CN207304797U - A kind of device for eliminating TV and disturbing speech recognition equipment - Google Patents
A kind of device for eliminating TV and disturbing speech recognition equipment Download PDFInfo
- Publication number
- CN207304797U CN207304797U CN201721073400.0U CN201721073400U CN207304797U CN 207304797 U CN207304797 U CN 207304797U CN 201721073400 U CN201721073400 U CN 201721073400U CN 207304797 U CN207304797 U CN 207304797U
- Authority
- CN
- China
- Prior art keywords
- audio
- unit
- signal
- video
- hdmi
- 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.)
- Active
Links
Landscapes
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
A kind of device for eliminating TV and disturbing speech recognition equipment, including for connecting the audio receiver of tv audio output terminal, the HDMI audio and video separator for connecting television set top box HDMI audio and video output terminals, video output unit, audio output Unit one, audio selector, echo cancellation unit, speech enhan-cement and noise reduction unit, audio collection unit, audio output Unit two, playing module；The echo cancellation unit is connected with audio selector, audio collection device, speech enhan-cement and noise reduction unit；The audio selector is connected with HDMI audio and video separator, audio receiver；Together with the utility model introduces television audio signals, the separation of HDMI audio and video, television video reduction, tv audio reduction, audio selection, echo cancellor, speech enhan-cement are combined with technologies such as noise reduction, audio collections, television set and intelligent sound equipment new technical solution used at the same time are constituted, solves the problems, such as that television set and intelligent sound equipment are used at the same time.
Description
Technical field
Ntelligent television technolog field is the utility model is related to, more particularly to a kind of TV that eliminates disturbs speech recognition equipment
Device.
Background technology
Joyful intelligent sound box equipment is increasingly popularized now.It is daily with us in basic playing function for intelligent sound box
The common speaker used is no different, and after adding intelligent concept, this speaker is no longer only the equipment for playing music, more various
Occupation mode and allow it to be subject to universal welcome with the interaction of user.For example, user can be against intelligent sound box
Say：" you put the song of one section of Liu De China to me ", intelligent sound will play the song of Liu De China automatically, without manually choosing
Select the process of song.Why intelligent sound box can distinguish the voice of people, be to rely on speech recognition technology.
But no matter the sound identification module inside intelligent sound box is provided at present, is also provided in outside intelligent sound box
The sound identification module in portion, its speech recognition technology is all there are limitation, when we see TV, when TV sound and
When the phonetic order that user sends occurs at the same time, speech recognition poor effect.The reason is that the collection of audio collection device is mixing
Sound, including the sound of TV and the voice of people, due to speech recognition poor effect, cause television set and intelligent sound box cannot be same
When use.Number of patent application：201611135608.0th, patent name " sound enhancement method, device and intelligent sound box, intelligence electricity
Depending on " a kind of sound enhancement method applied to intelligent sound box is proposed, this method uses following steps：Step 1: by microphone
The music primary sound and voice voice of array pickup are converted to multi-path digital signal through ADC；Step 2: obtain the multi-path digital letter
The digital signal all the way that number converted array is converted into；Step 3: the ginseng of echo cancellor is obtained from the digital signal all the way
Examine signal；Step 4: carrying out the elimination of the music primary sound using AEC algorithms based on the reference signal, voice voice is exported
Data.
This method there are the problem of be：The statement of " music primary sound " is wrong in one, step 1, music is primary be
What broadcasting was taken before is only music primary sound.The voice and music that he sends the user that microphone array column array is collected into are mixed in
Together, it is wrong to take music primary sound again from the sound to mix.The primary sound is made a phone call just as both sides, and A side is beaten
To B side, with hands-free, the sound of A side is sent first for B side, is released by hands-free after being obtained after sending by B side, first got again
The voice of A side be only primary sound, and it is clearly to violate " music primary sound " that above-mentioned patent document obtains primary sound from mixing sound
Definition；Secondly, key technology point be conceptual sentence, no referential：Repeatedly mention and " being picked up by microphone array in that patent
The music primary sound and people that the speaker that takes plays are spoken the voice voice of generation, and multi-path digital signal, Ran Houyou are converted to through ADC
The multi-path digital signal is changed into digital signal all the way by FPGA, is sent to CPU, is obtained by CPU from the digital signal all the way
The reference signal of echo cancellor." actually CPU how to obtain the reference signal of echo cancellor, this key point is in its specification
In full none word introduction, only conclusion and without particular content, without convincingness, therefore, which does not have without implementary value
There is comparativity.
To sum up, the prior art cannot still solve the problems, such as that television set and intelligent sound box equipment are used at the same time, or still cannot
Solve the problems, such as that third party's speech ciphering equipment and intelligent sound equipment characterized by user mutual are used at the same time.
The content of the invention
The utility model in view of the deficiencies of the prior art, proposes a kind of dress for eliminating TV and disturbing speech recognition equipment
Put, to solve the problems, such as that the prior art cannot solve television set and intelligent sound box equipment is used at the same time, or the 3rd cannot be solved
Square speech ciphering equipment and the intelligent sound equipment problem used at the same time characterized by user mutual.
The above-mentioned technical purpose of the utility model technical scheme is that：
A kind of device for eliminating TV and disturbing speech recognition equipment, including for connecting the sound of tv audio output terminal
Frequency receiver, the HDMI audio and video separator for connecting television set top box HDMI audio and video output terminals, video output unit,
Audio output Unit one, audio selector, echo cancellation unit, speech enhan-cement and noise reduction unit, audio collection unit, audio are defeated
Go out Unit two, playing module；
The input terminal connection HDMI audio and video separators of the video output unit, output terminal connection television set；
The input terminal of the audio selector connects HDMI audio and video separator unit and audio receiver unit, defeated respectively
Outlet connects echo cancellation unit and audio output Unit one；
The input terminal of the echo cancellation unit connects audio collection device and audio selector, output terminal connection language respectively
Sound strengthens and noise reduction unit；
The one unit input terminal of audio output connects audio selector, output terminal connection playing module；
Two unit input terminal of the audio output connection speech enhan-cement connects speech recognition system with noise reduction unit, output terminal
System.
It is preferred that the HDMI audio and video separator receives the sound from television set top box HDMI audio and video output terminals
Screen signal is simultaneously separated into audio signal, vision signal, its vision signal connects television set by video output unit and broadcast
Put, so as to complete the reduction of television set top box HDMI video signal；Its audio signal is defeated by audio selector unit, audio
Go out a unit, playing module plays out so that complete television set top box HDMI audio signal reduction；The broadcasting mould
Block plays the playing module of loudspeaker for acquisition tv audio signal and replacing TV.
It is preferred that the audio selector is used to receive the two-way from HDMI audio and video separator and audio receiver
Audio signal, can only receive wherein audio signal, and the audio signal all the way that synchronization is received all the way at the same moment
Echo cancellation unit is sent respectively to carry out echo cancellor；Audio output Unit one is sent to, to reduce audio output
The audio signal of Unit one；The audio signal of audio output Unit one is the audio signal from HDMI audio and video separators
Or the audio signal from audio receiver.
It is preferred that the signal of the audio collection unit collection includes being mixed by television audio signals and the voice signal of people
The mixed audio signal formed, the mixed audio signal are conveyed to echo cancellation unit by audio collection device, the echo cancellor list
Member balances out television audio signals from mixed audio signal and retains the voice signal of people according to the reference signal taken,
And the voice signal of the people of reservation is sent to speech enhan-cement and noise reduction unit；The reference signal taken is to come
From the input signal of audio selector.
It is preferred that the speech enhan-cement and noise reduction unit include being directed to the voice signal of the people retained after echo cancellor into
Row speech enhan-cement and noise reduction.
It is preferred that the HDMI audio and video separator, audio receiver, video output unit pass through wired mode respectively
Or wireless mode is connected with HDMI audio and video output terminal, tv audio output terminal, the television set of TV set-top box；The sound
By wired or be wirelessly connected with speech recognition system, the speech recognition system includes frequency output Unit two
Intelligent sound box speech recognition system, intelligent home voice identifying system.
A kind of intelligent home control system, it is characterised in that：Including one kind as described in claim 1 to 6 any one
Eliminate the device that TV disturbs speech recognition equipment.
A kind of intelligent sound box, it is characterised in that：The intelligent sound box connects such as claim 1 by wired or wireless mode
The device disturbed to a kind of elimination TV described in 6 any one speech recognition equipment.
A kind of smart television, it is characterised in that：Including intelligent sound box as claimed in claim 8.
The utility model has the advantages of
1st, the utility model introduces television audio signals, HDMI audio and video separate, television video reduces, television set sound
Together with the technical combinations such as frequency reduction, audio selection, echo cancellor, speech enhan-cement and noise reduction, audio collection, television set is constituted
With intelligent sound equipment new technical solution used at the same time, solve television set and intelligent sound equipment is used at the same time asks
Topic.
2nd, the utility model each several part is mutually supported, interdepends, and forms an organic whole：By in the present apparatus
Introduce television audio signals, television audio signals are incorporated into the input terminal of echo cancellation unit, solve and believe television audio
Number as echo cancellor reference signal possibility sex chromosome mosaicism；By setting HDMI audio and video separators, solve by HDMI sounds,
The problem of video is separated；By the combination of HDMI audio and video separator and screen output unit, solve TV and regard
Frequency reduction problem；By setting HDMI audio and video separator, audio receiver, audio selector, audio output Unit one, broadcasting
Module solves the problems, such as television audio reduction；By setting echo cancellation unit and the audio for sending audio selector to believe
Number it is sent to echo cancellation unit, the mixed audio signal that audio collection device gathers is sent to echo cancellation unit, solves
The problem of offsetting television audio signals and retaining voice signal；By setting speech enhan-cement and noise reduction unit, solves raising language
The problem of sound signal quality.
Brief description of the drawings
Fig. 1 is a kind of device circuit block diagram for eliminating TV and disturbing speech recognition equipment of the utility model.
Embodiment
The design philosophy and design principle of the utility model
1st, the purpose of this utility model is the interference for eliminating TV to speech recognition equipment, since being to eliminate doing for TV
Disturb it is necessary to which the audio signal of TV is removed from mixed audio signal.Just imagine, two sound are at least mixed with to one
The voice flow of sound, will separate them, then remove one of them, difficulty is what big.Just as one bottle of blue ink and one bottle it is red
Ink is poured on together, then needs red ink to extract, this probably can not possibly.But in fact, except this mixing
Signal, we can obtain producing the pervious original signal of mixed signal, this original signal is exactly reference signal.
2nd, in the utility model, the reference signal is both a television audio clean, without other audios
Signal, the reference signal is as the sample signal in " echo cancellor ", and as its name suggests, which is used to eliminate in " echo cancellor "
The television audio signals of sampled signal are corresponded in mixed audio.
3rd, it is determined that television audio signals are used as with reference to after signal, ensuing problem is how to obtain reference signal, ginseng
Examining signal must take before mixed audio signal is produced, or reference signal must be from an independent television audio letter
Number, the utility model method of sampling is exactly that the audio signal of TV is incorporated into the device of the utility model.
4th, TV signal is incorporated into the device of the utility model, there are several key points：Take the voice ginseng of TV
Traditional tv and the two class user of DTV with TV set-top box must be taken into account by examining signal, this two paths of signals is required for introducing
Into the device of the utility model；The vision signal and audio signal of traditional tv are separated, it is only necessary to introduce tradition electricity
Depending on audio signal, the audio-video signal line concentration on the TV set-top box of DTV on HDMI audio and video output terminals,
So the HDMI audio-video signals of TV set-top box are fully incorporated in the utility model device.HDMI audio-video signals draw
To be separated after entering, that is, the HDMI audio-video signal separators using the utility model；After audio-video signal separation
Video reduction and audio reduction are carried out, video reduction will be finally reduced to television set；Audio signal will be also reduced to TV
Machine, but since the utility model takes the audio signal for television set of knowing clearly, with the playing module replacing TV voluntarily set
Loudspeaker, therefore, audio signal are that the audio signal of television set is reduced by homemade playing module.Due to the utility model
Take into account the demand of conventional television user and digital TV user, access is two-way tv audio signal, and export to
" echo cancellor " unit and " unit of audio output one ", otherwise be the audio signal of conventional television, or be DTV
The audio signal of machine, so, export to " echo cancellor " unit and " unit of audio output one " must be one of which signal,
Therefore, need to set an audio selector in the utility model, for identifying different types of audio signal, targetedly
Carry out echo cancellor and audio reduction.Input, the output design of echo cancellation unit.Echo cancellation unit should have two
A input terminal：One input terminal is used to take clean reference signal, its clean reference signal comes from audio selector；It is another
A input terminal is used to connect audio collection device, and what audio collection device gathered is the sound of sound of television and the voice mixing of people；Return
Sound eliminates the design of unit output terminal：Signal after echo cancellor is the signal of the voice of the people remained, at this time, is retained
The signal of the voice of the people to get off can't be directly output to speech recognition system, because the voice signal of people can usually be subject to ring
Border is disturbed, such as reflection or the other electric appliances of wall（Air-conditioning）Noise can all reduce the voice quality of people, therefore, after echo cancellor
The voice signal of the people of reservation should be sent to speech enhan-cement and noise reduction unit.
Based on above design philosophy and design principle, the utility model devises a kind of TV that eliminates to speech recognition equipment
The device of interference, as shown in Figure 1：
Regarded including the audio receiver for connecting tv audio output terminal, for connecting television set top box HDMI sounds
HDMI audio and video separator, video output unit, audio output Unit one, audio selector, the echo cancellor list of frequency output terminal
Member, speech enhan-cement and noise reduction unit, audio collection unit, audio output Unit two, playing module；
The input terminal connection HDMI audio and video separators of the video output unit, output terminal connection television set；
The input terminal of the audio selector connects HDMI audio and video separator unit and audio receiver unit, defeated respectively
Outlet connects echo cancellation unit and audio output Unit one；
The input terminal of the echo cancellation unit connects audio collection device and audio selector, output terminal connection language respectively
Sound strengthens and noise reduction unit；
The one unit input terminal of audio output connects audio selector, output terminal connection playing module；
Two unit input terminal of the audio output connection speech enhan-cement connects speech recognition system with noise reduction unit, output terminal
System.
The HDMI audio and video separator receives the sound screen letter from television set top box HDMI audio and video output terminals
Number and be separated into audio signal, vision signal, its vision signal connects television set by video output unit and plays out so that
Complete the reduction of television set top box HDMI video signal；Its audio signal is single by audio selector unit, audio output one
Member, playing module play out, so as to complete the audio signal reduction of television set top box HDMI；The playing module is to obtain
Obtain tv audio signal and replacing TV plays the playing module after loudspeaker.
The audio selector is used to receive to be believed from the two-way audio of HDMI audio and video separator and audio receiver
Number, wherein audio signal, and the audio signal all the way that synchronization is received is sent out respectively all the way can only be received at the same moment
Echo cancellation unit is given to carry out echo cancellor；Audio output Unit one is sent to, to reduce audio output Unit one
Audio signal；The audio signal of audio output Unit one is the audio signal from HDMI audio and video separators or comes
From the audio signal of audio receiver.
The signal of the audio collection unit collection includes what is mixed by the voice signal of television audio signals and people
Mixed audio signal, the mixed audio signal are conveyed to echo cancellation unit by audio collection device, the echo cancellation unit according to
The reference signal taken balances out television audio signals from mixed audio signal and retains the voice signal of people, and will protect
The voice signal of the people stayed is sent to speech enhan-cement and noise reduction unit；The reference signal taken is to come from audio
The input signal of selector.
The voice signal that the speech enhan-cement and noise reduction unit include being directed to the people retained after echo cancellor carries out voice
Enhancing and noise reduction, the voice de-noising means are the noise reduction means taken for environmental noise.
The HDMI audio and video separator, audio receiver, video output unit pass through wired or wireless way respectively
HDMI audio and video output terminal, tv audio output terminal, television set with TV set-top box are connected；The audio output two is single
For member by wired or be wirelessly connected with speech recognition system, the speech recognition system includes intelligent sound box language
Sound identifying system, intelligent home voice identifying system.
A kind of intelligent home control system, it is characterised in that：Including one kind as described in claim 1 to 6 any one
Eliminate the device that TV disturbs speech recognition equipment.
A kind of intelligent sound box, it is characterised in that：The intelligent sound box connects such as claim 1 by wired or wireless mode
The device disturbed to a kind of elimination TV described in 6 any one speech recognition equipment.
A kind of smart television, it is characterised in that：Including intelligent sound box as claimed in claim 8.
It is emphasized that embodiment described in the utility model is illustrative, rather than it is limited, therefore this reality
With the new embodiment including being not limited to described in embodiment.
Claims (9)
- A kind of 1. device for eliminating TV and disturbing speech recognition equipment, it is characterised in that：Including for connecting tv audio The audio receiver of output terminal, the HDMI audio and video separator for connecting television set top box HDMI audio and video output terminals, regard Frequency output unit, audio output Unit one, audio selector, echo cancellation unit, speech enhan-cement and noise reduction unit, audio collection Unit, audio output Unit two, playing module；The input terminal connection HDMI audio and video separators of the video output unit, output terminal connection television set；The input terminal of the audio selector connects HDMI audio and video separator unit and audio receiver unit, output terminal respectively Connect echo cancellation unit and audio output Unit one；The input terminal of the echo cancellation unit connects audio collection device and audio selector respectively, and output terminal connection voice increases Strong and noise reduction unit；The one unit input terminal of audio output connects audio selector, output terminal connection playing module；Two unit input terminal of the audio output connection speech enhan-cement connects speech recognition system with noise reduction unit, output terminal.
- A kind of 2. device for eliminating TV and disturbing speech recognition equipment according to claim 1, it is characterised in that：It is described HDMI audio and video separator receive and the audio-video signal from television set top box HDMI audio and video output terminals and be separated into sound Frequency signal, vision signal, its vision signal connects television set by video output unit and plays out, so as to complete television set machine The reduction of top box HDMI video signal；Its audio signal by audio selector unit, audio output Unit one, playing module into Row plays, so as to complete the audio signal reduction of television set top box HDMI；The playing module is acquisition tv audio Signal and replacing TV play the playing module after loudspeaker.
- A kind of 3. device for eliminating TV and disturbing speech recognition equipment according to claim 1, it is characterised in that：It is described Audio selector be used to receiving the two-way audio signal from HDMI audio and video separator and audio receiver, when same Wherein audio signal, and the audio signal all the way that synchronization is received is sent respectively to echo cancellor list all the way can only be received quarter Member is carrying out echo cancellor；Audio output Unit one is sent to, to reduce the audio signal of television set；The audio output The audio signal of Unit one is the audio signal from HDMI audio and video separators or the audio signal from audio receiver.
- A kind of 4. device for eliminating TV and disturbing speech recognition equipment according to claim 1, it is characterised in that：It is described The signal of audio collection unit collection includes the mixed audio signal mixed by the voice signal of television audio signals and people, The mixed audio signal is conveyed to echo cancellation unit by audio collection device, and the echo cancellation unit is according to the reference taken Signal balances out television audio signals from mixed audio signal and retains the voice signal of people, and the voice of the people of reservation is believed Number it is sent to speech enhan-cement and noise reduction unit；The reference signal taken is the input letter from audio selector Number.
- A kind of 5. device for eliminating TV and disturbing speech recognition equipment according to claim 4, it is characterised in that：It is described Speech enhan-cement and noise reduction unit include being directed to the voice signal of people retained after echo cancellor and carry out speech enhan-cement and noise reduction.
- A kind of 6. device for eliminating TV and disturbing speech recognition equipment according to claim 1, it is characterised in that：It is described HDMI audio and video separator, audio receiver, video output unit pass through wired mode or the same television set of wireless mode respectively HDMI audio and video output terminal, tv audio output terminal, the television set of top box are connected；Audio output Unit two is by having Line is wirelessly connected with speech recognition system, and the speech recognition system includes intelligent sound box speech recognition system System, intelligent home voice identifying system.
- A kind of 7. intelligent home control system, it is characterised in that：Disappear including one kind as described in claim 1 to 6 any one The device disturbed except TV speech recognition equipment.
- A kind of 8. intelligent sound box, it is characterised in that：The intelligent sound box by wired or wireless mode connect as claim 1 to The device that a kind of elimination TV described in 6 any one disturbs speech recognition equipment.
- A kind of 9. smart television, it is characterised in that：Including intelligent sound box as claimed in claim 8.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2017209936745 | 2017-08-09 | ||
| CN201720993674 | 2017-08-09 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN207304797U true CN207304797U (en) | 2018-05-01 |
Family
ID=62438806
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201721073400.0U Active CN207304797U (en) | 2017-08-09 | 2017-08-24 | A kind of device for eliminating TV and disturbing speech recognition equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN207304797U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107371059A (en) * | 2017-08-09 | 2017-11-21 | 北京智网时代科技有限公司 | A kind of device for eliminating TV and disturbing speech recognition equipment |
| CN110493616A (en) * | 2018-05-15 | 2019-11-22 | 中国移动通信有限公司研究院 | A kind of acoustic signal processing method, device, medium and equipment |
-
2017
- 2017-08-24 CN CN201721073400.0U patent/CN207304797U/en active Active
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107371059A (en) * | 2017-08-09 | 2017-11-21 | 北京智网时代科技有限公司 | A kind of device for eliminating TV and disturbing speech recognition equipment |
| CN107371059B (en) * | 2017-08-09 | 2023-07-18 | 北京智网时代科技有限公司 | Device for eliminating interference of television on voice recognition device |
| CN110493616A (en) * | 2018-05-15 | 2019-11-22 | 中国移动通信有限公司研究院 | A kind of acoustic signal processing method, device, medium and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107371059A (en) | A kind of device for eliminating TV and disturbing speech recognition equipment | |
| CN204836533U (en) | Acoustic control intelligence audio amplifier with echo elimination function | |
| CN106128459B (en) | Phonetic controller and sound control method based on multi-microphone | |
| CN106792097A (en) | Audio signal captions matching process and device | |
| CN207304797U (en) | A kind of device for eliminating TV and disturbing speech recognition equipment | |
| CN106341719A (en) | Synchronized audio play method simultaneously using various kinds of play modules of equipment and apparatus thereof | |
| CN200987193Y (en) | Set up box | |
| CN101335848A (en) | Digital television receiving device and communication method thereof to mobile information apparatus | |
| CN208691406U (en) | Far field speech collecting system for smart television | |
| CN106569774A (en) | Method of eliminating noise, and terminal | |
| CN105933748A (en) | Implementation method and equipment for displaying barrages on live television program | |
| CN205987323U (en) | Audio processing circuit and terminal based on type c interface | |
| CN203423767U (en) | Multifunctional digital television system | |
| CN204168404U (en) | A kind of processing unit of ultra high-definition signal and full HD TV | |
| CN107424617A (en) | A kind of multimedia-data procession device and method | |
| CN201054689Y (en) | A multi-media receiving video/audio terminal | |
| CN207766522U (en) | A kind of circular audio-video system of wireless bass | |
| CN201957185U (en) | Television integrated machine with Cable Modem function | |
| CN201118750Y (en) | Home digital camera with network shooting function | |
| CN207731645U (en) | A kind of tone frequency channel wire and system for Real-time speech recognition | |
| CN201639688U (en) | Multimedia audiovisual receiving mobile television | |
| CN205943467U (en) | Audio playback device and terminal | |
| CN200956615Y (en) | Video-watching mobile phone | |
| CN107333214B (en) | A kind of audio frequency broadcast system based on area distribution formula speaker | |
| CN213028377U (en) | Mobile wireless IPTV set top box |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |