Movatterモバイル変換


[0]ホーム

URL:


CN106612367A - Speech wake method based on microphone and mobile terminal - Google Patents

Speech wake method based on microphone and mobile terminal
Download PDF

Info

Publication number
CN106612367A
CN106612367ACN201510698609.5ACN201510698609ACN106612367ACN 106612367 ACN106612367 ACN 106612367ACN 201510698609 ACN201510698609 ACN 201510698609ACN 106612367 ACN106612367 ACN 106612367A
Authority
CN
China
Prior art keywords
signal
wake
voice
mobile terminal
unit
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
Application number
CN201510698609.5A
Other languages
Chinese (zh)
Inventor
叶菁华
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Yutaixin Microelectronics Technology Shanghai Co Ltd
Zilltek Technology Corp
Original Assignee
Yutaixin Microelectronics Technology Shanghai Co Ltd
Zilltek Technology Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Yutaixin Microelectronics Technology Shanghai Co Ltd, Zilltek Technology CorpfiledCriticalYutaixin Microelectronics Technology Shanghai Co Ltd
Priority to CN201510698609.5ApriorityCriticalpatent/CN106612367A/en
Publication of CN106612367ApublicationCriticalpatent/CN106612367A/en
Pendinglegal-statusCriticalCurrent

Links

Classifications

Landscapes

Abstract

The invention relates to the technical field of sound signal processing, and particularly relates to a speech wake method based on a microphone and a mobile terminal. According to the invention, a detection unit collects the current environment every 10 milliseconds to form a detection signal; the detection unit of the mobile terminal does not need to always be standby, which reduces the use frequency of the detection unit; a control calculation unit calculates the 1-millisecond signal of the detection signal to form a calculation result; only 1-millisecond information is detected and judged; judging all information of 10 milliseconds is not needed; and the power consumption is reduced by 90%.

Description

A kind of voice awakening method and mobile terminal based on mike
Technical field
The present invention relates to a kind of sound signal processing technical field, more particularly to a kind of language based on mikeSound awakening method and mobile terminal.
Background technology
In the last few years, with the fast development of speech recognition technology, Siri is similar to, search dog voice assistant is soUtilization voice realize that control, the mobile terminal of semantic understanding or operating system start to emerge in multitude.But,Most of such applications are required for user to click on button can start mobile terminal or operating system to start,In some occasions, during such as driving a car, it is not easy to user and uses.It is exactly to solve that voice wakes upSuch demand and the function that produces.The technology for adding voice to wake up in systems, user does not need and equipmentProduce physical contact, it is only necessary to say keyword, such as " making a phone call ", mobile terminal or operating system areCan start.
Existing intelligent mobile terminal, with voice arousal function, it is mobile terminal in standbyUnder state, voice wakeup unit can at any time be detected and be presently in environment whether having wake-up signal, in working asContaining in the state of wake-up signal in the front environment, according to the wake-up signal activation mobile terminal, and controlMobile terminal processed performs the operation matched with wake-up signal, but is equipped with this kind of voice arousal functionMobile terminal has a larger defect --- and power consumption is increased, in stand-by mode, voice wakeup unitUnder being constantly in working condition, the power consumption that it is consumed is larger;Existing intelligent mobile terminal, its energy storage dressThe capacity put is mostly 1500mA~3500mA, intelligent mobile terminal, and the consumption of its electric energy is inherently veryGreatly, and under standby mode, the electric current about 0.1mA of consumption, if waking up the electric energy for consuming plus voice,Then the intelligent mobile terminal actually used time then further shorten, by taking i Phone as an example, in order to avoidVoice wakes up and consumes excessive electric energy, and existing i Phone can only be called out in the state of charging using voiceAwake function.
The content of the invention
For the deficiencies in the prior art, the present invention provides a kind of voice awakening method and shifting based on mikeDynamic terminal.
The technical purpose of the present invention is realized by following technological means:
A kind of voice awakening method based on mike, is applied to mobile terminal, wherein, including:
Step S1, it is under standby mode in the mobile terminal, control detector unit is predetermined every firstInterval time is acquired to being presently in environment, and forms a detection signal;
Step S2, control computing unit are counted to the signal in the second predetermined space of the detection signalCalculate, and form a result of calculation;
Step S3, the result of calculation is compared with a threshold signal, in the result of calculation institute is less thanWhen stating threshold signal, execution step S1;
Step S4, one voice wakeup unit of activation, control the voice wakeup unit acquisition and are presently in ringThe wake-up signal in border, and the control signal that output matches with the wake-up signal.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S1,First predetermined space is 10 milliseconds.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S2,Second predetermined space is 1 millisecond.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S1,Also specifically include:
Step S11, arranges the threshold signal of the detector unit.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S4,Specifically include:
Step S40, the activation voice wakeup unit, control the voice wakeup unit acquisition and are presently inThe wake-up signal of environment,
Step S41, judge that the wake-up signal is whether comprising effective voice signal;In the wake-up signalComprising under effective voice signal state, execution step S42 does not include described in the wake-up signalEfficient voice signal condition, execution step S1,
Step S42, controls the control signal that the wakeup unit output matches with the wake-up signal.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S4,Specifically include:
Step S40:Define and do not include the number of times of the efficient voice signal condition in the wake-up signal and beI, wherein I=0;
Step S41, the activation voice wakeup unit, control the voice wakeup unit acquisition and are presently inThe wake-up signal of environment,
Step S42, judge that the wake-up signal is whether comprising effective voice signal;In the wake-up signalComprising under effective voice signal state, execution step S43 does not include described in the wake-up signalEfficient voice signal condition, execution step S46,
Step S43, I=I++
Step S44, judge whether I≤5 are true;When I≤5, execution step S1;
Step S45, when I is not less than 5, improves the threshold signal of the detector unit, and performsStep S1;
Step S46, controls the control signal that the wakeup unit output matches with the wake-up signal.
Preferably, the above-mentioned voice awakening method based on mike, wherein, in step S2, alsoSpecifically include:
Step S20, arranges the collection frequency domain of the detector unit;
Specifically include:
Step S21:The frequency domain of presently described detection signal is detected,
Step S22:The collection frequency domain of the detector unit is adjusted, with the collection frequency domain modelEnclose the matching frequency domain.
A kind of mobile terminal, wherein, an including mike, the mike is specifically included:
Detection means, is under standby mode, every the first predetermined interval to working as in the mobile terminalFront local environment is acquired, and forms a detection signal;
Computing unit, calculates the signal in the second predetermined space of the detection signal, and forms oneResult of calculation;
Comparing unit, compares to the result of calculation with a threshold signal, and in the result of calculation institute is more thanWhen stating threshold signal, a wake-up signal is exported;
Voice wakeup unit, controls the voice wakeup unit and obtains presently described wake-up signal, and export withThe control signal that the wake-up signal matches.
Preferably, above-mentioned mobile terminal, wherein, also including an adjustment unit, to adjust the detectionThe threshold signal of unit, so that the threshold signal matches the presently described environment of the mobile terminal.
Preferably, above-mentioned mobile terminal, wherein, the adjustment unit is also to adjust the detector unitFrequency domain so that the frequency domain matches the frequency domain of the wake-up signal.
Compared with prior art, it is an advantage of the invention that:
In the application, detector unit is acquired every 10 milliseconds to being presently in environment, and forms an inspectionSurvey signal;Without mobile terminal detector unit be constantly in and await orders in the state of, reduce detector unitUsage frequency, control computing unit and 1 millisecond of the detection signal of signal calculated, and shapeInto a result of calculation, only detection judgement is carried out to 1 millisecond of information, without the institute to 10 millisecondsThere is information to be judged, make lower power consumption 90%.
Description of the drawings
Fig. 1 is a kind of voice awakening method schematic flow sheet based on mike of the present invention.
Fig. 2 is that a kind of a kind of embodiment flow process of the voice awakening method based on mike is illustrated in the present inventionFigure;
Fig. 3 is that a kind of a kind of embodiment flow process of the voice awakening method based on mike is illustrated in the present inventionFigure;
Fig. 4 is a kind of structural representation of mobile terminal in the present invention.
Specific embodiment
Below in conjunction with the accompanying drawings the invention will be further described with specific embodiment, but not as the present invention'sLimit.
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried outClearly and completely describe, it is clear that described embodiment is only a part of embodiment of the invention, andIt is not all, of embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art are without workThe every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that the embodiment and the spy in embodiment in the case where not conflicting, in the present inventionLevying to be mutually combined.
Below in conjunction with the accompanying drawings the invention will be further described with specific embodiment, but not as the present invention'sLimit.
As shown in figure 1, the voice awakening method based on mike, is applied to mobile terminal, wherein, bagInclude:
Step S1, in the mobile terminal be in standby mode under, control detector unit it is pre- every firstDetermine to be acquired to being presently in environment interval time, and form a detection signal;Further, it is describedFirst predetermined space is 10 milliseconds.
Step S2, control computing unit are carried out to the signal in the second predetermined space of the detection signalCalculate, and form a result of calculation;Further, second predetermined space is 1 millisecond.
Step S3, the result of calculation is compared with a threshold signal, in the result of calculation institute is less thanWhen stating threshold signal, execution step S1;
Step S4, one voice wakeup unit of activation, control the voice wakeup unit acquisition and are presently in ringThe wake-up signal in border, and the control signal that output matches with the wake-up signal.
In the application, control detector unit is adopted every the first predetermined interval to being presently in environmentCollection, and form a detection signal;Further, first predetermined space is 10 milliseconds, rather than is movedDynamic terminal is constantly in the state of awaiting orders, such as in 60 second time, using the present invention, mobile terminal is onlyNeed at most detect 6 times, and the mobile terminal of prior art needs detection 60 seconds, using work(of the present inventionConsumption is substantially reduced.Then control computing unit to enter the signal in the second predetermined space of the detection signalRow is calculated, and forms a result of calculation;Further, second predetermined space is 1 millisecond.This ShenPlease in, only detection judgement is carried out to 1 millisecond of information, enter without all information to 10 millisecondsRow judgement, makes lower power consumption 90%, but this kind of mode has a shortcoming, may exactly lose 90%Effective information, but the sound propagation velocity of people is relatively slow, and the loss of 9 milliseconds of data message is notAffect whole wake-up system.The result of calculation is compared with a threshold signal, in the result of calculationDuring less than the threshold signal, illustrate to be presently in environment no presence of effective wake-up signal, it is mobile wholeEnd is continued to operate under holding state, continues gathered data information, when the result of calculation is more than the thresholdDuring value signal;Illustrate to be presently in the presence of effective wake-up signal in environment, activate a voice wakeup unit,Control the voice wakeup unit and obtain the wake-up signal for being presently in environment, and export and the wake-up letterNumber control signal for matching.
In the application, detector unit is acquired every 10 milliseconds to being presently in environment, and forms an inspectionSurvey signal;Without mobile terminal detector unit be constantly in and await orders in the state of, reduce detector unitUsage frequency, control computing unit and 1 millisecond of the detection signal of signal calculated, and shapeInto a result of calculation, only detection judgement is carried out to 1 millisecond of information, without the institute to 10 millisecondsThere is information to be judged, make lower power consumption 90%.
As further preferred embodiment, the above-mentioned voice awakening method based on mike, wherein, inIn step S1, also specifically include:
Step S11, arranges the threshold signal of the detector unit.When the detection letter that detector unit is obtainedNumber mismatch threshold signal when, mobile terminal will not be activated, otherwise then be activated.Avoid detector unitWhen getting any acoustical signal (invalid acoustical signal), mobile terminal is activated.
As shown in Fig. 2 as further preferred embodiment, the above-mentioned voice wake-up side based on mikeMethod, wherein, in step S4, specifically include:
Step S40, the activation voice wakeup unit, control the voice wakeup unit acquisition and are presently inThe wake-up signal of environment,
Step S41, judge that the wake-up signal is whether comprising effective voice signal;In the wake-up signalComprising under effective voice signal state, execution step S42 does not include described in the wake-up signalEfficient voice signal condition, execution step S1,
Step S42, controls the control signal that the wakeup unit output matches with the wake-up signal.
As shown in figure 3, as further preferred embodiment, the above-mentioned voice wake-up side based on mikeMethod, wherein, in step S4, specifically include:
Step S40:Define and do not include the number of times of the efficient voice signal condition in the wake-up signal and beI, wherein I=0;
Step S41, the activation voice wakeup unit, control the voice wakeup unit acquisition and are presently inThe wake-up signal of environment,
Step S42, judge the wake-up signal whether comprising effective voice signal;In the wake-up signal bagContaining under effective voice signal state, execution step S43, not including in the wake-up signal described hasEffect voice signal state, execution step S46,
Step S43, I=I++
Step S44, judge whether I≤5 are true;When I≤5, execution step S1;
Step S45, when I is not less than 5, improves the threshold signal of the detector unit, and performsStep S1;
Step S46, controls the control signal that the wakeup unit output matches with the wake-up signal.
Above-mentioned steps, it mainly provides the method for adjusting threshold signal.
When user is among noisy environment, to avoid mobile terminal from being activated by " accident ", threshold value is adjustedSignal (heightens threshold value), when mobile terminal is activated more than 5 times by " accident ", that is, heightens threshold signal,This time 5 times is only a kind of way of example, and user could be arranged to other quantity.
As further preferred embodiment, the above-mentioned voice awakening method based on mike, wherein, inStep S2, also specifically includes:
Step S20, arranges the collection frequency domain of the detector unit;
Specifically include:
Step S21:The frequency domain of presently described detection signal is detected,
Step S22:The collection frequency domain of the detector unit is adjusted, with the collection frequency domain modelEnclose the matching frequency domain.
The scope of above-mentioned regulation frequency domain, to be set to frequency domain most to connect with user voice signalNear state, because the sound of each user there are different difference, is provided with frequency domain, keeps awayExempt from the acoustical signal activation mobile terminal that other sound sources send.
As shown in figure 4, present invention simultaneously provides a kind of mobile terminal, wherein, including a mike, it is describedMike is specifically included:
Detection means, is under standby mode, every the first predetermined interval to working as in the mobile terminalFront local environment is acquired, and forms a detection signal;
Computing unit, calculates the signal in the second predetermined space of the detection signal, and forms oneResult of calculation;
Comparing unit, compares to the result of calculation with a threshold signal, and in the result of calculation institute is more thanWhen stating threshold signal, a wake-up signal is exported;
Voice wakeup unit, controls the voice wakeup unit and obtains presently described wake-up signal, and export withThe control signal that the wake-up signal matches.
As further preferred embodiment, above-mentioned mobile terminal, wherein, also including an adjustment unit,To the threshold signal for adjusting the detector unit, so that the threshold signal matches the mobile terminal and works asThe front environment.
As further preferred embodiment, above-mentioned mobile terminal, wherein, the adjustment unit also toThe frequency domain of the detector unit is adjusted, so that the frequency domain matches the frequency domain of the wake-up signal.
A kind of mobile terminal, its operation principle with based on mike voice awakening method operation principle phaseSeemingly, here is omitted.
More than above are only preferred embodiments of the present invention, not thereby limit embodiments of the present invention andProtection domain, to those skilled in the art, should can appreciate that all utilization description of the inventionAnd the equivalent done by diagramatic content and the scheme obtained by obvious change, should includeWithin the scope of the present invention.

Claims (10)

CN201510698609.5A2015-10-232015-10-23Speech wake method based on microphone and mobile terminalPendingCN106612367A (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN201510698609.5ACN106612367A (en)2015-10-232015-10-23Speech wake method based on microphone and mobile terminal

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN201510698609.5ACN106612367A (en)2015-10-232015-10-23Speech wake method based on microphone and mobile terminal

Publications (1)

Publication NumberPublication Date
CN106612367Atrue CN106612367A (en)2017-05-03

Family

ID=58613229

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN201510698609.5APendingCN106612367A (en)2015-10-232015-10-23Speech wake method based on microphone and mobile terminal

Country Status (1)

CountryLink
CN (1)CN106612367A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN107590096A (en)*2017-08-312018-01-16联想(北京)有限公司Method and processor for processor in electronic equipment
CN109275056A (en)*2018-10-312019-01-25南昌与德软件技术有限公司A kind of method of microphone and reception sound
CN110148405A (en)*2019-04-102019-08-20北京梧桐车联科技有限责任公司Phonetic order processing method and processing device, electronic equipment and storage medium
CN111123799A (en)*2019-12-312020-05-08蔚蓝计划(北京)科技有限公司Voice garbage can and power-saving control method, device and system thereof
CN113038214A (en)*2021-03-032021-06-25深圳创维-Rgb电子有限公司Standby control method, terminal device and readable storage medium
CN116153341A (en)*2023-04-202023-05-23深圳锐盟半导体有限公司Control method and device of voice detection device

Citations (9)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN1707613A (en)*2005-05-272005-12-14曲哲Collecting apparatus and method for noise insulation audio frequency
CN103021411A (en)*2012-11-272013-04-03威盛电子股份有限公司Voice control device and voice control method
CN103052001A (en)*2011-10-172013-04-17联想(北京)有限公司Intelligent device and control method thereof
CN104023121A (en)*2014-04-302014-09-03小米科技有限责任公司Mode switching method and device
US20140278435A1 (en)*2013-03-122014-09-18Nuance Communications, Inc.Methods and apparatus for detecting a voice command
US20140334645A1 (en)*2013-05-072014-11-13Qualcomm IncorporatedMethod and apparatus for controlling voice activation
CN104269016A (en)*2014-09-222015-01-07北京奇艺世纪科技有限公司Alarm method and device
CN104866067A (en)*2015-05-112015-08-26联想(北京)有限公司Low power consumption control method and electronic device
CN205545492U (en)*2015-10-232016-08-31钰太芯微电子科技(上海)有限公司Mobile terminal

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN1707613A (en)*2005-05-272005-12-14曲哲Collecting apparatus and method for noise insulation audio frequency
CN103052001A (en)*2011-10-172013-04-17联想(北京)有限公司Intelligent device and control method thereof
CN103021411A (en)*2012-11-272013-04-03威盛电子股份有限公司Voice control device and voice control method
US20140278435A1 (en)*2013-03-122014-09-18Nuance Communications, Inc.Methods and apparatus for detecting a voice command
US20140334645A1 (en)*2013-05-072014-11-13Qualcomm IncorporatedMethod and apparatus for controlling voice activation
CN104023121A (en)*2014-04-302014-09-03小米科技有限责任公司Mode switching method and device
CN104269016A (en)*2014-09-222015-01-07北京奇艺世纪科技有限公司Alarm method and device
CN104866067A (en)*2015-05-112015-08-26联想(北京)有限公司Low power consumption control method and electronic device
CN205545492U (en)*2015-10-232016-08-31钰太芯微电子科技(上海)有限公司Mobile terminal

Cited By (8)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN107590096A (en)*2017-08-312018-01-16联想(北京)有限公司Method and processor for processor in electronic equipment
CN107590096B (en)*2017-08-312021-06-15联想(北京)有限公司Method for processor in electronic equipment and processor
CN109275056A (en)*2018-10-312019-01-25南昌与德软件技术有限公司A kind of method of microphone and reception sound
CN110148405A (en)*2019-04-102019-08-20北京梧桐车联科技有限责任公司Phonetic order processing method and processing device, electronic equipment and storage medium
CN110148405B (en)*2019-04-102021-07-13北京梧桐车联科技有限责任公司Voice instruction processing method and device, electronic equipment and storage medium
CN111123799A (en)*2019-12-312020-05-08蔚蓝计划(北京)科技有限公司Voice garbage can and power-saving control method, device and system thereof
CN113038214A (en)*2021-03-032021-06-25深圳创维-Rgb电子有限公司Standby control method, terminal device and readable storage medium
CN116153341A (en)*2023-04-202023-05-23深圳锐盟半导体有限公司Control method and device of voice detection device

Similar Documents

PublicationPublication DateTitle
CN106612367A (en)Speech wake method based on microphone and mobile terminal
TWI474317B (en)Signal processing apparatus and signal processing method
KR102362107B1 (en) Low Power Ambient Computing System with Machine Learning Capabilities
CN107360327B (en)Speech recognition method, apparatus and storage medium
CN110223691A (en)Voice wakes up the method for handover control and device of identification
CN107277672B (en)Method and device for supporting automatic switching of wake-up mode
CN103686962A (en)Low-power-consumption mobile terminal awakening method and device
CN103543814B (en) Signal processing device and signal processing method
CN108848264A (en)Microphone control method and device, storage medium and electronic equipment
CN106161755A (en)A kind of key word voice wakes up system and awakening method and mobile terminal up
CN105630161A (en)Android system-based gesture control ring and using method thereof
TW201843582A (en)Electronic device having wake on voice function and operating method thereof
CN114039398B (en)Control method and device of new energy camera equipment and storage medium
CN108810280A (en)Processing method, device, storage medium and the electronic equipment of voice collecting frequency
CN103901782A (en)Sound control method, electronic device and sound control apparatus
CN108509225B (en)Information processing method and electronic equipment
CN110473544A (en)A kind of low-power consumption voice awakening method and device
CN104050973A (en)Low power audio trigger via intermittent sampling
CN105892614A (en)Power consumption management and control method and system of intelligent terminal
CN103701981A (en)Method and device for implementing voice recognition function
CN111179924B (en)Method and system for optimizing awakening performance based on mode switching
CN113889109B (en)Voice wake-up mode adjusting method, storage medium and electronic equipment
CN110956968A (en) Voice wake-up and method, device and terminal device for triggering voice wake-up function
CN113096652A (en)Voice awakening method and chip system
Yurur et al.Adaptive sampling and duty cycling for smartphone accelerometer

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
RJ01Rejection of invention patent application after publication
RJ01Rejection of invention patent application after publication

Application publication date:20170503


[8]ページ先頭

©2009-2025 Movatter.jp