Recording system and its implementation based on multi-microphone array beam formingTechnical field
The present invention relates to mobile terminal recording technology field, more particularly to it is a kind of based on multi-microphone array wave beam atThe recording system and its implementation of shape.
Background technique
It is quick universal with mobile terminal, user for equipment function and it is intelligentized require it is higher and higher, such asWhat keeps mobile terminal more intelligent, specialized, diversification, and is more efficiently used in daily life, has becomeOne of current research direction.
Such as: essentially all of mobile terminal all standard configuration sound-recording functions.However, recording on existing mobile terminalThe disadvantages of there are quality to ensure for function, function is dull leads to this function relative to other dedicated sound pick-up outfits to finishEntirely at chicken ribs.
Have in view of MEMS (Micro-Electro-Mechanical Systems MEMS) microphone higherElectric property and higher audio performance, be able to ascend recording quality, have increasingly been used on mobile terminal.
There is the mobile terminal using four omni-directional microphones in the prior art, is set by the position to microphoneIt counts while the calculating of the environmental noise of entrance is analyzed using audio coprocessor, eliminate unwanted environmental noise, thus intoOne step promotes recording quality.However, above scheme still has certain shortcoming: in the case that mobile terminal does not move,When the sound of someone required for merely desiring to record or sound in some angle, just become some difficulties.
Have at present using four omni-directional microphones and their position is designed while being utilized at audio associationIt manages device to analyze the calculating of the environmental noise of entrance, and eliminates unwanted environmental noise, can further promote recording matterThe solution of amount.It is generally used although this method has begun.But in the case where not mobile hand-held device, merely desire to recordThe sound or the sound in some angle of someone required for oneself just becomes some difficulties.
In view of this, the existing technology needs to be improved and developed.
Summary of the invention
In view of above-mentioned deficiencies of the prior art, the purpose of the present invention is to provide one kind based on multi-microphone array wave beam atThe recording system and its implementation of shape can not change recording side when solving to be recorded using mobile terminal in the prior artTo.And the sound of some angle is only recorded, thus the problem of making troubles to user.
In order to achieve the above object, this invention takes following technical schemes:
A kind of recording system based on multi-microphone array beam forming, wherein include:
Several are used to acquire the microphone of external sound signal；
For will individually be converted into the mould of corresponding audio digital signals per the external sound signal of microphone input all the wayAnalog-to-digital conversion apparatus；
For by each railway digital voice signal and it is pre-set according to beamforming algorithm establish mathematical model intoRow compares, and filters out the audio digital signal processing device of corresponding audio digital signals；
For exporting the output device of the audio digital signals after screening；
The microphone, analog-digital commutator, audio digital signal processing device and output device are sequentially connected.
The recording system based on multi-microphone array beam forming, wherein further include one for after screeningAudio digital signals increase intensity of sound by dynamically reinforcing its phase and amplitude, and re-establish virtual digital space domainThe enhancement device of special angle sound reinforcement is formed with time-domain.
The recording system based on multi-microphone array beam forming, wherein the audio digital signals processing dressCorresponding audio digital signals are filtered out in setting to specifically include:
Each railway digital voice signal and the pre-set mathematical model established according to beamforming algorithm are comparedIt is right, the mixed characteristic data of audio digital signals are identified, and according to preset angle threshold as limiting, to digital languageSound signal is screened, and the audio digital signals within the scope of the angle threshold are retained, to the number outside the angle threshold rangeWord voice signal carries out anti-wave inhibition and filtering.
The recording system based on multi-microphone array beam forming, wherein the number of the microphone is 4,Comprising: which positive first microphone of mobile terminal and second microphone is arranged in, and it is arranged in the of the mobile terminal back sideThree microphones and the 4th microphone；
Wherein, the distance between first microphone and second microphone are between 100mm and 150mm；The thirdThe distance between microphone and the 4th microphone are between 100mm and 150mm；Between first microphone and third microphoneDistance be greater than 18mm；The distance between the second microphone and the 4th microphone are greater than 18mm.
The recording system based on multi-microphone array beam forming, wherein the mobile terminal is mobile phone.
A kind of way of recording based on multi-microphone array beam forming, wherein the described method comprises the following steps:
S1, external sound signal is acquired by several microphones；
S2, corresponding audio digital signals will be individually converted into per the external sound signal of microphone input all the way；
S3, each railway digital voice signal and the pre-set mathematical model established according to beamforming algorithm is enabled to carry outIt compares, filters out corresponding audio digital signals；
Audio digital signals after S4, output screening.
The way of recording based on multi-microphone array beam forming, wherein after the step S3, step S4It before further include step S31, the audio digital signals after screening are strong to increase sound by dynamically reinforcing its phase and amplitudeDegree, and virtual digital space domain and time-domain is re-established to form the enhancement device of special angle sound reinforcement.
The way of recording based on multi-microphone array beam forming, wherein filtered out in the step S2 correspondingAudio digital signals specifically include:
Each railway digital voice signal and the pre-set mathematical model established according to beamforming algorithm are comparedIt is right, the mixed characteristic data of audio digital signals are identified, and according to preset angle threshold as limiting, to digital languageSound signal is screened, and the audio digital signals within the scope of the angle threshold are retained, to the number outside the angle threshold rangeWord voice signal carries out anti-wave inhibition and filtering.
The way of recording based on multi-microphone array beam forming, wherein the number of the microphone is 4,Comprising: which positive first microphone of mobile terminal and second microphone is arranged in, and it is arranged in the of the mobile terminal back sideThree microphones and the 4th microphone；
Wherein, the distance between first microphone and second microphone are between 100mm and 150mm；The thirdThe distance between microphone and the 4th microphone are between 100mm and 150mm；Between first microphone and third microphoneDistance be greater than 18mm；The distance between the second microphone and the 4th microphone are greater than 18mm.
The way of recording based on multi-microphone array beam forming, wherein the mobile terminal is mobile phone.
Compared to the prior art, the recording system and its realization provided by the invention based on multi-microphone array beam formingMethod can change recording direction in real time.And the sound that can only record some angle, is arranged other people except this angleSound and background noise are all effectively eliminated.Meanwhile the position and direction by changing mobile terminal are not needed deliberately will yetAcoustic distance near point of the mobile terminal from needs, to realize the effect preferably recorded in most 5 meters.
Detailed description of the invention
Fig. 1 is the structural block diagram of the recording system provided by the invention based on multi-microphone array beam forming.
Fig. 2 a be the recording system provided by the invention based on multi-microphone array beam forming first embodiment in disappearSound except interference and effect diagram that special angle sound is reinforced.
Fig. 2 b be the recording system provided by the invention based on multi-microphone array beam forming second embodiment in disappearSound except interference and effect diagram that special angle sound is reinforced.
Fig. 3 a is the first wheat in the embodiment of the recording system provided by the invention based on multi-microphone array beam formingThe schematic diagram of gram wind and second microphone.
Fig. 3 b is third wheat in the embodiment of the recording system provided by the invention based on multi-microphone array beam formingThe schematic diagram of gram wind and the 4th microphone.
Fig. 4 is the flow chart of the way of recording provided by the invention based on multi-microphone array beam forming.
Specific embodiment
The present invention provides a kind of recording system and its implementation based on multi-microphone array beam forming.To make this hairBright objects, technical solutions and advantages are clearer, clear, as follows in conjunction with drawings and embodiments to of the invention further detailedExplanation.
Firstly, introducing multi-microphone array beam forming technique (also referred to as Beam forming technology).It is substantially formerIt manages as follows: individually adjusting the phase and wave amplitude of each unit received signal in multi-microphone array, making can be in joint outputThe maximum signal to noise ratio on some direction is obtained, similar to by the way of bandpass filtering, extracts desired signal in a frequency domain.But it is rightThe beam forming of microphone array, work are to occur in the spatial domain, are usually that can be configured to all directions.Due to everyIndividual answers the Fourier of static measurement values to change the received sound left and right ear pulsion phase of different distance and direction, binaural soundThe differential and ears time difference is all variant, to will lead to people in the differentiation of the range and orientation in the region that identification light beam is formedIt can be variant.Thus by the method that these otherness data are acquired and are classified, and the new number establishedModel avoids influence caused by the difference of head movement or number of people form and organ site.
Referring to Fig. 1, it is the structural frames of the recording system provided by the invention based on multi-microphone array beam formingFigure.As shown, the recording system based on multi-microphone array beam forming includes: several microphones 100, modulus turnChanging device 200, audio digital signal processing device 300 and output device 400；The microphone 100, analog-digital commutator 200,Audio digital signal processing device 300 and output device 400 are sequentially connected.
Specifically, microphone 100 is for acquiring external sound signal, wherein external sound signal include: voice andNoise (wherein, the noise includes: echo, reverberation and the low frequency noise for needing to filter etc.).The analog-digital commutator 200For will individually be converted into corresponding audio digital signals per the external sound signal of microphone input all the way, it is noted here thatGround is that analog-digital commutator is that (analog signal turns digital signal to independent sampling processing is carried out per external sound signal all the waySampling).Audio digital signal processing device 300 is used for each railway digital voice signal and pre-set according to beam formingThe mathematical model that algorithm is established is compared, and filters out corresponding audio digital signals.Due to each audio digital signals signalAll having time domain and spatial domain.Each railway digital voice signal requires to be compared with the mathematical model provided in algorithm, distinguishesNot Chu sound general direction, position, loudness, background, quality is far and near, the data of the most mixed characteristic such as tone.AboutBy the agency of has crossed technical principle to the beamforming algorithm in front.The output device 400 is used to export the number after screeningWord voice signal.
It further, can also include one for the audio digital signals after screening to be reinforced its phase and width by dynamicValue increases intensity of sound, and re-establishes virtual digital space domain and time-domain to form the reinforcement of special angle soundEnhancement device.
The recording system based on multi-microphone array beam forming through the invention can only record the sound of some angle,Other voice and background noise except this angle are arranged all effectively to eliminate.Output can be easier to distinguish for hearerKnow the sound in supported angle, filters the voice and environmental noise of the angle that do not support.To form special angle soundThe effect that sound is reinforced, to realize that the longer-distance sound in the angle of oneself setting is reinforced.
Illustrate it is how to carry out eliminating the sound of interference and be reinforced special angle sound below by diagram.As shown in Figure 2 a, setting needs the sound enrolled at 0 degree, and the sound of intruder is at 50 degree.Then circular diagram and petal shapeIndicate the angle for the reinforcing voice that can be supported and the boosting intensity of live sound: -30 degree are to 30 degree and 60 degree to 120Degree, -150 degree to 150 degree and -60 degree to -120 degree.Object sound in this region will do it amplification, angle in addition to thisDegree sound will do it filtering and delete.It is achieved that the selective recording of the specific region of some or multiple directions.Such as Fig. 2 b instituteShow, setting needs the sound enrolled at 0 degree, and the sound of intruder is at 950 degree.Then circular diagram and petal shape are represented and are supportedStrengthen the degree of angle -60 for recording supported to 60 degree and -120 degree to 120 degree, similarly the sound of this angle can be by phaseThe processing of position and amplitude is to amplify recording, and other regions need to filter or interference-shielded sound, can from petal shapeOut: angle being supported to obtain effective amplification.Other angle sound pressure levels are smaller.
Further, in the present embodiment, when the number of the microphone is 4 comprising: it is arranged in movementIn the present embodiment, the mobile terminal is mobile phone to terminal 1() positive first microphone 10 and second microphone 20, Yi JisheSet the third microphone 30 and the 4th microphone 40 at the mobile terminal back side；As best shown in figures 3 a and 3b.
Be shown experimentally that: the distance between first microphone 10 and second microphone 20 100mm and 150mm itBetween；The distance between the third microphone 30 and the 4th microphone 40 are between 100mm and 150mm；First microphoneThe distance between 10 and third microphone 30 are greater than 18mm；The distance between the second microphone 20 and the 4th microphone 40 are bigEffect is best when 18mm.
The present invention also provides a kind of ways of recording based on multi-microphone array beam forming, as shown in figure 4, the sideMethod the following steps are included:
S1, external sound signal is acquired by several microphones；
S2, corresponding audio digital signals will be individually converted into per the external sound signal of microphone input all the way；
S3, each railway digital voice signal and the pre-set mathematical model established according to beamforming algorithm is enabled to carry outIt compares, filters out corresponding audio digital signals；
Audio digital signals after S4, output screening.
It further, further include step S31 after the step S3, before step S4, by the digital speech after screeningSignal increases intensity of sound by dynamically reinforcing its phase and amplitude, and re-establishes virtual digital space domain and time-domainTo form the enhancement device of special angle sound reinforcement.
Further, corresponding audio digital signals are filtered out in the step S2 to specifically include:
Each railway digital voice signal and the pre-set mathematical model established according to beamforming algorithm are comparedIt is right, the mixed characteristic data of audio digital signals are identified, and according to preset angle threshold as limiting, to digital languageSound signal is screened, and the audio digital signals within the scope of the angle threshold are retained, to the number outside the angle threshold rangeWord voice signal carries out anti-wave inhibition and filtering.
Further, the number of the microphone is 4 comprising: it is arranged in positive first microphone of mobile terminalAnd second microphone, and the third microphone and the 4th microphone at the mobile terminal back side are set；Wherein, first MikeThe distance between wind and second microphone are between 100mm and 150mm；Between the third microphone and the 4th microphone away fromFrom between 100mm and 150mm；The distance between first microphone and third microphone are greater than 18mm；Second wheatThe distance between gram wind and the 4th microphone are greater than 18mm.
The particular technique details of above-mentioned each step has all been described in detail in above system, here just no longerRedundant.
In conclusion the recording system and its implementation provided by the invention based on multi-microphone array beam forming,Wherein, the recording system based on multi-microphone array beam forming includes: several for acquiring external sound signalMicrophone, analog-digital commutator, audio digital signal processing device and output device；The microphone, analog-digital commutator, soundFrequency digital signal processing device and output device are sequentially connected.Recording direction can be changed in real time.And it can only record some angleThe sound of degree is arranged other voice and background noise except this angle and all effectively eliminates.Meanwhile it not needing to pass through yetThe position and direction for changing mobile terminal carry out the deliberately acoustic distance near point by mobile terminal from needs, to realize most 5 metersThe interior effect preferably recorded.
It should be understood that the application of the present invention is not limited to the above for those of ordinary skills canWith improvement or transformation based on the above description, all these modifications and variations all should belong to the guarantor of appended claims of the present inventionProtect range.