger, it is determined that the third trigger is touched
Hair；In the second preset duration after the third trigger is triggered, if the third trigger is not again triggered,
Determine the processor whether in a dormant state；
The wake-up module is additionally operable to：If the processor is in the dormant state, the processor is waken up；
The setup module is used for：By the processor reset first trigger and/or it is described second touch
Send out the trigger threshold of device.
Optionally, the data operation module is additionally operable to：
First trigger and/or second trigger are reset by the processor in the setup module
Trigger threshold before, if the trigger threshold of first trigger need to be reset, it is determined that first trigger touch
Send out the legal trigger threshold of minimum that thresholding is greater than or equal to first trigger；Or
First trigger and/or second trigger are reset by the processor in the setup module
Trigger threshold before, if the trigger threshold of second trigger need to be reset, it is determined that second trigger touch
Send out the legal trigger threshold of minimum that thresholding is greater than or equal to second trigger.
Optionally,
The data operation module is additionally operable to：In the first preset duration after first trigger is triggered, determine
The energy of the audio signal detected by second trigger is respectively less than the trigger threshold of second trigger, it is determined that institute
The second trigger is stated not to be triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
The data path control module is additionally operable to：Close the data channel of the memory.
Optionally, the equipment further includes setup module；
The data operation module is additionally operable to：Determine that the triggering frequency of second trigger is higher than by the processor
Default triggering frequency；
The setup module is used for：The trigger threshold of second trigger is reset by the processor.
In the embodiment of the present invention, the trigger threshold of the first trigger is relatively low, therefore the first trigger is easier to be triggered,
After first trigger is triggered, you can enable the data channel of memory to acquire audio data, in this way, can subtract
The data of few leakage acquisition, that is, more audio datas can be acquired so that the audio data of acquisition is more complete, but also
Subsequent voice recognition result is more accurate.
Description of the drawings
It, below will be to embodiment or the prior art in order to illustrate more clearly of the present invention or technical solution in the prior art
Attached drawing needed in description is briefly described.
Fig. 1 is the flow chart of audio data collecting method provided in an embodiment of the present invention；
Fig. 2 is a kind of structural schematic diagram of audio data collecting equipment provided in an embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of audio data collecting equipment provided in an embodiment of the present invention.
Specific implementation mode
In order to make those skilled in the art more fully understand the technical solution in the embodiment of the present invention, and make the present invention's
Above objects, features, and advantages can be more obvious and easy to understand, makees below in conjunction with the accompanying drawings to technical solution in the present invention further detailed
Thin explanation.
It is introduced below with specific embodiment：
As shown in Figure 1, one embodiment of the invention provides a kind of audio data collecting method, the flow description of this method is such as
Under：
Step 101：During detecting audio signal by the first trigger, the second trigger and third trigger,
Determine that the energy of the audio signal detected by the first trigger is greater than or equal to the trigger threshold of the first trigger, it is determined that the
One trigger is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；
Step 102：The data channel of enabled memory, to be input to the audio of audio input system by memory acquisition
Data.
In general, the trigger threshold of trigger is level value, and level value is related with the energy of sound, it is possible to understand that
For the energy of audio signal is bigger, then corresponding level value is higher.
After device power-up, after the completion of each module register configuration, system (i.e. audio system, hereinafter referred to as system)
Into dormant state, waiting voice wakes up.At this point, the parts such as only the first trigger, the second trigger and audio input access
Dormant state is not entered, to detect audio signal.To reduce power consumption, the other component of equipment may be in closed state.
The embodiment of the present invention provides the first trigger and the second trigger, what the trigger threshold of the first trigger can be arranged
Relatively low, i.e. the triggering level value of the first trigger is relatively low, is, for example, less than the first predetermined threshold value, the audio signal of input is not necessarily to
Excessive energy can trigger the first trigger, it is seen then that the first trigger is easier to be triggered.It is sent out in audio signal
When changing, if the corresponding level value of the energy of audio signal is greater than or equal to the trigger threshold of the first trigger, first touches
Hair device is triggered, you can the data channel of enabled memory, memory here can be FIFO, be also likely to be other certainly
Memory, hereinafter by taking FIFO as an example.Then FIFO can acquire the audio data of input by data channel.Compared to existing
The trigger threshold of trigger is arranged for higher situation in technology, in the embodiment of the present invention thresholding of the first trigger compared with
Low, the audio data of acquisition is more, more completely, avoids leakage to acquire useful audio data as possible so as to pass through audio data
The voice messaging of identification more complete and accurate.
It can be seen that after the first trigger is triggered, only enable the data channel of FIFO, and and be not enabled on processing
Device, that is, can still reduce power consumption while acquiring audio data.Wherein, processor may include CPU or specific application collection
May include that one or more is used to control at circuit (Application Specific Integrated Circuit, ASIC)
The integrated circuit that processing procedure sequence executes, may include use site programmable gate array (Field Programmable Gate
Array, FPGA) exploitation hardware circuit, may include baseband chip.Hereinafter by taking CPU as an example.
Because the embodiment of the present invention additionally provides the second trigger, then the comparison that the trigger threshold of the second trigger can be arranged
The triggering level value of height, i.e. the second trigger can be higher, is greater than the second predetermined threshold value, the trigger threshold of the second trigger
More than the trigger threshold of the first trigger.First predetermined threshold value can be less than or equal to the second predetermined threshold value.
After the first trigger is triggered, other than can be to enable FIFO, the first timer can also be enabled so that the
One timer starts timing, and maximum time when the first timer is the first preset duration.
After the first trigger is triggered, until before the first timer to maximum value, that is, be triggered in the first trigger
In the first preset duration afterwards, if the energy of the audio signal of the second trigger detection is greater than or equal to the triggering of the second trigger
Thresholding, then the second trigger be also triggered.After second trigger is triggered, so that it may to open FIFO's by the second trigger
Detector persistently detects the data volume stored in FIFO by the detector of FIFO.If it is determined that the data volume stored in FIFO
Reach the maximum amount of data that FIFO may store, then CPU wake-up interrupts can be generated by the detector of FIFO, wakes up CPU and come
Read the audio data stored in FIFO.
That is, CPU is waken up after the second trigger is triggered, the trigger threshold of the second trigger is higher, generally not
Easily by false triggering, reduce the probability of false triggering.Such as second the trigger threshold of trigger voice input is formally carried out according to user,
I.e. the mouth of user is designed relatively close to the volume at audio collection end, then, if the second trigger is triggered, generally indicate that
User is carrying out voice input really, and waking up CPU at this time can be with timely processing audio data.
And after the first trigger is triggered, until before the first timer to maximum value, that is, touched in the first trigger
In the first preset duration after hair, if the energy of the audio signal of the second trigger detection is consistently less than the triggering of the second trigger
Thresholding, it is determined that the second trigger is not triggered, then can close the data channel of FIFO, and empties the number stored in FIFO
According to.It if the second trigger is not triggered, can be shown that user does not carry out voice input, then be not necessarily to wake up CPU, and close
The data channel of FIFO, to save power consumption.FIFO is emptied, is to avoid audio number to carry out audio data collecting next time
According to obscuring.
After the second trigger is triggered, if waking up CPU by the detector of FIFO to read the audio stored in FIFO
Data, then CPU can once read all audio frequency data stored in FIFO.After CPU carries out a data fifo reading, such as
Fruit determines data volume deficiency, that is, because the audio data for carrying out speech recognition needs is more, the maximum storage of FIFO is not necessarily
The demand for the data volume for carrying out speech recognition can be met, then CPU may determine that the audio data of reading is not enough to after once reading
Carry out speech recognition waits for the detector of FIFO to be waken up next time then CPU can go successively to dormant state.Pass through
This mode can be further reduced the idle waiting time of CPU, reduce power consumption.After CPU enters dormant state, FIFO's
Detector can still continue the audio data stored in detection FIFO, however, it is determined that the audio data stored in FIFO reaches FIFO institutes
Can storage maximum amount of data, then the detector of FIFO can generate CPU wake-up interrupts again to wake up CPU, to read in FIFO
Audio data.It recycles successively, until the audio data carried out needed for speech recognition reads completion, then CPU enters speech recognition
State.If speech recognition success, system is successfully waken up, into normal mode of operation.And if speech recognition fails,
Then return to step 101 is continued through the first trigger and the second trigger to detect audio data by system.It is recyclable to hold
Row, until voice is successfully identified.
In addition, at the beginning, the trigger threshold of the trigger threshold of the first trigger and the second trigger may be that system is silent
Recognize setting, or may be to be arranged in implementation procedure before.During carrying out audio data collecting, because
The environment that the different time carries out audio data collecting may be different, and environment sometimes is more quiet, environment more noise sometimes
It is miscellaneous, if the trigger threshold of trigger is constant always, the demand of varying environment may be cannot be satisfied.Therefore, implement in the present invention
In example, the trigger threshold of the first trigger and/or the trigger threshold of the second trigger can also be adjusted according to environmental factor, under
Face is introduced.
It, can be to enable the data channel and the first timer of FIFO, in addition, the first triggering after first trigger is triggered
After device is triggered, the second timer can also be enabled so that the second timer starts timing, when maximum when the second timer
A length of second preset duration.The magnitude relationship of first preset duration and the second preset duration is not restricted.For example, second it is default when
It is long to be more than the first preset duration.
The embodiment of the present invention also provides third trigger.After the first trigger is triggered, third trigger starts to detect
Audio signal.If the energy of the audio signal of third trigger detection meets the trigger threshold of third trigger, third is touched
Hair device is triggered, and can reset the second timer by the triggering of third trigger so that the second timer starts reclocking.
After the second timer starts reclocking, if remembered to third trigger before maximum value all not by again in the second timer
Secondary triggering, i.e., third trigger is not again triggered in the second preset duration after third trigger is triggered, it is determined that
Whether in a dormant state CPU, if CPU is in a dormant state, interrupt signal can be generated by the second timer, to wake up
CPU reconfigures the trigger threshold of the first trigger and/or the trigger threshold of the second trigger by CPU, is reconfiguring
After trigger threshold, corresponding trigger restarts work.If third trigger is not all by again in the second preset duration
Triggering, shows that the volume of the audio data actually entered is smaller, and possible environment is more quiet, then, after ensureing
Smoothly acquisition, CPU can turn the trigger threshold of the second trigger down, to meet current environment.
In addition, in the embodiment of the present invention, it may be possible to the trigger threshold of the trigger threshold ＜ third triggers of the first trigger
The trigger threshold of the second triggers of ＜, or may be the trigger threshold ＜ of the first triggers of trigger threshold ＜ of third trigger
The trigger threshold of second trigger.Second timer is triggered by third trigger, avoids the second timer by the second triggering
The limitation of the high trigger threshold of device.
In addition, adjusting the trigger threshold of the first trigger and/or the trigger threshold of the second trigger by CPU, before,
If it is determined that the trigger threshold of the first trigger need to be adjusted, it is determined that whether the current trigger threshold of the first trigger has been equal to first
The legal trigger threshold of minimum of trigger, if the minimum that the current trigger threshold of the first trigger is more than the first trigger is legal
Trigger threshold can then turn the trigger threshold of the first trigger down, and if the current trigger threshold of the first trigger is equal to
The legal trigger threshold of minimum of first trigger, then no longer adjust the trigger threshold of the first trigger.Second trigger is come
Say to be also the same, however, it is determined that the trigger threshold of the second trigger need to be adjusted, it is determined that the current trigger threshold of the second trigger is
The no legal trigger threshold of minimum for being equal to the second trigger, if the current trigger threshold of the second trigger is more than the second triggering
The legal trigger threshold of minimum of device, then can turn the trigger threshold of the second trigger down, and if the second trigger is current
Trigger threshold is equal to the legal trigger threshold of minimum of the second trigger, then no longer adjusts the trigger threshold of the second trigger.Triggering
The legal trigger threshold of minimum of device can be related with factors such as the performances of trigger, and in this way, guarantee was not in
The case where degree adjustment trigger threshold.
It is described above the feelings for turning the trigger threshold of the trigger threshold of the first trigger and/or the second trigger down
Condition perhaps further relates to touching the trigger threshold of the first trigger and/or the second trigger in the possible scene of other
The case where hair thresholding tunes up, is described below.
For example, can be monitored by processor, if finding, the frequency that the second trigger is triggered triggers frequency higher than default,
The number that i.e. the second trigger is triggered whithin a period of time is excessive, it is judged that the trigger threshold setting of the second trigger
It is too low, it is excessively easy to be triggered, may result in false triggering, also increase power consumption.So, second can be adjusted by CPU to touch
Send out the trigger threshold of device so that the trigger threshold of the second trigger is more suitable for current environment.At this point, the first trigger and/or
The trigger threshold of third trigger can also optionally determine adjustment or not adjust.
It should be noted that the embodiment of the present invention introduced the first trigger, the second trigger, third trigger, first
The detector of timer, the second timer and FIFO can be realized by software form, or can also pass through hardware shape
Formula realizes that the embodiment of the present invention is not restricted, and is realized to be integrated into audio system by integrated circuit form
For.
Equipment provided in an embodiment of the present invention is introduced below in conjunction with the accompanying drawings.
Refer to Fig. 2, one embodiment of the invention provides a kind of audio data collecting equipment, if by the first trigger, the
Two triggers, third trigger, the first timer, the second timer and FIFO detector be collectively regarded as triggering system, i.e.,
First trigger, the second trigger, third trigger, the first timer, the second timer and FIFO detector be all tactile
The component part of hair system, then the audio data collecting equipment also belongs to the triggering system, such as the audio data collecting is set
It is standby to be equivalent to controller in the triggering system.It is so predictable, the work(included by the audio data collecting equipment
Energy module can be realized by software, or can be realized by hardware.Processor (such as CPU) as previously mentioned and
FIFO belongs to audio input system.Triggering system and audio input system can be located in same equipment, for example, triggering system and
Audio input system can be integrated in a chip.The equipment includes data operation module 201 and data path control module
202。
Wherein, data operation module 201, for being detected by the first trigger, the second trigger and third trigger
During audio signal, determine that the energy of the audio signal detected by the first trigger is greater than or equal to described first and triggers
The trigger threshold of device, it is determined that first trigger is triggered；It is default that the trigger threshold of first trigger is less than first
Threshold value；
Data path control module 202, the data channel for enabling memory are inputted with being acquired by the memory
To the audio data of the audio input system.
In possible embodiment, which further includes detection module 203, wake-up module 204 and operation module 205,
Refer to Fig. 3.Because detection module 203, wake-up module 204 and operation module 205 are optional function modules, in order to it is essential
Function module mutually distinguish, detection module 203, wake-up module 204 and operation module 205 are depicted as dotted line in figure 3.
Data operation module 201 is additionally operable to：In the first preset duration after first trigger is triggered, determine logical
The energy for crossing the audio signal of the second trigger detection is greater than or equal to the trigger threshold of second trigger, it is determined that
Second trigger is triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
Detection module 203 is used for：Whether the data volume for detecting memory storage reaches the memory and may store
Maximum amount of data；
Wake-up module 204 is used for：If it is determined that the data volume of the memory storage has reached the memory and may store
The maximum amount of data, then wake-up processor；
Operation module 205 is used for：The audio data stored in the memory is read by the processor.
In possible embodiment, which further includes system suspend mode control module 206 and identification module 207, please after
Continue referring to Fig. 3.Because system suspend mode control module 206 and identification module 207 are optional function modules, in order to essential work(
Energy module is mutually distinguished, and system suspend mode control module 206 and identification module 207 are depicted as dotted line in figure 3.
Operation module 205 is additionally operable to：Determine that the audio data read is not enough to carry out speech recognition by the processor；
System suspend mode control module 206 is used for：Indicate that the processor enters dormant state；Or
Operation module 205 is additionally operable to：Determine that the audio data read is sufficient for speech recognition by the processor；
Identification module 207 is used for：Speech recognition is carried out according to the audio data of reading by the processor.
In possible embodiment, which further includes setup module 208, continuing with referring to Fig. 3.Because of setup module
208 be optional function module, and in order to mutually be distinguished with essential function module, setup module 208 is depicted as dotted line in figure 3.
Data operation module 201 is additionally operable to：After first trigger is triggered, determination is detected by third trigger
Audio signal energy be greater than or equal to the third trigger trigger threshold, it is determined that the third trigger is touched
Hair；In the second preset duration after the third trigger is triggered, if the third trigger is not again triggered,
Determine the processor whether in a dormant state；
Wake-up module 204 is additionally operable to：If the processor is in the dormant state, the processor is waken up；
Setup module 208 is used for：First trigger and/or second triggering are reset by the processor
The trigger threshold of device.
In possible embodiment, data operation module 201 is additionally operable to：
First trigger and/or second trigger are reset by the processor in setup module 208
Trigger threshold before, if the trigger threshold of first trigger need to be reset, it is determined that first trigger touch
Send out the legal trigger threshold of minimum that thresholding is greater than or equal to first trigger；Or
First trigger and/or second trigger are reset by the processor in setup module 208
Trigger threshold before, if the trigger threshold of second trigger need to be reset, it is determined that second trigger touch
Send out the legal trigger threshold of minimum that thresholding is greater than or equal to second trigger.
In possible embodiment,
Data operation module 201 is additionally operable to：In the first preset duration after first trigger is triggered, determine logical
The energy for crossing the audio signal of the second trigger detection is respectively less than the trigger threshold of second trigger, it is determined that described
Second trigger is not triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
Data path control module 202 is additionally operable to：Close the data channel of the memory.
In possible embodiment, which further includes setup module 208, as shown in Figure 3.
Data operation module 201 is additionally operable to：The triggering frequency of second trigger is determined by second trigger
Higher than default triggering frequency；
Setup module 208 is used for：The trigger threshold of second trigger is reset by the processor.
The audio data collecting equipment can be used for executing the method that embodiment shown in FIG. 1 provides, therefore this is set
The function etc. that each function module in standby can be realized can refer to the introduction of embodiment shown in FIG. 1, seldom repeat.
In the embodiment of the present invention, the trigger threshold of the first trigger is relatively low, therefore the first trigger is easier to be triggered,
After first trigger is triggered, you can enable the data channel of memory to acquire audio data, in this way, can subtract
The data of few leakage acquisition, that is, more audio datas can be acquired so that the audio data of acquisition is more complete, but also
Subsequent voice recognition result is more accurate.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement etc., should all be included in the protection scope of the present invention made by within refreshing and principle.
Claims (14)
1. a kind of audio data collecting method, which is characterized in that including：
During detecting audio signal by the first trigger, the second trigger and third trigger, determines and pass through first
The energy of the audio signal of trigger detection is greater than or equal to the trigger threshold of first trigger, it is determined that described first touches
Hair device is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；
The data channel of enabled memory, to be input to the audio number of the audio input system by memory acquisition
According to.
2. the method as described in claim 1, which is characterized in that the method further includes：
In the first preset duration after first trigger is triggered, the audio detected by second trigger is determined
The energy of signal is greater than or equal to the trigger threshold of second trigger, it is determined that second trigger is triggered；It is described
The trigger threshold of second trigger is more than the second predetermined threshold value；
Whether the data volume for detecting the memory storage reaches the maximum amount of data that the memory may store；
If it is determined that the data volume of the memory storage has reached the maximum amount of data that the memory may store, then call out
Awake processor；
The audio data stored in the memory is read by the processor.
3. method as claimed in claim 2, which is characterized in that stored in reading the memory by the processor
After audio data, further include：
Determine that the audio data read is not enough to carry out speech recognition by the processor；
Indicate that the processor enters dormant state；Or
Determine that the audio data read is sufficient for speech recognition by the processor；
Speech recognition is carried out according to the audio data of reading by the processor.
4. method as claimed in claim 2 or claim 3, which is characterized in that the method further includes：
After first trigger is triggered, determine that the energy of the audio signal detected by third trigger is greater than or equal to
The trigger threshold of the third trigger, it is determined that the third trigger is triggered；
In the second preset duration after the third trigger is triggered, if the third trigger is not triggered, really
In a dormant state whether the fixed processor；
If the processor is in the dormant state, the processor is waken up；
The trigger threshold of first trigger and/or second trigger is reset by the processor.
5. method as claimed in claim 4, which is characterized in that resetting first trigger by the processor
And/or before the trigger threshold of second trigger, further include：
If the trigger threshold of first trigger need to be reset, it is determined that the trigger threshold of first trigger be more than or
Equal to the legal trigger threshold of minimum of first trigger；Or
If the trigger threshold of second trigger need to be reset, it is determined that the trigger threshold of second trigger be more than or
Equal to the legal trigger threshold of minimum of second trigger.
6. the method as described in claim 1, which is characterized in that the method further includes：
In the first preset duration after first trigger is triggered, the audio detected by second trigger is determined
The energy of signal is respectively less than the trigger threshold of second trigger, it is determined that second trigger is not triggered；Described
The trigger threshold of two triggers is more than the second predetermined threshold value；
Close the data channel of the memory.
7. the method as described in claim 1, which is characterized in that the method further includes：
Determine the triggering frequency of second trigger higher than default triggering frequency by the processor；
The trigger threshold of second trigger is reset by the processor.
8. a kind of audio data collecting equipment, which is characterized in that including：
Data operation module, in the mistake for detecting audio signal by the first trigger, the second trigger and third trigger
Cheng Zhong determines that the energy of the audio signal detected by the first trigger is greater than or equal to the trigger gate of first trigger
Limit, it is determined that first trigger is triggered；The trigger threshold of first trigger is less than the first predetermined threshold value；
Data path control module, the data channel for enabling memory are described to be input to by memory acquisition
The audio data of audio input system.
9. equipment as claimed in claim 8, which is characterized in that the equipment further includes detection module, wake-up module and operation
Module；
The data operation module is additionally operable to：In the first preset duration after first trigger is triggered, determination passes through
The energy of the audio signal of the second trigger detection is greater than or equal to the trigger threshold of second trigger, it is determined that institute
The second trigger is stated to be triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
The detection module is used for：Whether the data volume for detecting memory storage reaches the memory and may store most
Big data quantity；
The wake-up module is used for：If it is determined that the data volume of the memory storage has reached the institute that the memory may store
Maximum amount of data is stated, then wake-up processor；
The operation module is used for：The audio data stored in the memory is read by the processor.
10. equipment as claimed in claim 9, which is characterized in that the equipment further includes system suspend mode control module and identification
Module；
The operation module is additionally operable to：Determine that the audio data read is not enough to carry out speech recognition by the processor；
The system suspend mode control module is used for：Indicate that the processor enters dormant state；Or
The operation module is additionally operable to：Determine that the audio data read is sufficient for speech recognition by the processor；
The identification module is used for：Speech recognition is carried out according to the audio data of reading by the processor.
11. the equipment as described in claim 9 or 10, which is characterized in that the equipment further includes setup module；
The data operation module is additionally operable to：After first trigger is triggered, what determination was detected by third trigger
The energy of audio signal is greater than or equal to the trigger threshold of the third trigger, it is determined that the third trigger is triggered；
In the second preset duration after the third trigger is triggered, if the third trigger is not again triggered, really
In a dormant state whether the fixed processor；
The wake-up module is additionally operable to：If the processor is in the dormant state, the processor is waken up；
The setup module is used for：First trigger and/or second trigger are reset by the processor
Trigger threshold.
12. equipment as claimed in claim 11, which is characterized in that the data operation module is additionally operable to：
Touching for first trigger and/or second trigger is reset by the processor in the setup module
Before sending out thresholding, if the trigger threshold of first trigger need to be reset, it is determined that the trigger gate of first trigger
Minimum legal trigger threshold of the limit more than or equal to first trigger；Or
Touching for first trigger and/or second trigger is reset by the processor in the setup module
Before sending out thresholding, if the trigger threshold of second trigger need to be reset, it is determined that the trigger gate of second trigger
Minimum legal trigger threshold of the limit more than or equal to second trigger.
13. equipment as claimed in claim 8, which is characterized in that
The data operation module is additionally operable to：In the first preset duration after first trigger is triggered, determination passes through
The energy of the audio signal of second trigger detection is respectively less than the trigger threshold of second trigger, it is determined that described the
Two triggers are not triggered；The trigger threshold of second trigger is more than the second predetermined threshold value；
The data path control module is additionally operable to：Close the data channel of the memory.
14. the equipment as described in claim 8-13 is any, which is characterized in that the equipment further includes setup module；
The data operation module is additionally operable to：Determine the triggering frequency of second trigger higher than default by the processor
Trigger frequency；
The setup module is used for：The trigger threshold of second trigger is reset by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710127263.2A CN108536413B (en) | 2017-03-06 | 2017-03-06 | Audio data acquisition method and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710127263.2A CN108536413B (en) | 2017-03-06 | 2017-03-06 | Audio data acquisition method and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108536413A true CN108536413A (en) | 2018-09-14 |
| CN108536413B CN108536413B (en) | 2021-03-26 |
Family
ID=63489561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710127263.2A Active CN108536413B (en) | 2017-03-06 | 2017-03-06 | Audio data acquisition method and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108536413B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113433926A (en) * | 2021-06-21 | 2021-09-24 | 广州鲁邦通物联网科技有限公司 | Ultra-low power consumption fault detection method and ultra-low power consumption elevator opening/closing door fault detection device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150046161A1 (en) * | 2013-08-07 | 2015-02-12 | Lenovo (Singapore) Pte. Ltd. | Device implemented learning validation |
| CN105094816A (en) * | 2015-07-09 | 2015-11-25 | 北京君正集成电路股份有限公司 | Method for reducing power consumption of intelligent device and intelligent device |
| CN105261368A (en) * | 2015-08-31 | 2016-01-20 | 华为技术有限公司 | Voice wake-up method and apparatus |
| CN105336330A (en) * | 2015-10-15 | 2016-02-17 | 上海易景信息科技有限公司 | Simple voice recognizing system |
| CN106062661A (en) * | 2014-03-31 | 2016-10-26 | 英特尔公司 | Location aware power management scheme for always-on-always-listen voice recognition system |
-
2017
- 2017-03-06 CN CN201710127263.2A patent/CN108536413B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150046161A1 (en) * | 2013-08-07 | 2015-02-12 | Lenovo (Singapore) Pte. Ltd. | Device implemented learning validation |
| CN106062661A (en) * | 2014-03-31 | 2016-10-26 | 英特尔公司 | Location aware power management scheme for always-on-always-listen voice recognition system |
| CN105094816A (en) * | 2015-07-09 | 2015-11-25 | 北京君正集成电路股份有限公司 | Method for reducing power consumption of intelligent device and intelligent device |
| CN105261368A (en) * | 2015-08-31 | 2016-01-20 | 华为技术有限公司 | Voice wake-up method and apparatus |
| CN105336330A (en) * | 2015-10-15 | 2016-02-17 | 上海易景信息科技有限公司 | Simple voice recognizing system |
Non-Patent Citations (2)
| Title |
|---|
| M. CHUAH ETAL.: "Smartphone Based Autism Social Alert System", 《IEEE》 * |
| 陈静等: "基于单片机和FPGA的多路音视频控制系统设计", 《数据采集与处理》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113433926A (en) * | 2021-06-21 | 2021-09-24 | 广州鲁邦通物联网科技有限公司 | Ultra-low power consumption fault detection method and ultra-low power consumption elevator opening/closing door fault detection device |
| CN113433926B (en) * | 2021-06-21 | 2022-05-17 | 广州鲁邦通物联网科技股份有限公司 | Ultra-low power consumption fault detection method and ultra-low power consumption elevator opening/closing door fault detection device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108536413B (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8464183B2 (en) | System and method for distinguishing multimodal commands directed at a machine from ambient human communications | |
| CN104035743B (en) | System for carrying out processor wake-up based on sensing data | |
| US8972252B2 (en) | Signal processing apparatus having voice activity detection unit and related signal processing methods | |
| US10651827B2 (en) | Apparatus and method for activating circuits | |
| KR101994569B1 (en) | Clock Switching on Constant-On Components | |
| US20140372737A1 (en) | Electronic device and method for automatically waking up operating system of electronic device | |
| WO2016168982A1 (en) | Method, apparatus and terminal device for setting interrupt threshold for fingerprint identification device | |
| CN105094816A (en) | Method for reducing power consumption of intelligent device and intelligent device | |
| US11620995B2 (en) | Voice interaction processing method and apparatus | |
| CN111679861A (en) | Wake-up apparatus and method for electronic device, computer device and storage medium | |
| US11614790B2 (en) | Electronic pen, method and device for controlling electronic pen, and computer-readable storage medium | |
| WO2018053714A1 (en) | Single-chip microcomputer system, and reset method for single-chip microcomputer system | |
| CN108845727B (en) | Mutual capacitance type touch screen based awakening method and device | |
| CN106453911B (en) | A kind of shutdown control method of mobile terminal, device and mobile terminal | |
| US8659563B2 (en) | Electronic device with a page turning function during a sleep mode of the electronic device | |
| TWI633425B (en) | Microphone apparatus | |
| CN111199733A (en) | Multi-stage recognition voice awakening method and device, computer storage medium and equipment | |
| CN105205377A (en) | Device, method and mobile terminal for preventing mistakenly starting of camera | |
| CN106650383A (en) | Terminal control method and terminal control device | |
| CN108536413A (en) | A kind of audio data collecting method and apparatus | |
| CN104598192A (en) | Information processing method and electronic equipment | |
| CN107436830A (en) | Use control method, system and electronic equipment | |
| CN111179924B (en) | Method and system for optimizing awakening performance based on mode switching | |
| CN108536412A (en) | A kind of audio data collecting method and apparatus | |
| US20120260116A1 (en) | External power-saving usb mass storage device and power-saving method thereof |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |