CN107863101A - A kind of speech recognition equipment of intelligent home device - Google Patents
A kind of speech recognition equipment of intelligent home device Download PDFInfo
- Publication number
- CN107863101A CN107863101A CN201711251101.6A CN201711251101A CN107863101A CN 107863101 A CN107863101 A CN 107863101A CN 201711251101 A CN201711251101 A CN 201711251101A CN 107863101 A CN107863101 A CN 107863101A
- Authority
- CN
- China
- Prior art keywords
- electric capacity
- module
- resistance
- operational amplifier
- electrically connected
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
Abstract
The invention discloses a kind of speech recognition equipment of intelligent home device, including speech recognition equipment shell and the circuit board for being arranged on speech recognition equipment enclosure, the circuit board is provided with anti-aliasing filtering module, A/D modular converters, endpoint detection module, pretreatment module, characteristic extracting module and HMM type selection modules, and anti-aliasing filtering module, A/D modular converters, endpoint detection module, pretreatment module, characteristic extracting module and HMM type selection modules are electrically connected with successively, the anti-aliasing filtering module includes operational amplifier U1, operational amplifier U1 pin 2 is electrically connected with electric capacity C3 one end, electric capacity C4 one end and resistance R2 one end.The present invention is economical and practical, anti-aliasing filtering module is increased newly between sound identification module and A/D modular converters, there is the phenomenon of aliased distortion in the data signal that can effectively avoid A/D modular converters from being changed, and the filter factor of anti-aliasing filtering module can be adjusted precisely by potentiometer RP.
Description
Technical field
The present invention relates to the speech recognition dress of speech recognition equipment technical field, more particularly to a kind of intelligent home device
Put.
Background technology
Speech recognition equipment is the device for user voice signal to be identified, with the continuous hair of science and technology
Exhibition, speech recognition equipment has on intelligent home device to be extremely widely applied, and the voice of existing intelligent home device is known
Other device includes sound identification module, A/D modular converters, endpoint detection module, pretreatment module, characteristic extracting module, template
Storehouse and HMM type selection modules, HMM type selection modules with equipment by being wirelessly attached, and ATL and the choosing of HMM types
Select module and use bidirectional data transfers, be electrically connected with due to being used between sound identification module and the progress of A/D modular converters, in language
Lack anti-aliasing filtering module between sound identification module and A/D modular converters so that the data signal of A/D modular converters output is easy
There is the phenomenon of fold-over distortion.
The content of the invention
Based on technical problem existing for background technology, the present invention proposes a kind of speech recognition dress of intelligent home device
Put.
A kind of speech recognition equipment of intelligent home device proposed by the present invention, including speech recognition equipment shell and setting
In the circuit board of speech recognition equipment enclosure, the circuit board is provided with anti-aliasing filtering module, A/D modular converters, end points
Detection module, pretreatment module, characteristic extracting module and HMM type selection modules, and anti-aliasing filtering module, A/D modular converters,
Endpoint detection module, pretreatment module, characteristic extracting module and HMM type selection modules are electrically connected with successively.
The anti-aliasing filtering module includes operational amplifier U1, and operational amplifier U1 pin 2 is electrically connected with electric capacity C3
One end, electric capacity C4 one end and resistance R2 one end, the resistance R2 other end is electrically connected with electric capacity C1 one end, electric capacity C2
One end and resistance R1 one end, the resistance R1 other end is electrically connected with sound identification module, and sound identification module is installed
In the side of speech recognition equipment shell on one side, operational amplifier U1 pin 4 is grounded, and operational amplifier U1 pin 8 is electrical
Potentiometer RP the first fixing end is connected with, and potentiometer RP the second fixing end is electrically connected with power supply VCC and potentiometer RP
Sliding end, operational amplifier U1 pin 3 is electrically connected with resistance R3 one end, and resistance R3 other end ground connection, computing
Amplifier U1 pin 1 is electrically connected with the electric capacity C1 other end, the electric capacity C2 other end, the electric capacity C3 other end, electric capacity C4
The other end and resistance R4 one end, the resistance R4 other end is electrically connected with electric capacity C5 one end, electric capacity C6 one end and electricity
R5 one end is hindered, and the resistance R5 other end is electrically connected with operational amplifier U2 pin 2, electric capacity C7 one end and electric capacity C8
One end, operational amplifier U2 pin 4 is grounded, and operational amplifier U2 pin 8 and potentiometer RP the first fixing end are electrical
Connection, operational amplifier U2 pin 3 are electrically connected with resistance R6 one end, and resistance R6 other end ground connection, operation amplifier
Device U2 pin 1 other end with electric capacity C5, the electric capacity C6 other end, the electric capacity C7 other end and electric capacity the C8 respectively other end
It is electrically connected with, and the input of operational amplifier U2 pin 1 and A/D modular converters is electrically connected with.
Preferably, the HMM type selection modules are electrically connected with ATL, and ATL and HMM type selection modules
Using bidirectional data transfers, HMM type selection modules are by being wirelessly connected with equipment.
Preferably, the model WH19 of the potentiometer RP, and power supply VCC is 5 volts of power supplys.
Preferably, the model of the operational amplifier U1 and operational amplifier U2 are LM346, and operational amplifier U1 and
Operational amplifier U2 is programmable operational amplifier.
Preferably, the resistance R1, resistance R2, resistance R3, resistance R4, resistance R5 and resistance R6 resistance are followed successively by 10K
Ω、10KΩ、2.4KΩ、10KΩ、10KΩ、2.4KΩ。
Preferably, the electric capacity C1, electric capacity C2, electric capacity C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7 and electric capacity C8
Capacitance is followed successively by 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F.
Preferably, the sound identification module is used to the sound that user sends be identified, then sound identification module
It is sent in the form of analog signal in anti-aliasing filtering module.
Preferably, the data signal that the endpoint detection module is exported using display mode to A/D modular converters is held
Point detection, then the data signal after detection is transported in pretreatment module and pre-processed by endpoint detection module, last pre-
Processing module is controlled according to the result after processing to characteristic extracting module.
Preferably, the HMM type selection modules are used to receive the signal characteristic transmitted by characteristic extracting module, then
Signal characteristic of the HMM type selection modules according to transmitted by characteristic extracting module is selected from ATL, last HMM types
Selecting module carries out controlled in wireless to equipment.
Beneficial effects of the present invention:(1)By having increased anti-aliasing filtering newly between sound identification module and A/D modular converters
Module, and anti-aliasing filtering module is by operational amplifier U1, operational amplifier U2, resistance R1, resistance R2, resistance R3, resistance R4, electricity
Hinder R5, resistance R6, electric capacity C1, electric capacity C2, electric capacity C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7, electric capacity C8, potentiometer RP
Formed with power supply VCC, the analog signal that anti-aliasing filtering module can export to sound identification module is removed the width in environment
Band random noise, then anti-aliasing filtering module the analog signal after processing is transported on A/D modular converters, greatly avoid
There is the phenomenon of aliased distortion in the data signal that A/D modular converters are changed;(2)By in power supply VCC and operational amplifier U1
Pin 8 and operational amplifier U2 pin 8 between increased potentiometer RP newly, user can be to anti-aliasing filtering by potentiometer RP
The filter factor of module is precisely adjusted.
The present invention is economical and practical, has increased anti-aliasing filtering module newly between sound identification module and A/D modular converters, can be with
There is the phenomenon of aliased distortion, and the filtering of anti-aliasing filtering module in the effective data signal for avoiding A/D modular converters from being changed
Coefficient can be adjusted precisely by potentiometer RP.
Brief description of the drawings
Fig. 1 is a kind of structural representation of the speech recognition equipment of intelligent home device proposed by the present invention.
Fig. 2 is a kind of fundamental diagram of the speech recognition equipment of intelligent home device proposed by the present invention.
Fig. 3 is that a kind of speech recognition equipment moderate resistance of intelligent home device proposed by the present invention mixes the inside electricity of filtration module
Lu Tu.
In figure:1- speech recognition equipments shell, 2- sound identification modules, 3- anti-aliasing filterings module, 4- A/D modular converters,
5- endpoint detection modules, 6- pretreatment modules, 7- characteristic extracting modules, 8- HMM type selection modules, 9- ATLs, 10- are set
It is standby.
Embodiment
The present invention is made with reference to specific embodiment further to explain.
Embodiment:With reference to figure 1-3, a kind of speech recognition equipment of intelligent home device is proposed in the present embodiment, including
Speech recognition equipment shell 1 and the circuit board being arranged on inside speech recognition equipment shell 1, circuit board are provided with anti-aliasing filtering mould
Block 3, A/D modular converters 4, endpoint detection module 5, pretreatment module 6, characteristic extracting module 7 and HMM type selection modules 8, and
Anti-aliasing filtering module 3, A/D modular converters 4, endpoint detection module 5, pretreatment module 6, characteristic extracting module 7 and the choosing of HMM types
Module 8 is selected to be electrically connected with successively;Anti-aliasing filtering module 3 includes operational amplifier U1, and operational amplifier U1 pin 2 is electrically connected with
There is one end of electric capacity C3 one end, electric capacity C4 one end and resistance R2, the resistance R2 other end is electrically connected with the one of electric capacity C1
The one end at end, electric capacity C2 one end and resistance R1, the resistance R1 other end is electrically connected with sound identification module 2, and voice is known
Other module 2 is arranged on the side of speech recognition equipment shell 1 on one side, and operational amplifier U1 pin 4 is grounded, operational amplifier U1
Pin 8 be electrically connected with potentiometer RP the first fixing end, and potentiometer RP the second fixing end is electrically connected with power supply VCC
With potentiometer RP sliding end, operational amplifier U1 pin 3 is electrically connected with resistance R3 one end, and the resistance R3 other end
Ground connection, operational amplifier U1 pin 1 be electrically connected with the electric capacity C1 other end, the electric capacity C2 other end, electric capacity C3 it is another
End, the electric capacity C4 other end and resistance R4 one end, the resistance R4 other end are electrically connected with electric capacity C5 one end, electric capacity C6
One end and resistance R5 one end, and the resistance R5 other end is electrically connected with operational amplifier U2 pin 2, electric capacity C7 one end
With electric capacity C8 one end, operational amplifier U2 pin 4 is grounded, and operational amplifier U2 pin 8 is solid with the first of potentiometer RP
Fixed end is electrically connected with, and operational amplifier U2 pin 3 is electrically connected with resistance R6 one end, and resistance R6 other end ground connection,
Operational amplifier U2 pin 1 other end with electric capacity C5, the electric capacity C6 other end, the electric capacity C7 other end and electric capacity C8 respectively
The other end be electrically connected with, and the input of operational amplifier U2 pin 1 and A/D modular converters 4 is electrically connected with, by language
Anti-aliasing filtering module 3 is increased between sound identification module 2 and A/D modular converters 4 newly, and anti-aliasing filtering module 3 is by operational amplifier
U1, operational amplifier U2, resistance R1, resistance R2, resistance R3, resistance R4, resistance R5, resistance R6, electric capacity C1, electric capacity C2, electric capacity
C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7, electric capacity C8, potentiometer RP and power supply VCC compositions, anti-aliasing filtering module 3 can be with
The analog signal exported to sound identification module 2 is removed the broadband random noise in environment, and then anti-aliasing filtering module 3 will
Analog signal after processing is transported on A/D modular converters 4, greatly avoids the data signal that A/D modular converters 4 are changed
There is the phenomenon of aliased distortion, by power supply VCC and operational amplifier U1 pin 8 and operational amplifier U2 pin 8 it
Between increased potentiometer RP newly, the filter factor that user can resist mixed filtration module 3 by potentiometer RP is precisely adjusted, this
Invent economical and practical, increased anti-aliasing filtering module 3 newly between sound identification module 2 and A/D modular converters 4, can be effective
There is the phenomenon of aliased distortion, and the filter factor of anti-aliasing filtering module 3 in the data signal for avoiding A/D modular converters 4 from being changed
Can precisely it be adjusted by potentiometer RP.
In the present embodiment, HMM type selection modules 8 are electrically connected with ATL 9, and ATL 9 and HMM type selecting moulds
Block 8 uses bidirectional data transfers, and HMM type selection modules 8 are by being wirelessly connected with equipment 10, potentiometer RP model
WH19, and power supply VCC is 5 volts of power supplys, operational amplifier U1 and operational amplifier U2 model are LM346, and operation amplifier
Device U1 and operational amplifier U2 is programmable operational amplifier, resistance R1, resistance R2, resistance R3, resistance R4, resistance R5 and electricity
Resistance R6 resistance is followed successively by 10K Ω, 10K Ω, 2.4K Ω, 10K Ω, 10K Ω, 2.4K Ω, electric capacity C1, electric capacity C2, electric capacity C3, electricity
The capacitance for holding C4, electric capacity C5, electric capacity C6, electric capacity C7 and electric capacity C8 is followed successively by 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33 μ F, 33
μ F, 33 μ F, sound identification module 2 are used to the sound that user sends be identified, and then sound identification module 2 is with analog signal
Form be sent in anti-aliasing filtering module 3, the numeral that endpoint detection module 5 is exported to A/D modular converters 4 using display mode
Signal carries out end-point detection, and then the data signal after detection is transported in pretreatment module 6 and carried out in advance by endpoint detection module 5
Processing, last pretreatment module 6 are controlled according to the result after processing to characteristic extracting module 7, HMM type selection modules 8
For receiving the signal characteristic transmitted by characteristic extracting module 7, then HMM type selection modules 8 are according to the institute of characteristic extracting module 7
The signal characteristic of transmission is selected from ATL 9, and last HMM type selection modules 8 carry out controlled in wireless to equipment 10, is led to
Cross and increased anti-aliasing filtering module 3 newly between sound identification module 2 and A/D modular converters 4, and anti-aliasing filtering module 3 is by computing
Amplifier U1, operational amplifier U2, resistance R1, resistance R2, resistance R3, resistance R4, resistance R5, resistance R6, electric capacity C1, electric capacity
C2, electric capacity C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7, electric capacity C8, potentiometer RP and power supply VCC compositions, anti-aliasing filtering mould
The analog signal that block 3 can export to sound identification module 2 is removed the broadband random noise in environment, then anti-aliasing filtering
Analog signal after processing is transported on A/D modular converters 4 by module 3, greatly avoids what A/D modular converters 4 were changed
There is the phenomenon of aliased distortion in data signal, passes through the pin 8 and operational amplifier U2 in power supply VCC and operational amplifier U1
Potentiometer RP is increased between pin 8 newly, the filter factor that user can resist mixed filtration module 3 by potentiometer RP is carried out precisely
Adjustment, the present invention is economical and practical, has increased anti-aliasing filtering module 3 newly between sound identification module 2 and A/D modular converters 4, can be with
There is the phenomenon of aliased distortion, and the filter of anti-aliasing filtering module 3 in the effective data signal for avoiding A/D modular converters 4 from being changed
Wave system number can be adjusted precisely by potentiometer RP.
In the present embodiment, when sound identification module 2 receives the sound instruction of user, sound identification module 2 can be to receiving
To sound instruction be identified, the broadband random noise being wherein mingled with the sound instruction of user in environment, then voice
Identification module 2 is sent in the form of analog signal by operational amplifier U1, operational amplifier U2, resistance R1, resistance R2, resistance
R3, resistance R4, resistance R5, resistance R6, electric capacity C1, electric capacity C2, electric capacity C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7, electric capacity
C8, potentiometer RP and power supply VCC composition anti-aliasing filterings module 3, what anti-aliasing filtering module 3 can export to sound identification module 2
Analog signal is removed the broadband random noise in environment, and then anti-aliasing filtering module 3 conveys the analog signal after processing
Onto A/D modular converters 4 so that the data signal that A/D modular converters 4 are changed is not in the phenomenon of aliased distortion, and A/D turns
Data signal after conversion is sent to endpoint detection module 5 by mold changing block 4, and endpoint detection module 5 is turned using display mode to A/D
Change the mold the data signal that block 4 exports and carry out end-point detection, then the data signal after detection is transported to pre- by endpoint detection module 5
Being pre-processed in processing module 6, last pretreatment module 6 is controlled according to the result after processing to characteristic extracting module 7,
Characteristic extracting module 7 sends signal characteristic to HMM type selection modules 8, and HMM type selection modules 8 are according to characteristic extracting module 7
Transmitted signal characteristic is selected from ATL 9, and last HMM type selection modules 8 carry out controlled in wireless to equipment 10;
When the filter factor inside anti-aliasing filtering module 3 needs adjustment in addition, it is only necessary to rotate the completion of potentiometer RP cans.
Claims (9)
1. a kind of speech recognition equipment of intelligent home device, including speech recognition equipment shell(1)Be arranged on speech recognition
Crust of the device(1)Internal circuit board, it is characterised in that:The circuit board is provided with anti-aliasing filtering module(3), A/D moduluss of conversion
Block(4), endpoint detection module(5), pretreatment module(6), characteristic extracting module(7)With HMM type selection modules(8), and it is anti-
Mixed filtration module(3), A/D modular converters(4), endpoint detection module(5), pretreatment module(6), characteristic extracting module(7)With
HMM type selection modules(8)It is electrically connected with successively;
The anti-aliasing filtering module(3)Including operational amplifier U1, operational amplifier U1 pin 2 is electrically connected with electric capacity C3's
One end of one end, electric capacity C4 one end and resistance R2, the resistance R2 other end are electrically connected with electric capacity C1 one end, electric capacity C2
One end and resistance R1 one end, the resistance R1 other end are electrically connected with sound identification module(2), and sound identification module(2)
Installed in speech recognition equipment shell(1)Side on one side, operational amplifier U1 pin 4 is grounded, and operational amplifier U1's draws
Pin 8 is electrically connected with potentiometer RP the first fixing end, and potentiometer RP the second fixing end is electrically connected with power supply VCC and electricity
Position device RP sliding end, operational amplifier U1 pin 3 are electrically connected with resistance R3 one end, and resistance R3 another termination
Ground, operational amplifier U1 pin 1 be electrically connected with the electric capacity C1 other end, the electric capacity C2 other end, electric capacity C3 the other end,
The electric capacity C4 other end and resistance R4 one end, the resistance R4 other end are electrically connected with electric capacity C5 one end, the one of electric capacity C6
End and resistance R5 one end, and the resistance R5 other end be electrically connected with operational amplifier U2 pin 2, electric capacity C7 one end and
Electric capacity C8 one end, operational amplifier U2 pin 4 are grounded, and operational amplifier U2 pin 8 is fixed with the first of potentiometer RP
End is electrically connected with, and operational amplifier U2 pin 3 is electrically connected with resistance R6 one end, and resistance R6 other end ground connection, fortune
Calculate amplifier U2 pin 1 other end with electric capacity C5, the electric capacity C6 other end, the electric capacity C7 other end and electric capacity C8 respectively
The other end is electrically connected with, and operational amplifier U2 pin 1 and A/D modular converters(4)Input be electrically connected with.
A kind of 2. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The HMM classes
Type selecting module(8)It is electrically connected with ATL(9), and ATL(9)With HMM type selection modules(8)Using bi-directional data
Transmission, HMM type selection modules(8)By being wirelessly connected with equipment(10).
A kind of 3. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The potentiometer
RP model WH19, and power supply VCC is 5 volts of power supplys.
A kind of 4. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The computing is put
Big device U1 and operational amplifier U2 model is LM346, and operational amplifier U1 and operational amplifier U2 are programmable fortune
Calculate amplifier.
A kind of 5. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The resistance
R1, resistance R2, resistance R3, resistance R4, resistance R5 and resistance R6 resistance be followed successively by 10K Ω, 10K Ω, 2.4K Ω, 10K Ω,
10KΩ、2.4KΩ。
A kind of 6. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The electric capacity
C1, electric capacity C2, electric capacity C3, electric capacity C4, electric capacity C5, electric capacity C6, electric capacity C7 and electric capacity C8 capacitance are followed successively by 33 μ F, 33 μ F, 33 μ
F、33μF、33μF、33μF、33μF、33μF。
A kind of 7. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The voice is known
Other module(2)For the sound that user sends to be identified, then sound identification module(2)Sent in the form of analog signal
To anti-aliasing filtering module(3)On.
A kind of 8. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The end points inspection
Survey module(5)Using display mode to A/D modular converters(4)The data signal of output carries out end-point detection, then end-point detection
Module(5)Data signal after detection is transported to pretreatment module(6)On pre-processed, last pretreatment module(6)Root
According to the result after processing to characteristic extracting module(7)It is controlled.
A kind of 9. speech recognition equipment of intelligent home device according to claim 1, it is characterised in that:The HMM classes
Type selecting module(8)For receiving characteristic extracting module(7)Transmitted signal characteristic, then HMM type selection modules(8)Root
According to characteristic extracting module(7)Transmitted signal characteristic is from ATL(9)In selected, last HMM type selection modules(8)
To equipment(10)Carry out controlled in wireless.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711251101.6A CN107863101A (en) | 2017-12-01 | 2017-12-01 | A kind of speech recognition equipment of intelligent home device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711251101.6A CN107863101A (en) | 2017-12-01 | 2017-12-01 | A kind of speech recognition equipment of intelligent home device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107863101A true CN107863101A (en) | 2018-03-30 |
Family
ID=61704699
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711251101.6A Pending CN107863101A (en) | 2017-12-01 | 2017-12-01 | A kind of speech recognition equipment of intelligent home device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107863101A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1996002911A1 (en) * | 1992-10-05 | 1996-02-01 | Matsushita Electric Industrial Co., Ltd. | Speech detection device |
CN102360187A (en) * | 2011-05-25 | 2012-02-22 | 吉林大学 | Chinese speech control system and method with mutually interrelated spectrograms for driver |
CN102589611A (en) * | 2011-01-13 | 2012-07-18 | 江苏东华测试技术股份有限公司 | Device for testing, analyzing and collecting dynamic signal |
CN103475551A (en) * | 2013-09-11 | 2013-12-25 | 厦门狄耐克电子科技有限公司 | Intelligent home system based on voice recognition |
CN105553799A (en) * | 2016-02-29 | 2016-05-04 | 深圳市广佳乐新智能科技有限公司 | Intelligent housing system based on voice recognition |
EP3091534A1 (en) * | 2014-03-17 | 2016-11-09 | Huawei Technologies Co., Ltd | Method and apparatus for processing speech signal according to frequency domain energy |
-
2017
- 2017-12-01 CN CN201711251101.6A patent/CN107863101A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1996002911A1 (en) * | 1992-10-05 | 1996-02-01 | Matsushita Electric Industrial Co., Ltd. | Speech detection device |
US5579431A (en) * | 1992-10-05 | 1996-11-26 | Panasonic Technologies, Inc. | Speech detection in presence of noise by determining variance over time of frequency band limited energy |
CN102589611A (en) * | 2011-01-13 | 2012-07-18 | 江苏东华测试技术股份有限公司 | Device for testing, analyzing and collecting dynamic signal |
CN102360187A (en) * | 2011-05-25 | 2012-02-22 | 吉林大学 | Chinese speech control system and method with mutually interrelated spectrograms for driver |
CN103475551A (en) * | 2013-09-11 | 2013-12-25 | 厦门狄耐克电子科技有限公司 | Intelligent home system based on voice recognition |
EP3091534A1 (en) * | 2014-03-17 | 2016-11-09 | Huawei Technologies Co., Ltd | Method and apparatus for processing speech signal according to frequency domain energy |
CN105553799A (en) * | 2016-02-29 | 2016-05-04 | 深圳市广佳乐新智能科技有限公司 | Intelligent housing system based on voice recognition |
Non-Patent Citations (2)
Title |
---|
刘洪涛等: "高品质抗混叠滤波器的设计", 《仪器仪表用户》 * |
闫三锋: "基于单片机语音识别系统设计", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104637488B (en) | The method and terminal device of acoustic processing | |
CN206061141U (en) | A kind of acoustical testing system of wireless audio products | |
CN104348960A (en) | Method for interaction control of intelligent mobile phone and sex appliance via bluetooth and system thereof | |
CN107864419A (en) | It is a kind of for the simulating of active noise reduction, numeral mixing noise reduction system | |
CN107863101A (en) | A kind of speech recognition equipment of intelligent home device | |
CN104599681A (en) | Method and device for processing audio file | |
CN103176005A (en) | System and method for processing signals | |
CN203787078U (en) | HDMI digital karaoke reverberator | |
CN109743658A (en) | A kind of information processing method and electronic equipment | |
CN204169196U (en) | A kind of Portable movable pays bracelet | |
CN104483716A (en) | Metal detector adjusting device | |
CN104199792B (en) | A kind of power taking communication apparatus, mobile terminal and its means of communication and system | |
CN207588983U (en) | A kind of concealed equipment of camera | |
CN110968001A (en) | High-speed analog acquisition board card based on FPGA + MCU | |
CN205754818U (en) | Audio frequency adapter | |
CN209030422U (en) | A kind of speaker numerical control system | |
CN211604687U (en) | Tea table type household electrical appliance control equipment based on far field voice | |
CN108197068A (en) | Embedded real-time data acquires and storage system and its method | |
CN206960528U (en) | Signal generation apparatus | |
CN203387650U (en) | Novel audio sharing device | |
CN207572064U (en) | A kind of speech recognition circuit and translation system | |
CN209102868U (en) | A kind of portable relay intelligent test device | |
CN204573502U (en) | A kind of computer USB jointing type camera | |
CN207475783U (en) | A kind of Super Bass Loudspeaker Enclosure and electronic equipment | |
CN205754816U (en) | Cloud audio sound card |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180330 |