CN104820556A - Method and device for waking up voice assistant - Google Patents

Method and device for waking up voice assistant Download PDF

Info

Publication number
CN104820556A
CN104820556A CN201510227622.2A CN201510227622A CN104820556A CN 104820556 A CN104820556 A CN 104820556A CN 201510227622 A CN201510227622 A CN 201510227622A CN 104820556 A CN104820556 A CN 104820556A
Authority
CN
China
Prior art keywords
facial image
described
ambient sound
voice assistant
whether
Prior art date
Application number
CN201510227622.2A
Other languages
Chinese (zh)
Inventor
张绍儒
Original Assignee
广州视源电子科技股份有限公司
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 广州视源电子科技股份有限公司 filed Critical 广州视源电子科技股份有限公司
Priority to CN201510227622.2A priority Critical patent/CN104820556A/en
Publication of CN104820556A publication Critical patent/CN104820556A/en

Links

Abstract

The invention relates to a method and a device for waking up a voice assistant. The method comprises the following steps that: a human face image and corresponding environment sound are obtained; whether the human face image meets a first preset condition or not is detected; if the human face image meets the first preset condition, whether the environment sound meets a second preset condition or not is detected; and if the environment sound meets the second preset condition, the voice assistant is waken up, and the environment sound is used as a voice operation instruction to be input into the voice assistant. The method and the device have the advantages that when the human face image and the environment sound meet the corresponding preset conditions, the voice assistant is automatically waken up; meanwhile, the environment sound is used as the voice operation instruction to be input into the voice assistant, i.e., a user can directly issue an operation instruction, so that the voice triggering redundancy step is avoided; the communication simplification is realized; and the electricity of a device provided with the voice assistant is saved.

Description

Wake method and the device of voice assistant up

Technical field

The present invention relates to communication technical field, particularly relate to and a kind ofly wake the method for voice assistant up, wake the device of voice assistant up.

Background technology

Speech recognition technology occurs from the 1950's, just start development slower, a small amount of isolated vocabulary can only be identified, until the nineties, this technology starts to achieve larger breakthrough in application and commercialization, become the emphasis of technical research, wherein apply the Siri (apple intelligent sound assistant) having Apple more widely, domestic University of Science and Technology news fly, Baidu's voice, search dog voice assistant etc.

In existing each voice assistant, generally need to input specific voice and complete triggering, thus make voice assistant be in voice state to be entered.Such as when connecing power supply, as long as say Hey Siri facing to ios (Mobile operating system of the Apple) equipment of band Siri, Siri service will be waken up.Speech trigger, as the breakdown action of phonetic entry, does not need contact arrangement, solves the problem of carrying out phonetic entry in certain circumstances well.

But, existing voice assistant needs special sound to wake up in use, and long-time opening voice arousal function can consume more electricity, and user needs first to wake voice assistant up by special sound, and then inputting corresponding voice operating instruction to voice assistant, process is redundancy comparatively.

Summary of the invention

Based on this, be necessary for the problems referred to above, a kind of simple to operate method waking voice assistant up and device are provided.

Wake a method for voice assistant up, comprise step:

Obtain the ambient sound of facial image and correspondence;

Whether detect described facial image meets first pre-conditioned;

If described facial image meets first pre-conditioned, detect described ambient sound and whether meet second pre-conditioned;

If described ambient sound meets second pre-conditioned, then wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.

Wake a device for voice assistant up, comprising:

Facial image acquisition module, for obtaining facial image;

Ambient sound acquisition module, for obtaining the ambient sound corresponding with facial image;

Whether facial image detection module, meet first pre-conditioned for detecting described facial image;

Ambient sound detection module, for meet at described facial image first pre-conditioned time, detect described ambient sound and whether meet second pre-conditioned;

Wake module, for described ambient sound meet second pre-conditioned time, wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.

The present invention wakes method and the device of voice assistant up, facial image and ambient sound meet corresponding pre-conditioned time, automatically voice assistant is waken up, and described ambient sound is inputted voice assistant as voice operating instruction simultaneously, that is to say that user directly can issue operational order, eliminate the redundancy step of speech trigger, achieve the simplification of communication, and save the electricity of the equipment being provided with voice assistant.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of the inventive method embodiment;

Fig. 2 is the schematic flow sheet of step S120 specific embodiment of the present invention;

Fig. 3 is the schematic flow sheet of step S130 specific embodiment of the present invention;

Fig. 4 is the structural representation of apparatus of the present invention embodiment;

Fig. 5 is the structural representation of face image detection module embodiment of the present invention;

Fig. 6 is the structural representation of the present invention first judging unit embodiment;

Fig. 7 is the structural representation of ambient sound detection module embodiment of the present invention.

Embodiment

The embodiment below in conjunction with accompanying drawing, the present invention being waken up to the method for voice assistant is described in detail.

As shown in Figure 1, a kind of method waking voice assistant up, comprises step:

The ambient sound of S110, acquisition facial image and correspondence;

S120, detect described facial image and whether meet first pre-conditioned;

If the described facial image of S130 meets first pre-conditioned, detect described ambient sound and whether meet second pre-conditioned;

If the described ambient sound of S140 meets second pre-conditioned, then wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.

Facial image can obtain according to camera, and ambient sound can according to acquisitions such as microphones.The camera image obtained and ambient sound need store, so that follow-up condition detects.In order to save storage space, can be set, so that after single job completes, useless data are removed in time, vacate storage space the reservation storage time of facial image and ambient sound.

Whether, after obtaining facial image and ambient sound, first can detect facial image and meet first pre-conditioned, wherein first pre-conditionedly can need to be set to various concrete form according to user.Such as, as shown in Figure 2, step S120 can comprise step:

S1201, judge whether facial image is front face image, if so, enters step S1202, otherwise returns step S110;

Whether S1202, the mouth judged in facial image have action, if there is action, enter step S1203, otherwise return step S110;

S1203, to judge that described facial image meets first pre-conditioned.

User is when using voice assistant, and general front, in the face of smart mobile phone or flat board etc., then sends phonetic order by mouth, that is to say that the facial image that camera captures is front face image, and when detecting that mouth has an action, empirically, the general possibility of voice assistant that uses is larger.Judge that whether facial image is that the mode of front face image has a variety of, such as, step S1201 comprises step:

Obtain the distance of eyes in facial image;

Judge that the distance of eyes is whether in preset range;

If so, then judge that facial image is front face image, otherwise judge that facial image is not front face image.

During owing to using voice assistant in reality, the impossible forward camera completely of user, so allow certain inclined value when the present invention detects, namely the present invention is not restricted to and determines that facial image is front face image, also can judge the whether approximate front face image of facial image.Determine that facial image is that front face image is also not limited to the above-mentioned method provided simultaneously, can also realize according to alternate manner in prior art.

Mouth is the main sound source of people, when needs send phonetic order to voice assistant, and can along with the action of mouth.So after judgement facial image is front face image, adds the judgement of mouth action feature, can accuracy be improved.Facial image according to obtaining judges whether mouth has action can realize according to mode existing in prior art.

When the front face image when facial image being detected and mouth have an action, namely can the detecting step of entered environment sound.Certain user can add other face verification condition as required, thus improves accuracy further, and the present invention will not describe in detail at this.

As shown in Figure 3, step S130 can comprise step:

S1301, judge that the volume of described ambient sound is whether in preset range;

S1302, to judge and whether the distance of sound source of described ambient sound is less than predetermined threshold value;

If the described volume of S1303 is in preset range and be less than predetermined threshold value with the distance of described sound source, then judges that described ambient sound meets second pre-conditioned, otherwise return step S110.

When general user uses voice assistant, the volume of sound is not too large or too little, and the distance of distance equipment is not too large, so present invention adds the judgement of volume and distance.Volume can obtain according to volume detector of the prior art etc., and the distance of equipment distance sound source can be determined according to the aerial decay formula qualitatively of sound.After environmentally sound obtains volume and distance, namely volume can be judged whether in preset range, whether distance is less than predetermined threshold value, thus it is second pre-conditioned to determine whether ambient sound meets, and wherein preset range and predetermined threshold value all can rule of thumb sets itself.

Ambient sound meet second pre-conditioned time, determine that user needs to carry out phonetic entry to smart machine, wake voice assistant up, and ambient sound is inputted voice assistant as voice operating instruction simultaneously, voice assistant directly performs corresponding operation, eliminate the step that special sound wakes voice assistant up, user operation is simple.

It should be noted that, the present invention is not limited the detection ordering of facial image and ambient sound, in addition, user can carry out the investigation of other session features according to actual conditions, such as detect that user not thinks when groaning song or talking to onself to meet second pre-conditioned etc., the present invention is not limited decision condition.

Implementation process for a better understanding of the present invention, is described below in conjunction with an embody rule scene.

When cooking in kitchen, we want to carry out menu displaying by panel computer, thus complete oneself unfamiliar cuisine together according to step above.After finishing one operation, we wish that the software in panel computer can show next process, at this time just to input instruction to it, but at this time likely both hands all busy or be stained with oil stain and be inconvenient to touch panel computer, at this moment we only need face to turn to panel computer to say " next step ", then voice provided by the invention wake software up and judge that we are in the voice assistant input instruction to panel computer, instead of others is spoken, so just the wake-up step of voice assistant is directly completed, and voice assistant can immediately by speech analysis out, and make the displaying of menu software translate into next step.

Based on same inventive concept, the present invention also provides a kind of device waking voice assistant up, is described in detail below in conjunction with the embodiment of accompanying drawing to apparatus of the present invention.

As shown in Figure 4, a kind of device waking voice assistant up, comprising:

Facial image acquisition module 410, for obtaining facial image;

Ambient sound acquisition module 420, for obtaining the ambient sound corresponding with facial image;

Whether facial image detection module 430, meet first pre-conditioned for detecting described facial image;

Ambient sound detection module 440, for meet at described facial image first pre-conditioned time, detect described ambient sound and whether meet second pre-conditioned;

Wake module 450, for described ambient sound meet second pre-conditioned time, wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.

Facial image acquisition module 410 can obtain facial image according to camera, and ambient sound acquisition module 420 can obtain ambient sound according to microphone etc.The camera image that facial image acquisition module 410 and ambient sound acquisition module 420 obtain and ambient sound need store, so that follow-up condition detects.In order to save storage space, can be set, so that after single job completes, useless data are removed in time, vacate storage space the reservation storage time of facial image and ambient sound.

After obtaining facial image and ambient sound, whether facial image detection module 430 detects facial image and meets first pre-conditioned, and wherein first pre-conditionedly can need to be set to various concrete form according to user.Such as, as shown in Figure 5, described facial image detection module 430 can comprise:

First judging unit 4301, for judging whether facial image is front face image;

Second judging unit 4302, during for the front face image when facial image, judges whether the mouth in facial image has action;

Identifying unit 4303, for when mouth has action, judges that described facial image meets first pre-conditioned.

First judging unit 4301 judges that whether facial image is that the mode of front face image has a variety of, and such as, as shown in Figure 6, described first judging unit 4301 can comprise:

Eyes distance acquiring unit 43011, for obtaining the distance of eyes in facial image;

Eyes Distance Judgment unit 43012, for judging that the distance of eyes is whether in preset range;

Facial image identifying unit 43013, for when the distance of eyes is in preset range, judges that facial image is front face image, otherwise judges that facial image is not front face image.

First judging unit 4301 is not restricted to determines that facial image is front face image, also can judge the whether approximate front face image of facial image.First judging unit 4301 determines that facial image is the mode that front face image is also not limited to as shown in Figure 6 simultaneously, can also realize according to alternate manner in prior art.After judgement facial image is front face image, second judging unit 4302 adds the judgement of mouth action feature, can improve accuracy, wherein according to the facial image obtained, the second judging unit 4302 judges whether mouth has action can realize according to mode existing in prior art.

Ambient sound detection module 440 facial image meet first pre-conditioned time, whether testing environment sound meets second pre-conditioned.It should be noted that, the present invention is not limited the execution sequence of ambient sound detection module 440 and facial image detection module 430.The implementation of ambient sound detection module 440 has a variety of, and such as, as shown in Figure 7, described ambient sound detection module 440 can comprise:

Volume judging unit 4401, for judging that the volume of described ambient sound is whether in preset range, wherein volume can obtain according to volume detector of the prior art etc.;

Distance Judgment unit 4402, for judging whether be less than predetermined threshold value with the distance of the sound source of described ambient sound, wherein Distance Judgment unit 4402 can determine the distance with sound source according to the aerial decay formula qualitatively of sound;

Sound identifying unit 4403, for in described volume in preset range and when being less than predetermined threshold value with the distance of described sound source, judge that described ambient sound meets second pre-conditioned, otherwise the camera image that facial image acquisition module 410 and ambient sound acquisition module 420 obtain again and ambient sound, wherein preset range and predetermined threshold value all can rule of thumb sets itself.

Wake module 450 ambient sound meet second pre-conditioned time, determine that user needs to carry out phonetic entry to smart machine, wake voice assistant up, and ambient sound is inputted voice assistant as voice operating instruction simultaneously, voice assistant directly performs corresponding operation, eliminate the step that special sound wakes voice assistant up, user operation is simple.

Each technical characteristic of the above embodiment can combine arbitrarily, for making description succinct, the all possible combination of each technical characteristic in above-described embodiment is not all described, but, as long as the combination of these technical characteristics does not exist contradiction, be all considered to be the scope that this instructions is recorded.

The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be construed as limiting the scope of the patent.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.

Claims (10)

1. wake a method for voice assistant up, it is characterized in that, comprise step:
Obtain the ambient sound of facial image and correspondence;
Whether detect described facial image meets first pre-conditioned;
If described facial image meets first pre-conditioned, detect described ambient sound and whether meet second pre-conditioned;
If described ambient sound meets second pre-conditioned, then wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.
2. the method waking voice assistant up according to claim 1, is characterized in that, detects described facial image and whether meets the first pre-conditioned step and comprise:
Judge whether facial image is front face image;
If facial image is front face image, judge whether the mouth in facial image has action, if facial image is not front face image, then return the step of the ambient sound obtaining facial image and correspondence;
If there is action, then judges that described facial image meets first pre-conditioned, otherwise return the step of the ambient sound obtaining facial image and correspondence.
3. the method waking voice assistant up according to claim 2, is characterized in that, judges that whether facial image is that the step of front face image comprises:
Obtain the distance of eyes in facial image;
Judge that the distance of eyes is whether in preset range;
If so, then judge that facial image is front face image, otherwise judge that facial image is not front face image.
4. the method waking voice assistant up according to claim 1, is characterized in that, detects described ambient sound and whether meets the second pre-conditioned step and comprise:
Judge that the volume of described ambient sound is whether in preset range;
Judge whether be less than predetermined threshold value with the distance of the sound source of described ambient sound;
If described volume is in preset range and be less than predetermined threshold value with the distance of described sound source, then judges that described ambient sound meets second pre-conditioned, otherwise return the step of the ambient sound obtaining facial image and correspondence.
5. the method waking voice assistant up according to claim 4, is characterized in that, determines according to the aerial decay formula of sound with the distance of described sound source.
6. wake a device for voice assistant up, it is characterized in that, comprising:
Facial image acquisition module, for obtaining facial image;
Ambient sound acquisition module, for obtaining the ambient sound corresponding with facial image;
Whether facial image detection module, meet first pre-conditioned for detecting described facial image;
Ambient sound detection module, for meet at described facial image first pre-conditioned time, detect described ambient sound and whether meet second pre-conditioned;
Wake module, for described ambient sound meet second pre-conditioned time, wake voice assistant up, and using described ambient sound as voice operating instruction input voice assistant.
7. the device waking voice assistant up according to claim 6, is characterized in that, described facial image detection module comprises:
First judging unit, for judging whether facial image is front face image;
Second judging unit, for when facial image is front face image, judges whether the mouth in facial image has action;
Identifying unit, for when mouth has action, judges that described facial image meets first pre-conditioned.
8. the device waking voice assistant up according to claim 7, is characterized in that, described first judging unit comprises:
Eyes distance acquiring unit, for obtaining the distance of eyes in facial image;
Eyes Distance Judgment unit, for judging that the distance of eyes is whether in preset range;
Facial image identifying unit, for when the distance of eyes is in preset range, judges that facial image is front face image, otherwise judges that facial image is not front face image.
9. the device waking voice assistant up according to claim 6, is characterized in that, described ambient sound detection module comprises:
Volume judging unit, for judging that the volume of described ambient sound is whether in preset range;
Distance Judgment unit, for judging whether be less than predetermined threshold value with the distance of the sound source of described ambient sound;
Sound identifying unit, in described volume in preset range and when being less than predetermined threshold value with the distance of described sound source, judges that described ambient sound meets second pre-conditioned.
10. the device waking voice assistant up according to claim 9, is characterized in that, described Distance Judgment unit determines the distance with described sound source according to the aerial decay formula of sound.
CN201510227622.2A 2015-05-06 2015-05-06 Method and device for waking up voice assistant CN104820556A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510227622.2A CN104820556A (en) 2015-05-06 2015-05-06 Method and device for waking up voice assistant

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510227622.2A CN104820556A (en) 2015-05-06 2015-05-06 Method and device for waking up voice assistant

Publications (1)

Publication Number Publication Date
CN104820556A true CN104820556A (en) 2015-08-05

Family

ID=53730864

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510227622.2A CN104820556A (en) 2015-05-06 2015-05-06 Method and device for waking up voice assistant

Country Status (1)

Country Link
CN (1) CN104820556A (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105204628A (en) * 2015-09-01 2015-12-30 涂悦 Voice control method based on visual awakening
CN105700363A (en) * 2016-01-19 2016-06-22 深圳创维-Rgb电子有限公司 Method and system for waking up smart home equipment voice control device
CN105912092A (en) * 2016-04-06 2016-08-31 北京地平线机器人技术研发有限公司 Voice waking up method and voice recognition device in man-machine interaction
WO2017035768A1 (en) * 2015-09-01 2017-03-09 涂悦 Voice control method based on visual wake-up
CN106782524A (en) * 2016-11-30 2017-05-31 深圳讯飞互动电子有限公司 One kind mixing awakening method and system
CN106847285A (en) * 2017-03-31 2017-06-13 上海思依暄机器人科技股份有限公司 A kind of robot and its audio recognition method
CN107315561A (en) * 2017-06-30 2017-11-03 联想(北京)有限公司 A kind of data processing method and electronic equipment
CN107517313A (en) * 2017-08-22 2017-12-26 珠海市魅族科技有限公司 Awakening method and device, terminal and readable storage medium storing program for executing
CN107678793A (en) * 2017-09-14 2018-02-09 珠海市魅族科技有限公司 Voice assistant starts method and device, terminal and computer-readable recording medium
CN108055617A (en) * 2017-12-12 2018-05-18 广东小天才科技有限公司 A kind of awakening method of microphone, device, terminal device and storage medium
CN108098767A (en) * 2016-11-25 2018-06-01 北京智能管家科技有限公司 A kind of robot awakening method and device
CN108154140A (en) * 2018-01-22 2018-06-12 北京百度网讯科技有限公司 Voice awakening method, device, equipment and computer-readable medium based on lip reading
CN108363557A (en) * 2018-02-02 2018-08-03 刘国华 Man-machine interaction method, device, computer equipment and storage medium

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000347692A (en) * 1999-06-07 2000-12-15 Sanyo Electric Co Ltd Person detecting method, person detecting device, and control system using it
EP1215658A2 (en) * 2000-12-05 2002-06-19 Hewlett-Packard Company Visual activation of voice controlled apparatus
CN102298443A (en) * 2011-06-24 2011-12-28 华南理工大学 Intelligent home control system combines voice and video channel control method
CN102945672A (en) * 2012-09-29 2013-02-27 深圳市国华识别科技开发有限公司 Voice control system for multimedia equipment, and voice control method
CN103472994A (en) * 2013-09-06 2013-12-25 乐得科技有限公司 Operation control achieving method, device and system based on voice
US20140222436A1 (en) * 2013-02-07 2014-08-07 Apple Inc. Voice trigger for a digital assistant
CN104078041A (en) * 2014-06-26 2014-10-01 美的集团股份有限公司 Voice recognition method and system
CN104103274A (en) * 2013-04-11 2014-10-15 纬创资通股份有限公司 Apparatus and method for voice processing
CN104428832A (en) * 2012-07-09 2015-03-18 Lg电子株式会社 Speech recognition apparatus and method
CN104423992A (en) * 2013-09-03 2015-03-18 冠捷投资有限公司 Speech recognition startup method for display

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000347692A (en) * 1999-06-07 2000-12-15 Sanyo Electric Co Ltd Person detecting method, person detecting device, and control system using it
EP1215658A2 (en) * 2000-12-05 2002-06-19 Hewlett-Packard Company Visual activation of voice controlled apparatus
US6970824B2 (en) * 2000-12-05 2005-11-29 Hewlett-Packard Development Company, L.P. Enabling voice control of voice-controlled apparatus using a head mounted camera system
CN102298443A (en) * 2011-06-24 2011-12-28 华南理工大学 Intelligent home control system combines voice and video channel control method
CN104428832A (en) * 2012-07-09 2015-03-18 Lg电子株式会社 Speech recognition apparatus and method
CN102945672A (en) * 2012-09-29 2013-02-27 深圳市国华识别科技开发有限公司 Voice control system for multimedia equipment, and voice control method
US20140222436A1 (en) * 2013-02-07 2014-08-07 Apple Inc. Voice trigger for a digital assistant
CN104103274A (en) * 2013-04-11 2014-10-15 纬创资通股份有限公司 Apparatus and method for voice processing
CN104423992A (en) * 2013-09-03 2015-03-18 冠捷投资有限公司 Speech recognition startup method for display
CN103472994A (en) * 2013-09-06 2013-12-25 乐得科技有限公司 Operation control achieving method, device and system based on voice
CN104078041A (en) * 2014-06-26 2014-10-01 美的集团股份有限公司 Voice recognition method and system

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105204628A (en) * 2015-09-01 2015-12-30 涂悦 Voice control method based on visual awakening
WO2017035768A1 (en) * 2015-09-01 2017-03-09 涂悦 Voice control method based on visual wake-up
CN105700363B (en) * 2016-01-19 2018-10-26 深圳创维-Rgb电子有限公司 A kind of awakening method and system of smart home device phonetic controller
CN105700363A (en) * 2016-01-19 2016-06-22 深圳创维-Rgb电子有限公司 Method and system for waking up smart home equipment voice control device
CN105912092A (en) * 2016-04-06 2016-08-31 北京地平线机器人技术研发有限公司 Voice waking up method and voice recognition device in man-machine interaction
CN108098767A (en) * 2016-11-25 2018-06-01 北京智能管家科技有限公司 A kind of robot awakening method and device
CN106782524A (en) * 2016-11-30 2017-05-31 深圳讯飞互动电子有限公司 One kind mixing awakening method and system
CN106847285A (en) * 2017-03-31 2017-06-13 上海思依暄机器人科技股份有限公司 A kind of robot and its audio recognition method
CN107315561A (en) * 2017-06-30 2017-11-03 联想(北京)有限公司 A kind of data processing method and electronic equipment
CN107517313A (en) * 2017-08-22 2017-12-26 珠海市魅族科技有限公司 Awakening method and device, terminal and readable storage medium storing program for executing
CN107678793A (en) * 2017-09-14 2018-02-09 珠海市魅族科技有限公司 Voice assistant starts method and device, terminal and computer-readable recording medium
CN108055617A (en) * 2017-12-12 2018-05-18 广东小天才科技有限公司 A kind of awakening method of microphone, device, terminal device and storage medium
CN108154140A (en) * 2018-01-22 2018-06-12 北京百度网讯科技有限公司 Voice awakening method, device, equipment and computer-readable medium based on lip reading
CN108363557A (en) * 2018-02-02 2018-08-03 刘国华 Man-machine interaction method, device, computer equipment and storage medium
WO2019149160A1 (en) * 2018-02-02 2019-08-08 刘国华 Human-machine interaction method and device, computer apparatus, and storage medium

Similar Documents

Publication Publication Date Title
US9117449B2 (en) Embedded system for construction of small footprint speech recognition with user-definable constraints
AU2013289660B2 (en) Method for updating voiceprint feature model and terminal
KR101726945B1 (en) Reducing the need for manual start/end-pointing and trigger phrases
EP2930716B1 (en) Speech recognition using electronic device and server
US20160019886A1 (en) Method and apparatus for recognizing whisper
EP2669889A2 (en) Method and apparatus for executing voice command in electronic device
US20140244253A1 (en) Systems and Methods for Continual Speech Recognition and Detection in Mobile Computing Devices
BR102014003975A2 (en) speech keyword detection system and associated method
CN105138171B (en) Touch key-press and fingerprint identification method
JP6311194B2 (en) Contact grouping method and apparatus
CN103049209A (en) Method and device for starting phone camera in screen extinguishing state of mobile phone
CN104951159A (en) Touch key and fingerprint identification method
RU2643460C2 (en) Touch keys and method for implementing fingerprint recognition, apparatus and terminal device
CN104090652B (en) A kind of pronunciation inputting method and device
US9652017B2 (en) System and method of analyzing audio data samples associated with speech recognition
CN103795850B (en) Electronic equipment and audio recognition method
WO2015127825A1 (en) Expression input method and apparatus and electronic device
KR20160001964A (en) Operating Method For Microphones and Electronic Device supporting the same
CN102594999A (en) Method and system for performing adaptive mobile phone energy conservation through face identification
CN105956518A (en) Face identification method, device and system
KR20150079829A (en) Gesture-based conversation processing method, apparatus, and terminal device
CN105632491A (en) Method and electronic device for voice recognition
CN103391366B (en) Mobile terminal and identification method thereof
WO2017206916A1 (en) Method for determining kernel running configuration in processor and related product
WO2015078151A1 (en) Method and apparatus for image processing and terminal device

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
EXSB Decision made by sipo to initiate substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20150805

RJ01 Rejection of invention patent application after publication