CN105847584A - Method for intelligent device to identify private conversations - Google Patents
Method for intelligent device to identify private conversations Download PDFInfo
- Publication number
- CN105847584A CN105847584A CN201610318560.0A CN201610318560A CN105847584A CN 105847584 A CN105847584 A CN 105847584A CN 201610318560 A CN201610318560 A CN 201610318560A CN 105847584 A CN105847584 A CN 105847584A
- Authority
- CN
- China
- Prior art keywords
- signal
- secret words
- words
- secret
- obtains
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
- H04M1/72454—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions according to context-related or environment-related conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/72—Mobile telephones; Cordless telephones, i.e. devices for establishing wireless links to base stations without route selection
- H04M1/724—User interfaces specially adapted for cordless or mobile telephones
- H04M1/72448—User interfaces specially adapted for cordless or mobile telephones with means for adapting the functionality of the device according to specific conditions
Abstract
The invention discloses a method for an intelligent device to identify private conversations. The method comprises the following steps: a central control module obtains private conversation start signals, and whether the obtained signals are private conversation mode start signals is determined; if yes, the central control module turns on a unidirectivity microphone, and a conversation device in a routine conversation mode is turned off; the central control module obtains private conversation closing signals, and whether the obtained signals are private conversation mode closing signals is determined; if yes, the central control module turns off the unidirectivity microphone, and the conversation device in the routine conversation mode is turned on. The method for the intelligent device to identify the private conversations enables the intelligent device to start and close a private conversation mode of the intelligent device in a reliable manner according to the private conversation start signals and the private conversation closing signals that are collected, and mistake start-up of the private conversation mode can be prevented.
Description
Technical field
The present invention relates to smart machine function technical field, particularly relate to a kind of smart machine identification secret words
Method.
Background technology
During some smart machine exchanges and dialogues, having following two situation needs smart machine startup quiet
Quiet words pattern.
The first, the when that people being many most around, will seek advice from some problems comparing secret to robot,
Or child asks that some compare the problem of child's simplicity, but in the case of having people too embarrassed to say for fear of surrounding;
It two is, before child falls asleep, can there is a need to the gentle talk of sound little with robot or listen machine
People tells a story, in the case of needing to compare the sight of secret.
Summary of the invention
The present invention provides a kind of method of smart machine identification secret words, and the method makes smart machine according to collection
The signal that the secret words enabling signal arrived and secret words are closed, starts reliably and closes smart machine silently
Words pattern.
For solving above-mentioned technical problem, the technical scheme is that
A kind of method of smart machine identification secret words comprises the following steps: central control module obtains secret words
The signal started;Judge whether the signal obtained is secret words mode starting signal;If signal is secret words mould
Formula enabling signal, central control module starts single directive microphone, closes the call under Regular conversation pattern
Equipment;Central control module obtains the signal that secret words are closed;Judge whether the signal obtained is secret words mould
Formula shutdown signal;If signal is secret words pattern shutdown signals, central control module closes unidirectivity Mike
Wind, starts the verbal system under described Regular conversation pattern.
Optimal way is, the signal that the signal of described secret words startup or secret words are closed is: optical pickocff
The speaker's hand-type obtained and the hand-type image identification signal corresponding with face position relationship, and Distance-sensing
Speaker's voltage signal corresponding with the spacing of smart machine that device obtains or digital signal.
Optimal way is, described secret words mode starting signal is: described optical pickocff obtains hands and is placed on mouth
Speak image identification signal corresponding to shape in bar side, and the distance that described range sensor obtains is less than
Voltage signal that 10cm is corresponding or digital signal.
Optimal way is, described secret words pattern shutdown signal is: the hand-type that described optical pickocff obtains is
Hands is not placed on face side and speaks image identification signal corresponding to shape, and/or described range sensor obtains
The distance voltage signal more than or equal to 10cm or digital signal.
Optimal way is, the signal that the signal of described secret words startup or secret words are closed is: optical pickocff
The speaker's hand-type obtained and the hand-type image identification signal corresponding with face position relationship, and voice module
Voltage signal corresponding to voice command obtained.
Optimal way is, described secret words mode starting signal is: the hand-type that described optical pickocff obtains is
It is placed on face side to speak image identification signal corresponding to shape, and the startup that described voice module obtains is quiet
The voltage signal that quiet words mode speech order is corresponding.
Optimal way is, described secret words pattern shutdown signal is: the hand-type that described optical pickocff obtains is
It is not placed on face side to speak image identification signal corresponding to shape, and/or the closedown that described voice module obtains
The voltage signal that secret words mode speech order is corresponding.
Optimal way is, the signal that the signal of described secret words startup or secret words are closed is: range sensor
The speaker's voltage signal corresponding with the spacing of smart machine obtained or digital signal, and voice module
Voltage signal corresponding to voice command obtained.
Optimal way is, described secret words mode starting signal is: the distance that described range sensor obtains is little
In voltage signal corresponding for 10cm or digital signal, and the startup secret words pattern that described voice module obtains
The voltage signal that voice command is corresponding.
Optimal way is, described secret words pattern shutdown signal is: the distance that described range sensor obtains is big
In or equal to voltage signal corresponding for 10cm or digital signal, and/or the closedown that described voice module obtains is quiet
The voltage signal that quiet words mode speech is corresponding.
After using technique scheme, the invention has the beneficial effects as follows: due to the smart machine identification of the present invention
The method of secret words comprises the following steps: central control module obtains the signal that secret words start;Judge to obtain
Signal whether be secret words mode starting signal;If signal is secret words mode starting signal, central authorities control
Module starts single directive microphone, closes the verbal system under Regular conversation pattern;Central control module obtains
Take the signal that secret words are closed;Judge whether the signal obtained is secret words pattern shutdown signals;If signal is
Secret words pattern shutdown signal, central control module is closed single directive microphone, is started described Regular conversation
Verbal system under pattern.The method of the smart machine identification secret words of the present invention, makes smart machine according to adopting
The secret words enabling signal that collection arrives, starts and closes the secret words pattern of smart machine reliably.
Accompanying drawing explanation
Fig. 1 is the structured flowchart of the method for smart machine identification secret words of the present invention;
Fig. 2 is the flow chart of embodiment one;
Fig. 3 is the flow chart of embodiment two;
Fig. 4 is the flow chart of embodiment three.
Detailed description of the invention
In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with accompanying drawing and reality
Execute example, the present invention is further elaborated.Only should be appreciated that specific embodiment described herein
Only in order to explain the present invention, it is not intended to limit the present invention.
Embodiment one:
As depicted in figs. 1 and 2, a kind of method of smart machine identification secret words, comprise the following steps:
In the step s 100, central control module obtains the signal that secret words start.Wherein secret words start
The signal that signal is sensed by optical pickocff and range sensor, the preferred version of the present embodiment is in intelligence
Two range sensors are set on equipment, certainly with multiple range sensor can be set, be arranged on intelligence and set
Standby diverse location, makes the distance detection between people and the smart machine of large-scale intelligent equipment more accurate.
In step S101, it is judged that whether the signal of acquisition is secret words mode starting signal.Wherein secret words
Mode starting signal refers to: speaker's hand-type and hand-type that optical pickocff obtains are corresponding with face position relationship
Image identification signal, and speaker's electricity corresponding with the spacing of smart machine that range sensor obtains
Pressure signal or digital signal.
In step s 102, if signal is secret words mode starting signal, central control module starts single sensing
Property mike, close the verbal system under Regular conversation pattern.The figure that in this step, optical pickocff will obtain
Shape signal is transferred to central control module, central control module process image identification signal and compare,
The voltage signal obtained or digital data transmission to central control module, are controlled by same range sensor by central authorities
Voltage signal is processed and compares by molding block.When central control module relatively after it is concluded that be: hands is put
The image identification signal that shape of speaking on face side is corresponding, the distance between people and intelligent set is less than simultaneously
10cm, central control module makes smart machine cut secret words pattern, simultaneously closes off the logical of Regular conversation pattern
Words equipment, reduces the volume of loudspeaker, switches to the sound of a kind of tenderness.
In step s 103, central control module obtains the signal that secret words are closed, with above-mentioned steps S100
In identical.
In step S104, it is judged that whether the signal of acquisition is secret words pattern shutdown signals, with above-mentioned steps
S101 is essentially identical.
In step S105, if signal is secret words pattern shutdown signals, central control module closes single sensing
Property mike, start the verbal system under Regular conversation pattern.Essentially identical with above-mentioned steps S102, different
Part be: when central control module relatively after it is concluded that be: optical pickocff acquisition hand-type be that hands is not put
The image identification signal that shape of speaking on face side is corresponding, and/or range sensor obtain distance more than or
When voltage signal equal to 10cm or digital signal, central control module single directive microphone, wherein conventional
Verbal system under call mode.
From the foregoing, the method for the smart machine identification secret words of the present invention, make smart machine according to collection
The secret words enabling signal arrived, starts and closes the secret words pattern of smart machine reliably, will not error starting
Secret words pattern.
Embodiment two:
As shown in figures 1 and 3, the present embodiment is essentially identical with embodiment one, and difference is:
In the step s 100, central control module obtains the signal that secret words start.Wherein secret words start letter
Number for optical pickocff obtain induced signal and voice module obtain voice command.
In step S101, it is judged that whether the signal of acquisition is secret words mode starting signal.Wherein secret words
Mode starting signal refers to: speaker's hand-type and hand-type that optical pickocff obtains are corresponding with face position relationship
Image identification signal, and voltage signal corresponding to voice command that voice module obtains.
In step s 102, if signal is secret words mode starting signal, central control module starts single sensing
Property mike, close the verbal system under Regular conversation pattern.The figure that in this step, optical pickocff will obtain
As identifying that signal is transferred to central control module, central control module telecommunication signal processed and compare,
The voltage signal of acquisition is transferred to central control module, by central control module to voltage by same voice module
Signal processes and compares.When central control module relatively after it is concluded that be: optical pickocff obtain
Hand-type is that hands is placed on face side and speaks image identification signal corresponding to shape, and opening of obtaining of voice module
During voltage signal corresponding to dynamic secret words mode speech order, central control module starts single directive microphone,
Close the verbal system under Regular conversation pattern.
In step s 103, central control module obtains the signal that secret words are closed.With the step in the present embodiment
Rapid S100 is essentially identical.
In step S104, it is judged that whether the signal of acquisition is secret words pattern shutdown signals, with the present embodiment
In step S101 essentially identical.
In step S105, if signal is secret words pattern shutdown signals, central control module closes single sensing
Property mike, start the verbal system under Regular conversation pattern.Essentially identical with the present embodiment step S102.
Difference is: when central control module relatively after it is concluded that be: optical pickocff obtain hand-type be hands
It is not placed on face side to speak image identification signal corresponding to shape, and/or the closedown that voice module obtains is silently
When talking about voltage signal corresponding to mode speech order or digital signal, central control module makes smart machine enter
Regular conversation pattern.
Embodiment three:
As shown in Figure 1 and Figure 4, the present embodiment is essentially identical with embodiment one, and difference exists:
In the step s 100, central control module obtains the signal that secret words start.Wherein secret words start letter
Number for range sensor sensing induced signal and voice module obtain voice command.
In step S101, it is judged that whether the signal of acquisition is secret words mode starting signal.Wherein secret words
Mode starting signal refers to: the speaker of range sensor acquisition and the voltage signal of the spacing of smart machine
Or digital signal, and voltage signal corresponding to voice command that voice module obtains.
In step s 102, if signal is secret words mode starting signal, central control module starts single sensing
Property mike, close the verbal system under Regular conversation pattern.The electricity that in this step, range sensor will obtain
Telecommunication signal, to central control module, is processed by pressure signal or digital data transmission by central control module
With compare, the voltage signal of acquisition is transferred to central control module by same voice module, central authorities control mould
Voltage signal is processed and compares by block.When central control module relatively after it is concluded that be: speaker and
The spacing of smart machine is less than 10cm, when voice command is " opening secret words pattern " simultaneously, and central authorities' control
Molding block makes smart machine cut secret words pattern, simultaneously closes off the verbal system under Regular conversation pattern, fall
The volume of low loudspeaker, switches to the sound of a kind of tenderness.
In step s 103, central control module obtains the signal that secret words are closed.Step with the present embodiment
In S100 essentially identical.
In step S104, it is judged that whether the signal of acquisition is secret words pattern shutdown signals, with the present embodiment
In step S101 essentially identical.
In step S105, if signal is secret words pattern shutdown signals, central control module closes single sensing
Property mike, start the verbal system under Regular conversation pattern.It is essentially identical with the present embodiment step S102,
Difference is: when central control module relatively after it is concluded that be: range sensor obtain distance be more than
Or equal to voltage signal corresponding for 10cm or digital signal, and/or voice command is " closing secret words pattern "
Voltage signal time, central control module make smart machine enter Regular conversation pattern.
The preferred embodiment of present invention described above, not in order to limit the present invention, all the present invention's
Any amendment of being made within spirit and principle, etc. the method structure of same smart machine identification secret words
Improve, should be included within the scope of the present invention.
Claims (10)
1. the method for smart machine identification secret words, it is characterised in that comprise the following steps:
Central control module obtains the signal that secret words start;
Judge whether the signal obtained is secret words mode starting signal;
If signal is secret words mode starting signal, central control module starts single directive microphone, closes
Verbal system under Regular conversation pattern;
Central control module obtains the signal that secret words are closed;
Judge whether the signal obtained is secret words pattern shutdown signals;
If signal is secret words pattern shutdown signals, central control module closes single directive microphone, starts
Verbal system under described Regular conversation pattern.
The method of smart machine identification secret words the most according to claim 1, it is characterised in that described
The signal that the signal of secret words startup or secret words are closed is: speaker's hand-type of optical pickocff acquisition and hands
The image identification signal that type is corresponding with face position relationship, and range sensor obtain speaker and intelligence
Voltage signal that the spacing of equipment is corresponding or digital signal.
The method of smart machine identification secret words the most according to claim 2, it is characterised in that described
Secret words mode starting signal is: described optical pickocff obtains hands, and to be placed on face side shape of speaking corresponding
Image identification signal, and the distance that described range sensor obtains is less than voltage signal corresponding for 10cm or number
Word signal.
The method of smart machine identification secret words the most according to claim 2, it is characterised in that described
Secret words pattern shutdown signal is: the hand-type that described optical pickocff obtains is that hands is not placed on face side and says
The image identification signal that words shape is corresponding, and/or the distance that described range sensor obtains is more than or equal to 10cm
Voltage signal or digital signal.
The method of smart machine identification secret words the most according to claim 1, it is characterised in that described
The signal that the signal of secret words startup or secret words are closed is: speaker's hand-type of optical pickocff acquisition and hands
The image identification signal that type is corresponding with face position relationship, and the voice command that voice module obtains is corresponding
Voltage signal.
The method of smart machine identification secret words the most according to claim 5, it is characterised in that described
Secret words mode starting signal is: the hand-type that described optical pickocff obtains is that hands is placed on face side and speaks
The image identification signal that shape is corresponding, and the startup secret words mode speech order pair that described voice module obtains
The voltage signal answered.
The method of smart machine identification secret words the most according to claim 5, it is characterised in that described
Secret words pattern shutdown signal is: the hand-type that described optical pickocff obtains is that hands is not placed on face side and says
The image identification signal that words shape is corresponding, and/or the closedown secret words mode speech order that described voice module obtains
Corresponding voltage signal.
The method of smart machine identification secret words the most according to claim 1, it is characterised in that described
The signal that the signal of secret words startup or secret words are closed is: speaker and intelligence that range sensor obtains set
Voltage signal that standby spacing is corresponding or digital signal, and the voice command that voice module obtains is corresponding
Voltage signal.
The method of smart machine identification secret words the most according to claim 8, it is characterised in that described
Secret words mode starting signal is: the distance that described range sensor obtains is less than voltage signal corresponding for 10cm
Or digital signal, and described voice module obtain start secret words mode speech order corresponding voltage letter
Number.
The method of smart machine identification secret words the most according to claim 8, it is characterised in that institute
Stating secret words pattern shutdown signal is: the distance that described range sensor obtains is corresponding more than or equal to 10cm
Voltage signal or digital signal, and/or described voice module obtain close electricity corresponding to secret words mode speech
Pressure signal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610318560.0A CN105847584B (en) | 2016-05-12 | 2016-05-12 | A kind of method of smart machine identification secret words |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610318560.0A CN105847584B (en) | 2016-05-12 | 2016-05-12 | A kind of method of smart machine identification secret words |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105847584A true CN105847584A (en) | 2016-08-10 |
CN105847584B CN105847584B (en) | 2019-03-05 |
Family
ID=56591816
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610318560.0A Active CN105847584B (en) | 2016-05-12 | 2016-05-12 | A kind of method of smart machine identification secret words |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105847584B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020244355A1 (en) * | 2019-06-03 | 2020-12-10 | 清华大学 | Microphone signal-based voice interaction wake-up electronic device, method, and medium |
CN113450769A (en) * | 2020-03-09 | 2021-09-28 | 杭州海康威视数字技术股份有限公司 | Voice extraction method, device, equipment and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0462599A (en) * | 1990-07-02 | 1992-02-27 | N T T Data Tsushin Kk | Noise removing device |
JP2011257627A (en) * | 2010-06-10 | 2011-12-22 | Murata Mach Ltd | Voice recognition device and recognition method |
US8438023B1 (en) * | 2011-09-30 | 2013-05-07 | Google Inc. | Warning a user when voice input to a device is likely to fail because of background or other noise |
CN104038591A (en) * | 2014-06-27 | 2014-09-10 | 宇龙计算机通信科技(深圳)有限公司 | Recording method and terminal |
CN104301664A (en) * | 2013-07-19 | 2015-01-21 | 松下电器产业株式会社 | Directivity control system, directivity control method, sound collection system and sound collection control method |
-
2016
- 2016-05-12 CN CN201610318560.0A patent/CN105847584B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0462599A (en) * | 1990-07-02 | 1992-02-27 | N T T Data Tsushin Kk | Noise removing device |
JP2011257627A (en) * | 2010-06-10 | 2011-12-22 | Murata Mach Ltd | Voice recognition device and recognition method |
US8438023B1 (en) * | 2011-09-30 | 2013-05-07 | Google Inc. | Warning a user when voice input to a device is likely to fail because of background or other noise |
CN104301664A (en) * | 2013-07-19 | 2015-01-21 | 松下电器产业株式会社 | Directivity control system, directivity control method, sound collection system and sound collection control method |
CN104038591A (en) * | 2014-06-27 | 2014-09-10 | 宇龙计算机通信科技(深圳)有限公司 | Recording method and terminal |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020244355A1 (en) * | 2019-06-03 | 2020-12-10 | 清华大学 | Microphone signal-based voice interaction wake-up electronic device, method, and medium |
CN113450769A (en) * | 2020-03-09 | 2021-09-28 | 杭州海康威视数字技术股份有限公司 | Voice extraction method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN105847584B (en) | 2019-03-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101422020B1 (en) | Method for recognizing voice, and apparatus for implementing the same | |
CN107656461A (en) | A kind of method and washing machine based on age of user regulation voice | |
CN110364156A (en) | Voice interactive method, system, terminal and readable storage medium storing program for executing | |
CN109272991B (en) | Voice interaction method, device, equipment and computer-readable storage medium | |
CN109688276A (en) | A kind of incoming call filter system and method based on artificial intelligence technology | |
CN105100484B (en) | A kind of methods, devices and systems for terminating voice call | |
CN109036393A (en) | Wake-up word training method, device and the household appliance of household appliance | |
JP2014191029A (en) | Voice recognition system and method for controlling voice recognition system | |
KR20190049260A (en) | Device and method for recognizing voice of vehicle | |
CN109308900B (en) | Earphone device, voice processing system and voice processing method | |
US20180158462A1 (en) | Speaker identification | |
CN106254612A (en) | A kind of sound control method and device | |
US20230164477A1 (en) | Noise cancellation processing method, device and apparatus | |
WO2014137341A1 (en) | Method for training an automatic speech recognition system | |
CN104202485B (en) | A kind of safety call method, device and mobile terminal | |
CN105847584A (en) | Method for intelligent device to identify private conversations | |
CN109389978A (en) | A kind of audio recognition method and device | |
US9680977B2 (en) | Voice only phone and method of operation | |
US10818298B2 (en) | Audio processing | |
CN109584877B (en) | Voice interaction control method and device | |
US20190304457A1 (en) | Interaction device and program | |
CN111182412B (en) | Electronic equipment, data processing method for electronic equipment and conference system equipment | |
JP2019219509A (en) | Robot, control method of the same, and program | |
CN103929532A (en) | Information processing method and electronic equipment | |
KR102417899B1 (en) | Apparatus and method for recognizing voice of vehicle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: 261031 Dongfang Road, Weifang high tech Industrial Development Zone, Shandong, China, No. 268 Applicant after: Goertek Inc. Address before: 261031 Dongfang Road, Weifang high tech Industrial Development Zone, Shandong, China, No. 268 Applicant before: Goertek Inc. |
|
COR | Change of bibliographic data | ||
GR01 | Patent grant | ||
GR01 | Patent grant |