CN205987367U - Sound pick up equipment based on distributing type microphone - Google Patents
Sound pick up equipment based on distributing type microphone Download PDFInfo
- Publication number
- CN205987367U CN205987367U CN201620703824.XU CN201620703824U CN205987367U CN 205987367 U CN205987367 U CN 205987367U CN 201620703824 U CN201620703824 U CN 201620703824U CN 205987367 U CN205987367 U CN 205987367U
- Authority
- CN
- China
- Prior art keywords
- module
- mike
- microphone
- wireless communication
- primary dcreening
- 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
The utility model discloses a sound pick up equipment based on distributing type microphone, the main control unit who is connected including distributed microphone, with each microphone radio communication, the microphone includes audio acquisition module, analog -to -digital conversion module, network equipment clock and wireless communication module, main control unit includes prescreening module and data analysis module, and the prescreening module is used for carrying out a plurality of data signal that the preliminary treatment obtained the prescreening according to the time characteristic value of each microphone, data analysis module is used for adopting the maximum likelihood estimation method to carry out assay to a plurality of data signal of prescreening, and the data signal who gets the biggest probability is as speech output, wherein, each microphone transmits respective data signal and time characteristic value to main control unit respectively. Distributed microphone is regarded as the pickup matrix to this sound pick up equipment for the microphone is close to the user more, can reduce the demand to the speech input acoustic pressure, in order to improve the speech recognition rate of accuracy, improves user's experience simultaneously and feels.
Description
Technical field
The present invention relates to pickup technology field, a kind of more particularly to sound pick up equipment based on distributed mike.
Background technology
It is frequently accompanied by various noises in our residing environment, for example, the television set sound in room and fan sound, automobile
Running car sound on interior electromotor sound, road, Babble noise in coffee shop etc..Noise is to multiple voice processing system
Have a negative impact.And, current speech collecting system either far field or near field is required for enough acoustic pressure inputs, special
It is not for speech recognition system；In speech recognition system, noise and low acoustic pressure input so that phonetic recognization rate decline,
Even make identifying system complete failure, the experience sense of impact interaction.
Content of the invention
In order to solve above-mentioned the deficiencies in the prior art, the invention provides a kind of pickup dress based on distributed mike
Put, pickup matrix is used as so that mike is closer to user by distributed mike, can reduce for phonetic entry
The demand of acoustic pressure；According to temporal characteristics value, primary dcreening operation is carried out to the voice messaging of each mike collection, and to primary dcreening operation result using
The maximum-likelihood estimation technique is analyzed, and finally obtains the pure and high voice output of degree of accuracy, to improve speech recognition accuracy, with
The experience sense of Shi Tigao user.
The technical problem to be solved is achieved by the following technical programs：
A kind of sound pick up equipment based on distributed mike, including：Distributed mike and each mike radio communication are even
The master controller connecing；Described mike includes audio collection module, analog-to-digital conversion module, network equipment clock and radio communication mold
Block, described audio collection module is connected with wireless communication module through analog-to-digital conversion module, described network equipment clock and channel radio
Letter module connects；Described master controller includes primary dcreening operation module and data analysis module, and described primary dcreening operation module is used for according to each Mike
The temporal characteristics value of wind carries out multiple digital signals that pretreatment obtains primary dcreening operation；Described data analysis module is used for many to primary dcreening operation
Individual digital signal is analyzed using maximum likelihood estimate, and the digital signal taking maximum of probability is as voice output；Wherein, institute
State each mike to transmit respective digital signal and temporal characteristics value to described master controller respectively.
Further, described mike also includes echo noise reduction process module, its respectively with described audio collection module and
Analog-to-digital conversion module connects.
The present invention has the advantages that：Pickup matrix is used as so that mike more leans on by distributed mike
Nearly user, can reduce the demand for phonetic entry acoustic pressure；The voice messaging that each mike is gathered is according to temporal characteristics
Value carries out primary dcreening operation, and primary dcreening operation result is analyzed using maximum likelihood estimate, finally obtains the pure and high language of degree of accuracy
Sound exports, and to improve speech recognition accuracy, improves the experience sense of user simultaneously.
Brief description
Fig. 1 is the schematic flow sheet of pickup screening technique of the present invention；
Fig. 2 is the principle schematic of sound pick up equipment of the present invention.
Specific embodiment
The present invention will be described in detail with reference to the accompanying drawings and examples.
As shown in figure 1, a kind of pickup screening technique based on distributed mike, specifically include following steps：
Step a, the distributed mike of initialization, carry out time synchronized to each mike and master controller；Each mike
Respectively with described master controller wireless communication connection；
Step b, each mike collection voice messaging simultaneously change into digital signal, by digital signal and the transmission of temporal characteristics value
To described master controller；
When implementing, arrange multiple mikes, set up be made up of multiple mikes and master controller based on Ad-hoc
The distributed mike network of network (also known as interim self-organizing network), each mike is in communication with each other with master controller respectively.Should
Master controller is arranged in speech recognition equipment or electronic equipment.
In mike network, network node is each mike；Each mike includes audio collection module, echo
Noise reduction process module, analog-to-digital conversion module, wireless communication module, described audio collection module is through echo noise reduction process module, mould
Number modular converter is connected with wireless communication module；Wherein, audio collection module is used for gathering the voice of user in local environment
Information, the voice messaging that described echo noise reduction process module is used for collection carries out echo noise reduction process, described analog-to-digital conversion module
For the voice messaging after echo noise reduction process is changed into digital signal；Also include network equipment clock, with radio communication mold
Block connects, and it is used for the time of synchronous each mike and main controller module, is synchronized based on ntp network time protocol；Should
Network equipment clock is preferably provided on described wireless communication module, and when audio collection module starts to gather audio frequency, reading should
Network clocking on wireless communication module, as temporal characteristics value；Described wireless communication module is used for being responsible for mike and master control
Data transfer between device processed, this data includes digital signal and time eigenvalue.
, according to the temporal characteristics value of each mike, pretreatment obtains multiple numeral letters of primary dcreening operation for step c, described master controller
Number；
When implementing, meeting synchronous joining day parameter during the sampling of audio collection module commencing signal, i.e. time eigenvalue,
The time parameter of this time parameter and described master controller is synchronous that is to say, that described master controller and multiple mikes
It is that unification carrys out sampled data using a clock.Temporal characteristics value away from the mike of sound source and the mike near sound source
Temporal characteristics value is different, and close near the temporal characteristics value of multiple mikes of sound source, then can pass through to receive several
Temporal characteristics value, the preliminary position delimiting sound source, can be according to the numeral letter of the optimized multiple primary dcreening operations of corresponding algorithm picks
Number.
Described algorithm is to judge the distance between sound source and mike according to the biography ripple time difference of sound；Multiple primary dcreening operations
The corresponding mike of digital signal recently or is closer to from sound source.
Step d, the multiple digital signals to primary dcreening operation carry out maximum likelihood estimate analysis, take the digital signal of maximum of probability
As voice output, complete pickup.
When implementing, described maximum likelihood estimate is specially：The digital signal of multiple primary dcreening operations is compared, and analyzes
Background noise, then wiping out background noise, obtain clearly phonetic order data, more each road phonetic order data,
Thus obtaining the digital signal of maximal possibility estimation as voice output, complete pickup.
It should be noted that in described maximum likelihood estimate calculating process, the step algorithm being adopted all can be by existing
Technology is realized, and is not particularly limited in the present invention.
This pickup screening technique is used as pickup matrix so that mike is closer to user by distributed mike,
The demand for phonetic entry acoustic pressure can be reduced；The voice messaging of each mike collection is carried out just according to temporal characteristics value
Sieve, and adopt maximum likelihood estimate to analyze primary dcreening operation result, finally obtain the pure and high voice output of degree of accuracy, to improve
Speech recognition accuracy, improves the experience sense of user simultaneously.
As shown in Fig. 2 a kind of sound pick up equipment based on distributed mike, including：Distributed mike and each mike
The master controller of wireless communication connection；Described mike include audio collection module, analog-to-digital conversion module, network equipment clock and
Wireless communication module, described audio collection module is connected with wireless communication module through analog-to-digital conversion module, during the described network equipment
Clock is connected with wireless communication module；Described master controller includes primary dcreening operation module and data analysis module, and described primary dcreening operation module is used for
Multiple digital signals that pretreatment obtains primary dcreening operation are carried out according to the temporal characteristics value of each mike；Described data analysis module is used for
Maximum likelihood estimate analysis is carried out to multiple digital signals of primary dcreening operation, the digital signal taking maximum of probability is as voice output；
Wherein, described each mike transmits respective digital signal and temporal characteristics value to described master controller respectively.
Described master controller also includes memory module, is connected with described primary dcreening operation module and data analysis module respectively, is used for
Data before and after storage analysis.Described master controller is preferably provided in a certain mike, or is arranged on and described sound pick up equipment
On the electronic equipment using cooperatively, such as speech recognition equipment etc..
Described mike also includes echo noise reduction process module, its respectively with described audio collection module and analog digital conversion mould
Block connects.
Embodiment described above only have expressed embodiments of the present invention, and its description is more concrete and detailed, but can not
Therefore it is interpreted as the restriction to the scope of the claims of the present invention, as long as the skill being obtained in the form of equivalent or equivalent transformation
Art scheme, all should fall within the scope and spirit of the invention.
Claims (2)
1. a kind of sound pick up equipment based on distributed mike is it is characterised in that include：Distributed mike and each mike
The master controller of wireless communication connection；Described mike include audio collection module, analog-to-digital conversion module, network equipment clock and
Wireless communication module, described audio collection module is connected with wireless communication module through analog-to-digital conversion module, during the described network equipment
Clock is connected with wireless communication module；Described master controller includes primary dcreening operation module and data analysis module, and described primary dcreening operation module is used for
Multiple digital signals that pretreatment obtains primary dcreening operation are carried out according to the temporal characteristics value of each mike；Described data analysis module is used for
Multiple digital signals of primary dcreening operation are analyzed using maximum likelihood estimate, take the digital signal of maximum of probability defeated as voice
Go out；Wherein, described each mike transmits respective digital signal and temporal characteristics value to described master controller respectively.
2. the sound pick up equipment based on distributed mike according to claim 1 is it is characterised in that described mike also wraps
Include echo noise reduction process module, it is connected with described audio collection module and analog-to-digital conversion module respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201620703824.XU CN205987367U (en) | 2016-07-06 | 2016-07-06 | Sound pick up equipment based on distributing type microphone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201620703824.XU CN205987367U (en) | 2016-07-06 | 2016-07-06 | Sound pick up equipment based on distributing type microphone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN205987367U true CN205987367U (en) | 2017-02-22 |
Family
ID=58022876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201620703824.XU Active CN205987367U (en) | 2016-07-06 | 2016-07-06 | Sound pick up equipment based on distributing type microphone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN205987367U (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106707788A (en) * | 2017-03-09 | 2017-05-24 | 上海电器科学研究院 | Intelligent and household voice control and recognition system |
| CN107592600A (en) * | 2016-07-06 | 2018-01-16 | 深圳市三诺声智联股份有限公司 | A kind of pickup screening technique and sound pick up equipment based on distributed microphone |
| CN107600111A (en) * | 2017-11-06 | 2018-01-19 | 郑州铁路局郑州通信段 | A kind of alarm device system |
| CN109936798A (en) * | 2017-12-19 | 2019-06-25 | 展讯通信（上海）有限公司 | The method, apparatus and server of pickup are realized based on distributed MIC array |
| CN112767947A (en) * | 2021-01-21 | 2021-05-07 | 厦门科路德科技有限公司 | Sound and gas composite detection device |
-
2016
- 2016-07-06 CN CN201620703824.XU patent/CN205987367U/en active Active
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107592600A (en) * | 2016-07-06 | 2018-01-16 | 深圳市三诺声智联股份有限公司 | A kind of pickup screening technique and sound pick up equipment based on distributed microphone |
| CN107592600B (en) * | 2016-07-06 | 2024-04-02 | 深圳市三诺声智联股份有限公司 | Pickup screening method and pickup device based on distributed microphones |
| CN106707788A (en) * | 2017-03-09 | 2017-05-24 | 上海电器科学研究院 | Intelligent and household voice control and recognition system |
| CN106707788B (en) * | 2017-03-09 | 2019-05-28 | 上海电器科学研究院 | A kind of intelligent home voice control identifying system and method |
| CN107600111A (en) * | 2017-11-06 | 2018-01-19 | 郑州铁路局郑州通信段 | A kind of alarm device system |
| CN109936798A (en) * | 2017-12-19 | 2019-06-25 | 展讯通信（上海）有限公司 | The method, apparatus and server of pickup are realized based on distributed MIC array |
| CN112767947A (en) * | 2021-01-21 | 2021-05-07 | 厦门科路德科技有限公司 | Sound and gas composite detection device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN205987367U (en) | Sound pick up equipment based on distributing type microphone | |
| CN111223497B (en) | Nearby wake-up method and device for terminal, computing equipment and storage medium | |
| CN105632486B (en) | Voice awakening method and device of intelligent hardware | |
| US10373630B2 (en) | Systems and methods for energy efficient and low power distributed automatic speech recognition on wearable devices | |
| CN105744434B (en) | A kind of intelligent sound box control method and system based on gesture identification | |
| CN106504762B (en) | Bird community number estimation system and method | |
| CN103456305A (en) | Terminal and speech processing method based on multiple sound collecting units | |
| CN105448303A (en) | Voice signal processing method and apparatus | |
| CN102111314A (en) | Smart home voice control system and method based on Bluetooth transmission | |
| CN111540347A (en) | Cable tunnel monitoring method and system based on audio | |
| WO2016173132A1 (en) | Method and device for voice recognition, and user equipment | |
| CN107592600A (en) | A kind of pickup screening technique and sound pick up equipment based on distributed microphone | |
| CN107680594B (en) | Distributed intelligent voice acquisition and recognition system and acquisition and recognition method thereof | |
| CN105182763A (en) | Intelligent remote controller based on voice recognition and realization method thereof | |
| CN105244042B (en) | A kind of speech emotional interactive device and method based on finite-state automata | |
| CN106328151A (en) | Environment de-noising system and application method | |
| CN109935226A (en) | A kind of far field speech recognition enhancing system and method based on deep neural network | |
| CN109874096A (en) | A kind of ears microphone hearing aid noise reduction algorithm based on intelligent terminal selection output | |
| CN111142066A (en) | Direction-of-arrival estimation method, server, and computer-readable storage medium | |
| CN109831732A (en) | Intelligent chauvent's criterion device and method based on smart phone | |
| CN113158923A (en) | Bus transfer reminding system based on face recognition | |
| CN110970020A (en) | Method for extracting effective voice signal by using voiceprint | |
| CN112420079B (en) | Voice endpoint detection method and device, storage medium and electronic equipment | |
| CN113259793B (en) | Intelligent microphone and signal processing method thereof | |
| CN105405447A (en) | Telephone transmitter respiration noise shielding method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |