CN208422407U - It is a kind of for far saying the echo eliminating device of speech recognition - Google Patents
It is a kind of for far saying the echo eliminating device of speech recognition Download PDFInfo
- Publication number
- CN208422407U CN208422407U CN201820891897.5U CN201820891897U CN208422407U CN 208422407 U CN208422407 U CN 208422407U CN 201820891897 U CN201820891897 U CN 201820891897U CN 208422407 U CN208422407 U CN 208422407U
- Authority
- CN
- China
- Prior art keywords
- echo
- far
- echo cancellation
- loudspeaker
- signal
- 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
- Circuit For Audible Band Transducer (AREA)
Abstract
It is a kind of for far saying the echo eliminating device of speech recognition, it include: microphone array column unit, it includes acoustic microphones sensor, microphone array column unit acquires the first audio signal mixed by far saying voice signal and echo signal in the acquisition position for positioning extraneous sound source, and is sent to DSP echo cancellation mould group；TV loudspeaker sound pick-up, is mounted near the loudspeaker of television set, and TV loudspeaker sound pick-up acquires second audio signal of the stereo sound of TV loudspeaker as echo source sound, and is sent to DSP echo cancellation mould group；And DSP echo cancellation mould group, the DSP echo cancellation mould group receive two groups of audio signals, fidelity is exported after echo cancellation far says voice signal to speech recognition module.The utility model creative use sound pick-up directly acquires distal reference signal of the source of sound as echo cancellation of TV loudspeaker, haves no need to change the hardware circuit and structure of television set, simple and effective, so that it may to realize echo cancellation function.
Description
Technical field
The utility model belongs to field of speech recognition, in particular to the echo eliminating device in a kind of speech recognition scene,
Innovative usage sound pick-up acquires echo source of sound.
Background technique
With the development of artificial intelligence technology, many smart machines using interactive voice as a kind of mode of human-computer interaction,
The especially voice application scene of television set needs user to say voice far to operate control television set, is far saying voice control field
Jing Zhong, the audio signal that television local plays can generate space reflection, reenter after forming echo for acquiring voice control
The microphone of system, superposition become the received audio signal a part of microphone, and the presence meeting jamming equipment of echo far says the external world
The identification of voice.Echo cancellation module is applied at this time, can balance out the echo of TV, echo from the mixed audio of microphone
It is eliminated in signal and only retains the external world and far say voice signal, so as to which remote speech identification, voice control etc. is better achieved
Using.
The echo cancellation AEC principle applied at present: microphone, which can be received, far says voice and reality by the external world actually required
The mixed audio signal of border unwanted echo synthesis, the broadcasting echo cancellation of mixed audio signal, however television audio
Collected distal reference signal (i.e. local audio signal) and echo are discrepant on decoding line, if directly joined with distal end
It examines signal and reversely subtracts each other mixed audio signal, the external world can be made far to say that voice changes so much that one loses one's identity, institute is in favor of distal reference signal
With the high correlation of echo, echo cancellation model solution function F is established, obtaining function F can be from distal reference signal meter
Calculation obtains the real echo signal that microphone receives, and then reversely subtracts from mixed audio signal and is calculated by function F
Echo signal, voice signal is far said by the external world of remaining i.e. height reduction.
In existing echo cancellation scheme, acquisition television local audio signal is acquired from television audio decoding line road,
At this moment the hardware circuit and structure for needing to change original television set, it is convenient to be unsatisfactory for the terminal devices such as the television set of market volume
Realization far says voice identification control function, high, time-consuming and laborious to finished product television set improvement cost.
Utility model content
Technical problem to be solved in the utility model is to overcome above-mentioned the deficiencies in the prior art, is provided a kind of for far saying
The echo eliminating device of speech recognition far says voice to easily improve to increase for terminal devices such as the television sets of market volume
Control function.
The technical scheme in the invention for solving the above technical problem is: a kind of for far saying the echo of speech recognition
Cancellation element, comprising:
Microphone array column unit comprising acoustic microphones sensor, the acquisition of microphone array column unit is by far saying that voice is believed
Number and the first audio signal for mixing of echo signal, and be sent to DSP echo cancellation mould group；
TV loudspeaker sound pick-up, is mounted near the loudspeaker of television set, and TV loudspeaker sound pick-up acquires TV loudspeaker
Second audio signal of the stereo sound as echo source sound, and it is sent to DSP echo cancellation mould group；
And DSP echo cancellation mould group, the DSP echo cancellation mould group receives two groups of audio signals, by echo cancellation
Export fidelity afterwards far says voice signal to speech recognition module.
The DSP echo cancellation mould group include: two groups of audio signals of interim synchronous storage inputoutput buffer unit,
To two groups of audio signal gains control to guarantee the automatic gain control unit of output amplitude, reversed neutralize in the first audio signal
The active noise reduction unit and echo cancellation unit of ambient noise, the echo cancellation unit are obtained back by the second audio signal
Sound signal, voice signal is far said in acquisition after carrying out echo signal counteracting to the first audio signal.
The DSP echo cancellation mould group further include: digital audio converting unit is connected with USB joint, and digital audio turns
Changing unit will far say that voice signal is converted to USB digital audio, and USB joint is by the USB interface of television set by USB digital audio
It is input to speech recognition module.
The TV loudspeaker sound pick-up selects the sound pick-up of sensitivity -50dB~30dB and rational design.
The acoustic microphones sensor is provided with 2-6.
The microphone array column unit includes auditory localization module, far field identification module and filtering noise reduction module.
Compared with prior art, the utility model has the beneficial effects that the utility model creative use sound pick-up is directly adopted
Collect distal reference signal of the source of sound of TV loudspeaker as echo cancellation, have no need to change the hardware circuit and structure of television set,
It is simple and effective, so that it may to realize echo cancellation function；And far say the microphone array of voice collecting, TV loudspeaker sound pick-up,
DSP echo cancellation mould group, USB joint form the peripheral device of a television set, and the audio signal after echo cancellation is connect by USB
Mouth is input to the speech recognition module of television set, and voice control function is far said in the television set increase convenient for market volume.
Detailed description of the invention
It, below will be to required in embodiment description in order to illustrate more clearly of the technical scheme in the embodiment of the utility model
Attached drawing to be used is briefly described, in which:
Fig. 1 is the echo eliminating device structural block diagram of the utility model preferred embodiment；
Fig. 2 is the DSP echo cancellation modular structure block diagram of the utility model preferred embodiment；
Fig. 3 is the echo eliminating device functional block diagram of the utility model preferred embodiment.
Specific embodiment
Below in conjunction with attached drawing and specific embodiment, the utility model is described further.In preferred embodiment
Cited such as "upper", "lower", "left", "right", " centre " and " one " term, only being illustrated convenient for narration, rather than to
The enforceable range of the utility model is limited, relativeness is altered or modified, under the content of no substantial changes in technology, when also
It is considered as the enforceable scope of the utility model.
Referring to Figure 1 to Fig. 3, a kind of echo for far saying speech recognition of the utility model preferred embodiment design disappears
Except device, microphone array column unit 1, TV loudspeaker sound pick-up 2, DSP echo cancellation mould group 3 and USB joint are specifically included that
4, casing is packed by microphone array column unit 1, TV loudspeaker sound pick-up 2 and DSP echo cancellation mould group 3 and forms television set etc.
The peripheral device of terminal device.
Microphone array column unit 1 is mainly made of the acoustic microphones sensor of certain amount, and microphone array column unit 1 is also
With far field identification module and sound source locating module, microphone array column unit 1 determines extraneous sound source according to sound source positioning function
Position, using the position as acquisition position, the voice signal acquired in acquisition position is as the first audio signal, the first audio
Signal is sent to DSP echo cancellation mould group 3 by far saying that the echo signal of voice signal and television set mixes.
Preferably, microphone array column unit 1 further includes filtering noise reduction module, can be according to receiving user voice
Time it is different, position the direction where user, filter the noise in other directions, the identification of auxiliary far field reaches more accurate knowledge
Other effect.
Preferably, acoustic microphones sensor includes 2-6.
TV loudspeaker sound pick-up 2 is mounted near the left and right loudspeaker of television set, it is therefore an objective to acquire the stereo of TV loudspeaker
Second audio signal of the sound as echo source sound, and it is sent to DSP echo cancellation mould group 3.
Preferably, the pickup of muting sensitivity -50dB~30dB and rational design can be selected in TV loudspeaker sound pick-up 2
Device effectively prevent the interference of other noises at a distance convenient for the loudspeaker source of sound in locking effective range.
DSP echo cancellation mould group 3 includes: inputoutput buffer unit, automatic gain control unit, active noise reduction list
Member, echo cancellation unit and digital audio converting unit.Inputoutput buffer unit, by the first audio signal and the second sound
Frequency signal temporarily stores, and makes synchronizing information；Automatic gain control unit carries out certainly the first audio signal and the second audio signal
Dynamic gain control, to guarantee output amplitude requirement；Active noise reduction unit generates the reversed sound wave equal with ambient noise, by the
Noise in one audio signal neutralizes；Echo cancellation unit obtains echo signal using the second audio signal, believes the first audio
It number carries out echo signal to offset and back drop filtration treatment, accurately far says voice signal to obtain；Digital audio converting unit
It is connected to USB joint 4, will far say that voice signal is converted to USB digital audio, USB digital audio can prevent audio disturbed,
Guarantee that the audio signal of high-fidelity, last USB joint 4 are input to speech recognition module by the USB interface of television set and carry out language
Sound identifying processing.
The above descriptions are merely preferred embodiments of the present invention, not makees in any form to the utility model
Limitation, anyone skilled in the art are not departing within the scope of technical solutions of the utility model, the skill according to the utility model
Art any brief introduction modification, equivalent variations and modification substantially to the above embodiments, still fall within the utility model technical side
In the range of case.
Claims (6)
1. a kind of for far saying the echo eliminating device of speech recognition characterized by comprising
Microphone array column unit comprising acoustic microphones sensor, microphone array column unit acquisition by far say voice signal and
The first audio signal that echo signal mixes, and it is sent to DSP echo cancellation mould group；
TV loudspeaker sound pick-up, is mounted near the loudspeaker of television set, and TV loudspeaker sound pick-up acquires the solid of TV loudspeaker
Second audio signal of the acoustic sound as echo source sound, and it is sent to DSP echo cancellation mould group；
And DSP echo cancellation mould group comprising echo cancellation unit, the echo cancellation unit receive two groups of audio signals
Afterwards, echo signal is obtained by the second audio signal, obtains the remote of fidelity after carrying out echo signal counteracting to the first audio signal
Say voice signal；
The peripheral hardware that casing forms television set is assembled by microphone array column unit, TV loudspeaker sound pick-up and DSP echo cancellation mould
Device.
2. according to claim 1 for far saying the echo eliminating device of speech recognition, which is characterized in that the DSP is returned
Sound eliminates mould group further include: digital audio converting unit is connected with USB joint, and digital audio converting unit will far say voice
Signal is converted to USB digital audio, and USB digital audio is input to speech recognition module by USB interface by USB joint.
3. according to claim 1 for far saying the echo eliminating device of speech recognition, which is characterized in that the DSP is returned
Sound eliminates mould group further include: interim synchronize stores the inputoutput buffer unit of two groups of audio signals, to two groups of audio signals
Gain control is to guarantee the automatic gain control unit of output amplitude and neutralize the active of ambient noise in the first audio signal
Noise reduction unit.
4. according to any one of claims 1 to 3 for far saying the echo eliminating device of speech recognition, which is characterized in that
The TV loudspeaker sound pick-up selects the sound pick-up of sensitivity -50dB~30dB and rational design.
5. according to any one of claims 1 to 3 for far saying the echo eliminating device of speech recognition, which is characterized in that
The acoustic microphones sensor is provided with 2-6.
6. according to any one of claims 1 to 3 for far saying the echo eliminating device of speech recognition, which is characterized in that
The microphone array column unit includes auditory localization module, far field identification module and filtering noise reduction module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820891897.5U CN208422407U (en) | 2018-06-07 | 2018-06-07 | It is a kind of for far saying the echo eliminating device of speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201820891897.5U CN208422407U (en) | 2018-06-07 | 2018-06-07 | It is a kind of for far saying the echo eliminating device of speech recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN208422407U true CN208422407U (en) | 2019-01-22 |
Family
ID=65110709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201820891897.5U Active CN208422407U (en) | 2018-06-07 | 2018-06-07 | It is a kind of for far saying the echo eliminating device of speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN208422407U (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110138991A (en) * | 2019-06-21 | 2019-08-16 | 苏州思必驰信息科技有限公司 | Echo removing method and device |
| CN110300318A (en) * | 2019-05-15 | 2019-10-01 | 深圳康佳电子科技有限公司 | A kind of USB terminal interface circuit and USB terminal interface circuit control method |
| CN111756943A (en) * | 2020-06-05 | 2020-10-09 | 中原工学院 | Self-adaptive stereo acoustic echo cancellation method for sub-band with increased complex value and proportional value |
| CN113286182A (en) * | 2021-04-02 | 2021-08-20 | 福州智象信息技术有限公司 | Method and system for eliminating echo between TV and sound pickup peripheral |
| CN113555029A (en) * | 2021-07-21 | 2021-10-26 | 歌尔科技有限公司 | Voice equipment control method, system, medium and voice equipment |
-
2018
- 2018-06-07 CN CN201820891897.5U patent/CN208422407U/en active Active
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110300318A (en) * | 2019-05-15 | 2019-10-01 | 深圳康佳电子科技有限公司 | A kind of USB terminal interface circuit and USB terminal interface circuit control method |
| CN110138991A (en) * | 2019-06-21 | 2019-08-16 | 苏州思必驰信息科技有限公司 | Echo removing method and device |
| CN110138991B (en) * | 2019-06-21 | 2021-05-25 | 思必驰科技股份有限公司 | Echo cancellation method and device |
| CN111756943A (en) * | 2020-06-05 | 2020-10-09 | 中原工学院 | Self-adaptive stereo acoustic echo cancellation method for sub-band with increased complex value and proportional value |
| CN113286182A (en) * | 2021-04-02 | 2021-08-20 | 福州智象信息技术有限公司 | Method and system for eliminating echo between TV and sound pickup peripheral |
| CN113555029A (en) * | 2021-07-21 | 2021-10-26 | 歌尔科技有限公司 | Voice equipment control method, system, medium and voice equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN208422407U (en) | It is a kind of for far saying the echo eliminating device of speech recognition | |
| CN203563160U (en) | Interactive intelligent set top box | |
| CN102044253B (en) | Echo signal processing method and system as well as television | |
| CN205249484U (en) | Microphone linear array reinforcing directive property adapter | |
| TWI595792B (en) | Multi-channel digital microphone | |
| CN110349582A (en) | Display device and far field speech processing circuit | |
| CN107124647A (en) | A kind of panoramic video automatically generates the method and device of subtitle file when recording | |
| CN105208189A (en) | Audio processing method and mobile terminal | |
| CN110223707A (en) | Display device | |
| CN108510997A (en) | Electronic equipment and echo cancel method applied to electronic equipment | |
| CN103702258A (en) | Microphone device and microphone setting method for eliminating interference of near-field sound source | |
| CN203057371U (en) | Camera device having echo elimination function and television set | |
| CN111402910B (en) | Method and equipment for eliminating echo | |
| CN104540015A (en) | Automatic volume adjustment method and device applied to online class system | |
| CN103559878A (en) | Method for eliminating noise in audio information and device thereof | |
| CN103974168A (en) | Information processing method and electronic devices | |
| CN110366067B (en) | Far field voice module echo cancellation circuit and device | |
| CN109378011B (en) | On-site audio playing and collecting system and echo eliminating method | |
| CN103152686B (en) | A kind of digital deaf-aid of customizable functionality pattern and its implementation | |
| CN214851543U (en) | Recording and broadcasting equipment | |
| CN105554633A (en) | Sound signal processing method for built-in sound effect device and electronic device | |
| CN106419954B (en) | One kind being suitable for stethoscopic vibration restoring method | |
| CN209419784U (en) | A kind of circular array microphone suitable for raspberry pie | |
| WO2021004067A1 (en) | Display device | |
| CN211089900U (en) | K sings earphone |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |