CN101079261B - Voice extraction circuit for suppressing environment noise - Google Patents
Voice extraction circuit for suppressing environment noise Download PDFInfo
- Publication number
- CN101079261B CN101079261B CN2007100622599A CN200710062259A CN101079261B CN 101079261 B CN101079261 B CN 101079261B CN 2007100622599 A CN2007100622599 A CN 2007100622599A CN 200710062259 A CN200710062259 A CN 200710062259A CN 101079261 B CN101079261 B CN 101079261B
- Authority
- CN
- China
- Prior art keywords
- circuit
- signal
- amplifier
- control circuit
- differential amplifier
- 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
Images
Abstract
The invention discloses a phonetic extracting circuit to inhibit environmental noise for kinds of communicating products, which comprises the following parts: AF signal gathering circuit, denoise amplifying circuit, control circuit, wherein the middle AF signal gathering circuit adopts individual single-direction microphone as AF sensor, which adds the AF signal gain control circuit among the control circuits of AF signal gathering circuit and denoise amplifying circuit and denoise amplifying circuit; two paths of gathering circuit and AF gain control circuit is fixed with AF signal with two fixed spread, which is transmitted to the denoise amplifying circuit to do time-sharing denoise and amplification; the AF signal with fixed spread is transmitted from the AF gain control circuit to the control circuit at two paths, which realizes the on-line condition control of denoise amplifying circuit through amplifying, comparing conversing and logical disposing and extracting condition.
Description
Technical field
The invention belongs to the high-quality speech communication field.Be a kind of voice extraction circuit that suppresses neighbourhood noise, be applicable to that the stronger occasion of various neighbourhood noises uses as high-level efficiency speech de-noising processing unit.
Background technology
Communication is changed in modern times and has been obtained development at full speed in the society, and the direct interchange of language still is the main mode of modern communication.Come directly actuated electronic equipment also to become the scheme of selecting for use in the automatic control process with voice messaging.So high s/n ratio, the transmission of voice messaging have accurately become crucial scientific and technical means.But ambient noise is inevitably when using, and how to suppress neighbourhood noise, makes the accurate glitch-free transmission of voice messaging become crucial problem.Even the microphone through particular design is also very undesirable when ambient noise level surpasses 70dB.For solving this difficult problem, after the acoustics expert utilizes complex processing software that the voice of sneaking into neighbourhood noise are decomposed, classify, discern, handle by computer technology neighbourhood noise is removed, form artificial intelligence ear technology.Though respond well, technical sophistication, use that supporting technology is loaded down with trivial details, cost is high, at present and be not suitable for the mobile communication industry of normal domestic product and widespread use.
The inventor once designed a kind of simple and easy, otherwise effective technique scheme (seeing the ZL031296888 patent documentation); Two full directional microphones that the certain distance placement is arranged in the space of taking in the document are as the audio signal sample circuit in this circuit structure; The differential amplifier circuit that has the high cmrr of floating ground through employing is formed the denoising amplifying circuit; Adopt the control circuit that extracts valid voice information to control the denoising amplifying circuit simultaneously, obtained obvious effects.Though circuit design is simple, at the various complex environments of inhibition, when comprising the noise in the multiple reflection environment, effect is remarkable.This project organization is simple, with low cost, it is supporting with various voice transfer circuit to be easy to, need not software supporting, thereby have extremely application prospects.
But, be difficult to aborning select as two identical microphones of performance index.Particularly use maximum electret microphones.When concrete supporting debugging, need a microphone wherein receive attach the one deck of subsides sound dampening layer in the sound hole place, and the delay that this damping layer is caused sound, suitable to guarantee the time of transmitting between two microphones.Otherwise each item index of circuit overall design is difficult to reach theoretical value.Cause the discreteness of technical indicator in the bulk article.In addition, because the output of available circuit sound intermediate frequency signal has delay mutually, and voice are a kind of low-frequency composite signals with a plurality of octaves, and the phase differential that identical delay causes on different frequencies is different, and is that low frequency is low, high frequency is high.In 300HZ to the 3.4KHZ scope of concentrating with speech energy, phase differential will differ tens times.Through behind the calculus of differences, the voice signal amplitude of output can differ 100 times, even more.
The actual process of above analytical proof is roughly the same signal " poor " in the difference of carrying out, and can cause very big decay and distortion to the low frequency part of voice.Therefore, the signal to noise ratio (S/N ratio) of circuit output voice signal can degenerate, and shows low-frequency range especially.Further analysis also can prove; It is not accurate enough that this patent design also exists the too coarse voice control signal that causes of voice control signal extraction; With can't adapt to the obvious signal to noise ratio (S/N ratio) that the voice signal changes in amplitude causes and reduce and distorted signals, all be the place that needs further improvement.
Summary of the invention
The objective of the invention is the important improvement that the available circuit structure is made and design again to overcome the defective of prior art; Accomplish a kind of voice that can suppress neighbourhood noise and extract electricity and realize Hi-Fi voice amplifying circuit design, and adaptation and various voice transfer circuit are complementary.
Basic design of the present invention is to comprise the audio signal sample circuit in the structure; The denoising amplifying circuit that comprises high common mode differential amplifier; The control circuit that denoising amplifying circuit state is regulated and control by the efficient voice signal; Crucial improvement is that the audio signal sample circuit has adopted the separate single directional microphone as audio sensor; Between audio signal sample circuit and denoising amplifying circuit and control circuit, set up the sound signal gain control circuit to denoising amplifying circuit duty; The signal of from the audio signal sample circuit, drawing is two-way with original signal and sends into the alternate treatment that denoising and amplification are carried out in the timesharing of denoising amplifying circuit after the quantitative attenuation processing of sound signal gain control circuit; In the sound signal gain control circuit, divide two-way to send into control circuit the sound signal of carrying out respectively after the attenuation processing, through amplify, relatively conversion and logical process are extracted the voice status control signal and are delivered to the denoising amplifying circuit.
Owing to used a uni-directional microphone instead as audio sensor; And after having added the quantitative attenuation processing of gain control circuit; So the two-way audio signal of the high common mode differential amplifier of input denoising amplifying circuit just no longer has the phase place difference that causes because of postponing; And have basicly stable amplitude difference, and help greatly accurately proposing the efficient voice control signal, realize that high-quality voice amplify and noise removing.From but the present invention produces the breakthrough of matter in above-mentioned technical indicator.
Below in conjunction with structured flowchart in the given description of drawings and embodiment electrical schematic diagram, further specify and the present invention seeks to how to realize.
Description of drawings
Fig. 1 is a structured flowchart of the present invention.
Fig. 2 is the electrical schematic diagram of audio signal sample circuit.
Fig. 3 is the electrical schematic diagram of signal gain control circuit (2) embodiment.
Fig. 4 is the electrical schematic diagram of the embodiment of the amplifier (3A1,3B1) in the control circuit (3).
Fig. 5 is the electrical schematic diagram of the embodiment of the pulse converter (3A2,3B2) in the control circuit (3).
Fig. 6 is the electrical schematic diagram of the embodiment of the logic processing circuit (3C) in the control circuit (3).
Fig. 7 is the electrical schematic diagram that the pulsewidth in the control circuit (3) is provided with the embodiment of circuit (3D).
Fig. 8 is the electrical schematic diagram of the embodiment of the control signal change-over circuit (3E) in the control circuit (3).
Fig. 9 is the control circuit (electrical schematic diagram to the embodiment of gain control trigger circuit (3F) in 3.
Figure 10 is the electrical schematic diagram of embodiment of the differential amplifier circuit (4A, 4B) of high cmrr.
Figure 11 is the electrical schematic diagram of the embodiment of summitor (4C) in the denoising amplifying circuit (4).
Figure 12 is the electrical schematic diagram of the embodiment of phase inverter (4D) in the denoising amplifying circuit (4).
Label 1 in the accompanying drawing 1 is represented the audio signal sample circuit, and 2 representation signal gain control circuits, 2A are represented controlled dynamic attenuation circuit, and 2B represents the fixed attenuation circuit; 3 represent control circuit, and 3A1,3B1 represent amplification coefficient, amplifier that the dynamic linear parameter is identical, and 3A2,3B2 representative have the pulse converter of same datum voltage; 3C represents logic processing circuit, and on behalf of pulsewidth, 3D circuit is set, and 3E represents the control signal change-over circuit; 3F represents the gain control trigger circuit, and 3F1 represents comparer, and 3F2 represents pulse producer; 4 represent the denoising amplifying circuit, 4A,, 4B represents the differential amplifier circuit of high cmrr; 4C represents summitor, and 4D represents the duty change-over circuit, and 4E represents phase inverter.
M represents microphone in the accompanying drawing 2, and R1-R26 is a resistance, and C1-C5 is an electric capacity, and U1-U22 represents IC chip.SW1-SW3 is an electronic switch.D1-D5 represents diode.
Embodiment
Can be clearly seen that with reference to accompanying drawing and to comprise audio signal sample circuit 1 in this circuit structure; Denoising amplifying circuit 4; Control circuit 3 to denoising amplifying circuit state; Crucial improvement is that audio signal sample circuit 1 has adopted the separate single directional microphone as audio sensor; Between audio signal sample circuit 1 and denoising amplifying circuit 4 and control circuit 3, set up sound signal gain control circuit 2 to denoising amplifying circuit duty; The signal of drawing from audio signal sample circuit 1 is two-way with original signal and sends into the alternate treatment that denoising and amplification are carried out in 4 timesharing of denoising amplifying circuit after sound signal gain control circuit 2 attenuation processing; Sound signal after sound signal gain control circuit 2 carries out attenuation processing respectively divides two-way to send into control circuit 3, through amplify, relatively conversion and logical process are extracted state control signal to the denoising amplifying circuit.
The secondary attenuator circuit that sound signal gain control circuit 2 is made up of controlled dynamic attenuation circuit 21 and fixed attenuation circuit 22 polyphones; Fixed attenuation circuit 22 output terminals from sound signal gain control circuit 2 are drawn sound signal after the attenuation processing and are sent into and deliver to denoising amplifying circuit 4; The spread signal of drawing from controlled dynamic attenuation circuit 21 and 22 liang of output terminals of fixed attenuation circuit is sent into two input ends of control circuit 3 respectively, and the control trigger pip that comes from control circuit 3 is connected on the trigger end of controlled dynamic attenuation circuit 21.
Comprise two audio signal amplifiers (3A1,3B1) and two pulse converters (3A2,3B2) of contacting respectively in the control circuit 3; The two-way spread signal of after the secondary attenuation processing, drawing from sound signal gain control circuit 2 is sent into the input end of two audio signal amplifiers (3A1,3B1) respectively; The pulse signal of two pulse converters (3A2,3B2) output is delivered to logic processing circuit 3C; The voice pulse signal that is extracted is delivered to pulsewidth circuit 3D is set; The phonetic feature pulse signal that circuit 3D is set through pulsewidth gets into control signal change-over circuit 3E; Send into the inhibition of denoising amplifying circuit 4 timesharing control ground unrest and the alternate treatment that voice amplify by the control signal of this circuit output; Also be provided with the trigger circuit 3F of control audio signal gain control circuit 2 duties in the control circuit 3; This circuit is made up of the pulse producer 3F2 of comparer 3F1 and polyphone, and reference voltage and the signal of drawing from amplifier 3A are got into by comparer 3F1 input end, and control signal causes the trigger end of sound signal gain control circuit 2 from pulse producer 3F2.
The differential amplifier circuit (4A, 4B) that comprises two high cmrrs in the denoising amplifying circuit 4; (4A, 4B) relatively amplifies the two paths of signals of usefulness as differential amplifier circuit; One road a input end from audio signal sample circuit 1 output terminal anti-phase each other introducing differential amplifier circuit (4A, 4B); The inverting input of the correspondence of differential amplifier circuit (4A, 4B) is drawn, is led to respectively at input, the output terminal of anti-phase controller 4E on another road from sound signal gain control circuit 2; Wherein differential amplifier circuit 4B is controlled by duty change-over circuit 4D; Control signal is drawn the corresponding trigger end that is connected on anti-phase controller 4E and duty change-over circuit 4D from the output terminal of control circuit 3; Two output terminals of differential amplifier circuit (4A, 4B) link to each other with two input ends of summitor 4C, and the voice signal after denoising is amplified is drawn from the output terminal of summitor 4C.
Technical scheme after the improvement divides circuit for four major parts.At first be audio signal sample circuit 1, it is the audio sensor that a uni-directional microphone is formed.It is an audio signal sample circuit that possesses high input impedance and low output impedance.The sound signal of coming out from above circuit is divided into two-way; One the tunnel is input to denoising amplifying circuit 4; Another road incoming audio signal gain control circuit 2, be treated to have fixing width of cloth difference but the identical two paths of signals of other parameters gets into two the identical amplifiers of design parameter (3A1,3B1) in the control circuit 3.Must be pointed out that this two paths of signals and the amplitude difference between them are the people for working it out, but accurately extract the key of phonetic feature pulse.We at first analyze its flow process.(3A1 3B1) is identical two amplifiers of parameter designing to amplifier.They are amplified to certain level to the two paths of signals of being imported with the multiple that equates fully respectively, and the linear two paths of signals that keeps is exported the amplitude difference of fashionable formation.(3A2 3B2) is identical two comparers of design to pulse converter.Choosing of its reference voltage is less than from the microphone front and receives the signal level amplitude, and greater than the signal level amplitude of receiving from behind.In addition, require comparer that big linear dynamic range is arranged, become the precision of pulse width variations to guarantee changes in amplitude.The difference of uni-directional microphone output voice, noise signal, when carrying out pulses switch through comparer, the width that the pulse width of voice signal conversion can be changed greater than noise signal out and away.Become and the multi-frequency Combination when considering the amplitude of sound, this must make the variation of commutation pulse width of output bigger.But the setting of fixed attenuation amount is carried out after XOR handles this two-way pulse with regard to establishing through logic processing circuit 3C, just can obtain wanting the fixed pulse width of learning.We can be according to this known pulsewidth; Design the correlation parameter that pulsewidth is provided with window among the circuit 3D; Extract exactly and receive frequency, changes in amplitude to influence and make pulsewidth have phonetic feature pulse, convert control signal removes to control denoising amplifying circuit 4 from control signal change-over circuit 3E output duty then into change.
Accompanying drawing 2-accompanying drawing 12 has provided concrete embodiment electrical schematic diagram.
Audio signal sample circuit 1 is the bleeder circuit of a uni-directional microphone M and resistance R 1 composition and the emitter follower composition that amplifier U1 is barricaded as, and sound signal is taken out through capacitor C 1 and exported through emitter follower.
Controlled dynamic attenuation circuit 21 in sound signal gain control circuit 2 structures is by amplifier U2 and is connected in the resistance (R2 between amplifier U2 input end and the earth point; R3) be barricaded as; Resistance R 2 is connected in parallel on electronic switch SW1 two ends; Electronic switch SW1 control signal is drawn Self Control circuit 3; Fixed attenuation circuit 22 is by amplifier U3 and is connected in resistance between amplifier U3 input end and the earth point (R4 R5) is barricaded as, and the output terminal of amplifier U2 is connected in the secondary attenuator circuit that the input end of amplifier U3 is formed through resistance R 4.
Included two audio signal amplifiers (3A1,3B1) all are by amplifier U4 and supporting electric capacity resistance (R6 in the control circuit 3; R7) amplifying circuit that is barricaded as; Have identical amplification coefficient and dynamic linear parameter; Respectively the polyphone two pulse converters (3A2,3B2) all be element (R8 to be set by amplifier U5 and reference voltage; D1) pulse shaping circuit that pressure limiting circuit that the comparator circuit that is barricaded as, diode D2 and resistance R 9 are formed and Schmidt trigger U6 form is connected in series successively; Logic processing circuit 3C is made up of XOR gate U7; It is by four two input nand gates (U8-U11) and multi resonant oscillator (U12-U13) and supporting Resistor-Capacitor Unit (R10, C2 that pulsewidth is provided with circuit 3D; R11, C3) the pulsewidth threshold circuit that is overlapped to form; The phonetic feature pulse signal of being drawn by Sheffer stroke gate (U11) is delivered to the control signal change-over circuit 3E that is barricaded as by phase inverter U14 and multivibrator U15 and Resistor-Capacitor Unit (R12, C4) thereof, and the control signal of the anti-phase each other of output of changing is divided into two-way and delivers to denoising amplifying circuit 4, and the reference voltage that comparator circuit 3F1 is made up of amplifier U16 and resistance R 13 and diode D3 among the trigger circuit 3F is provided with that circuit is barricaded as; Pulse producer 3F2 is the stretch circuit that is overlapped to form by multivibrator U17 and Resistor-Capacitor Unit (R14, C5), and trigger pip is drawn the signal input end of delivering to sound signal gain control circuit 2 from the output terminal of multivibrator U15.
The differential amplifier circuit of two high cmrrs in the denoising amplifying circuit 4 (4A, 4B) all is to be barricaded as by three amplifiers (U18, U19, U20) and supporting resistive element (R15-R21); The output terminal of two differential amplifier circuits (4A, 4B) is connected in parallel by what amplifier U21 was barricaded as by two substitutional resistances (R22, R23) respectively and penetrates the input end with formula summitor 4C; Anti-phase controller 4E is combined with phase inverter and the parallelly connected electronic switch SW2 that supporting resistance (R24, R25, R26) is barricaded as by amplifier U22; Duty change-over circuit 4D is an electronic switch SW3 who is connected in parallel on level amplifier U20 input, output terminal behind the differential amplifier 4B, and the trigger end of electronic switch SW2 and electronic switch SW3 is connected the trigger pip output terminal of two anti-phases each other of control circuit 3.
The damping capacity of fixed attenuation circuit 22 can not be excessive, the signal that microphone front, decay back is received, less than the signal that the back receives, the signal that front, back are received mix mutually around.Certainly it is too small to decay, to such an extent as to the amplitude difference device too small and mimic channel of the two paths of signals that generates receives the error that drift caused of environmental baselines such as voltage, temperature, air pressure suitable, and causes the erroneous judgement of circuit.When calculating, can find that the two paths of signals amplitude differs more little, voice output is bigger and noise is exported more little.So the damping capacity of fixed attenuator 22 should make one's options between general purpose single directional microphone directive property 10dB 1 decibel of the minimum precision of analog circuitry processes.
The decay index of summing up the controlled dynamic attenuation circuit 21 in the sound signal gain control circuit 2 according to great deal of experiment data is the 4-10 decibel, and the decay index of fixed attenuation circuit 22 is 2-5 decibels.
Patent art be with the voice characteristic pulse control voice enhancing, noise reduction and to the extraction of voice, the last output of voice contains the non-linear conversion circuit.In the time of simply, practicality, produce effects is that talker's voice, intonation, distance all might cause interrupted or distortion.Particularly the talker causes that the variation of signal speech commutation pulse width is fairly obvious when microphone distance or tone intensity change.It is the difficulty that can't overcome in the prior art.In the practical application, also can not change in the posture that requires the teller and the tone.Adopted the signal amplitude stable scheme among the present invention, but non-common automatic gain is technological, poor because it can reduce or eliminate voice and noise amplitude that uni-directional microphone receives, and present technique is reduced or lose noise suppressing function.Because distance or voice Strength Changes are very slow from the angle of time, the present invention adopts the automatic amplitude fading scheme of closed loop very suitable to present technique.The amplitude of decay is decided by among the comparator circuit 3F1 reference voltage that determined by resistance R 13 and diode D3, and reference value is the negative 2.5-3.5 decibel of employed supply voltage.This magnitude of voltage has been set the specified output amplitude of signal for us; When sound signal when amplifier 3A1 output reaches this range value; Obtain control signal output through comparer 3F1 and pulse producer 3F2; Go to control the breakdown of dynamic attenuation device 21 and close, reach the stable purpose of signal amplitude.
Further specify be two pulse converters (3A2,3B2) in control circuit 3 structures all be by amplifier U5 and reference voltage be provided with element (R8, the pressure limiting circuit that the comparator circuit that D1) is barricaded as, diode D2 and resistance R 9 are formed and and the pulse shaping circuit that forms of Schmidt trigger U6 be connected in series successively.Wherein reference voltage be provided with element (R8, D1) decision reference voltage be the negative 12-14db of stabilized voltage supply voltage.
Following Theoretical Calculation data can explain that the present invention has significant technique effect:
General purpose single directional microphone directive property is: 10db; The fixed attenuation amount is: 3dB.The amplitude output signal of microphone is made as: 1; Signal amplitude behind fixed attenuator is: 0.707; Voice signal is output as through control circuit summitor 4C when the efficient voice signal occurs: 1.707; The average noise signal is output as through summitor 4C: 0.292.
The inhibition degree of uni-directional microphone can improve: 20log1.707-20log0.292=15db.
That is to say that the system made from patent art is to the minimum 25db that accomplishes of inhibition degree of neighbourhood noise.According to the transmission theory of present voice acoustics, when voice and noise occur simultaneously,, will produce each other and cover as long as amplitude exists bigger difference between the two, make the obvious influence of people's the imperceptible noise of the sense of hearing.Theoretic index is as long as guarantee that speech energy is greater than noise 10db.The present invention can be considerably beyond this index.
Claims (10)
1. the voice extraction circuit that suppresses neighbourhood noise; Comprise audio signal sample circuit (1) in this circuit structure; Denoising amplifying circuit (4); The control circuit (3) of control denoising amplifying circuit state; It is characterized in that audio signal sample circuit (1) has adopted the separate single directional microphone as audio sensor; Between audio signal sample circuit (1) and denoising amplifying circuit (4) and control circuit (3), set up sound signal gain control circuit (2) to denoising amplifying circuit duty; The signal of drawing from audio signal sample circuit (1) is two-way with original signal and sends into the alternate treatment that denoising and amplification are carried out in denoising amplifying circuit (4) timesharing after sound signal gain control circuit (2) attenuation processing, the sound signal after sound signal gain control circuit (2) carries out attenuation processing respectively divides two-way to send into control circuit (3), through amplify, relatively conversion and logical process are extracted state control signal and are delivered to denoising amplifying circuit (4).
2. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 1; It is characterized in that the secondary attenuator circuit that sound signal gain control circuit (2) is made up of controlled dynamic attenuation circuit (21) and fixed attenuation circuit (22) polyphone; The sound signal that fixed attenuation circuit (22) output terminal from sound signal gain control circuit (2) is drawn after the attenuation processing is sent into denoising amplifying circuit (4); The spread signal of drawing from controlled dynamic attenuation circuit (21) and fixed attenuation circuit (22) two output terminals is sent into two input ends of control circuit (3) respectively, and the control trigger pip that comes from control circuit (3) is connected on the trigger end of controlled dynamic attenuation circuit (21).
3. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 1; It is characterized in that comprising in the control circuit (3) two audio signal amplifiers (3A1,3B1); I.e. first audio signal amplifier (3A1) and second note amplifier (3B1); Two pulse converters (3A2,3B2) of contacting respectively; The two-way spread signal of after the secondary attenuation processing, drawing from sound signal gain control circuit (2) is sent into the input end of two audio signal amplifiers (3A1,3B1) respectively; The pulse signal of two pulse converters (3A2,3B2) output is delivered to logic processing circuit (3C); The voice pulse signal that logic processing circuit (3C) extracts is delivered to pulsewidth circuit (3D) is set; The phonetic feature pulse that circuit (3D) is set through pulsewidth gets into control signal change-over circuit (3E); Send into denoising amplifying circuit (4) timesharing control environment Noise Suppression and the alternate treatment that voice amplify by the control signal of control signal change-over circuit (3E) output, also be provided with the trigger circuit (3F) of control audio signal gain control circuit (2) duty in the control circuit (3), trigger circuit (3F) are made up of the comparer (3F1) and the pulse producer (3F2) of contacting; Reference voltage and the signal of drawing from first audio signal amplifier (3A1) are got into by comparer (3F1) input end, and control signal causes the trigger end of sound signal gain control circuit (2) from pulse producer (3F2).
4. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 1; It is characterized in that comprising in the denoising amplifying circuit (4) first differential amplifier circuit (4A) and second differential amplifier circuit (4B) of two high cmrrs; Relatively amplify the two paths of signals of usefulness as first differential amplifier circuit (4A) and second differential amplifier circuit (4B); One the road introduces an input end of first differential amplifier circuit (4A) and second differential amplifier circuit (4B) respectively from the anti-phase each other of audio signal sample circuit (1) output terminal; The inverting input of the correspondence of first differential amplifier circuit (4A) and second differential amplifier circuit (4B) is drawn, is led to respectively at input, the output terminal of anti-phase controller (4E) on another road from sound signal gain control circuit (2); Second differential amplifier circuit (4B) is to be controlled by duty change-over circuit (4D) simultaneously; Control signal is drawn the corresponding trigger end that is connected on anti-phase controller (4E) and duty change-over circuit (4D) from the output terminal of control circuit (3); Two output terminals of first differential amplifier circuit (4A) and second differential amplifier circuit (4B) link to each other with two input ends of summitor (4C), and the voice signal after denoising is amplified is drawn from the output terminal of summitor (4C).
5. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 1; It is characterized in that audio signal sample circuit (1) is that bleeder circuit and first amplifier (U1) that a uni-directional microphone (M) and resistance R 1 formed formed with the emitter follower that electric capacity (C1) is barricaded as, sound signal is exported through emitter follower through electric capacity (C1) taking-up.
6. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 2; It is characterized in that controlled dynamic attenuation circuit (21) in sound signal gain control circuit (2) structure is to be barricaded as by second amplifier (U2) and second resistance (R2) and the 3rd resistance (R3) that are connected between second amplifier (U2) input end and the earth point; Second resistance (R2) is connected in parallel on first electronic switch (SW1) two ends; First electronic switch (SW1) control signal is drawn Self Control circuit (3); Fixed attenuation circuit (22) is to be barricaded as by the 3rd amplifier (U3) and the 4th resistance (R4) and the 5th resistance (R5) that are connected between the 3rd amplifier (U3) input end and the earth point, and the output terminal of second amplifier (U2) is connected in the secondary attenuator circuit that the input end of the 3rd amplifier (U3) is formed through the 4th resistance (R4).
7. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 3; It is characterized in that included two audio signal amplifiers (3A1,3B1) in the control circuit (3) all are the comparison amplifying circuits that are barricaded as by four high guaily unit (U4) and supporting the 6th resistance (R6) and the 7th resistance (R7); Have identical amplification coefficient and dynamic linear parameter; Respectively the polyphone two pulse converters (3A2,3B2) all be element (R8 to be set by the 5th amplifier (U5) and reference voltage; D1) pulse shaping circuit that pressure limiting circuit that the comparison amplifying circuit that is barricaded as, second diode (D2) and the 9th resistance (R9) are formed and Schmidt trigger (U6) form is connected in series successively; Logic processing circuit (3C) is made up of XOR gate (U7); It is by first Sheffer stroke gate (U8), second Sheffer stroke gate (U9), the 3rd Sheffer stroke gate (U10), the 4th Sheffer stroke gate (U11) and first multivibrator (U12), second multivibrator (U13) and supporting Resistor-Capacitor Unit (R10, C2 that pulsewidth is provided with circuit (3D); R11, C3) the pulsewidth threshold circuit that is overlapped to form; The phonetic feature pulse signal of being drawn by the 4th Sheffer stroke gate (U11) is delivered to the control signal change-over circuit (3E) that is barricaded as by phase inverter (U14) and the 3rd multivibrator (U15) and Resistor-Capacitor Unit (R12, C4) thereof; The control signal of the output of changing is divided into two-way and delivers to denoising amplifying circuit (4); The reference voltage that comparer (3F1) is made up of the 6th amplifier (U16) and the 8th resistance (R13) and the 3rd diode (D3) in the trigger circuit (3F) is provided with that circuit is barricaded as; Pulse producer (3F2) is the stretch circuit that is overlapped to form by the 4th multivibrator (U17) and Resistor-Capacitor Unit (R14, C5), and trigger pip is drawn the signal input end of delivering to sound signal gain control circuit (2) from the output terminal of the 4th multivibrator (U17).
8. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 4; First differential amplifier circuit (4A) of two high cmrrs of its characteristic in denoising amplifying circuit (4) and second differential amplifier circuit (4B) all are to be barricaded as by the 8th amplifier (U18), the 9th amplifier (U19), the tenth amplifier (U20) and supporting resistive element (R15-R21); The output terminal of first differential amplifier circuit (4A) and second differential amplifier circuit (4B) is connected in parallel by what the 11 amplifier (U21) was barricaded as by two substitutional resistances (R22, R23) respectively and penetrates the input end with formula summitor (4C); Anti-phase controller (4E) is to be combined with phase inverter and parallelly connected second electronic switch (SW2) that the tenth resistance (R24), the 11 resistance (R25), the 12 resistance (R26) are barricaded as by the 12 amplifier (U22); Duty change-over circuit (4D) is the 3rd electronic switch (SW3) that is connected in parallel on second differential amplifier circuit (4B) back level the tenth amplifier (U20) input, output terminal, and the trigger end of second electronic switch (SW2) and the 3rd electronic switch (SW3) is connected the trigger pip output terminal of two anti-phases each other that control circuit (3) draws.
9. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 2; The decay index that it is characterized in that the controlled dynamic attenuation circuit (21) in the sound signal gain control circuit (2) is the 4-10 decibel, and the decay index of fixed attenuation circuit (22) is the 2-5 decibel.
10. according to the voice extraction circuit of the said inhibition neighbourhood noise of claim 3; It is characterized in that two pulse converters (3A2,3B2) in control circuit (3) structure all are by the 5th amplifier (U5) and reference voltage element (R8 to be set; D1) pressure limiting circuit that the comparison amplifying circuit that is barricaded as, second diode (D2) and the 9th resistance (R9) are formed and be connected in series successively with the pulse shaping circuit of Schmidt trigger (U6) formation; Wherein reference voltage be provided with element (R8, D1) decision reference voltage be the negative 12-14db of supply voltage.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007100622599A CN101079261B (en) | 2007-07-02 | 2007-07-02 | Voice extraction circuit for suppressing environment noise |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2007100622599A CN101079261B (en) | 2007-07-02 | 2007-07-02 | Voice extraction circuit for suppressing environment noise |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101079261A CN101079261A (en) | 2007-11-28 |
| CN101079261B true CN101079261B (en) | 2012-06-13 |
Family
ID=38906695
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2007100622599A Active CN101079261B (en) | 2007-07-02 | 2007-07-02 | Voice extraction circuit for suppressing environment noise |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101079261B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101667426A (en) * | 2009-09-23 | 2010-03-10 | 中兴通讯股份有限公司 | Device and method for eliminating environmental noise |
| CN103594092A (en) * | 2013-11-25 | 2014-02-19 | 广东欧珀移动通信有限公司 | Single microphone voice noise reduction method and device |
| CN105719656A (en) * | 2014-12-03 | 2016-06-29 | 广州汽车集团股份有限公司 | Vehicle-mounted voice recognition system |
| CN107580397A (en) * | 2017-10-26 | 2018-01-12 | 深圳市银河风云网络系统股份有限公司 | Lamp light control method and lamp dimmer |
| CN113257263B (en) * | 2020-04-10 | 2024-03-12 | 徐忠义 | Self-adaptive suppression circuit for voice environment noise |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2256610Y (en) * | 1996-05-31 | 1997-06-18 | 绍兴市华越通讯设备有限公司 | Background noise silencer for voice receiving system |
| CN1450739A (en) * | 2003-04-21 | 2003-10-22 | 徐忠义 | Phoneme background noise inhibitor |
| CN2682532Y (en) * | 2003-09-16 | 2005-03-02 | 陈修志 | Convenient anti-noise speech recognition electronic installation |
-
2007
- 2007-07-02 CN CN2007100622599A patent/CN101079261B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2256610Y (en) * | 1996-05-31 | 1997-06-18 | 绍兴市华越通讯设备有限公司 | Background noise silencer for voice receiving system |
| CN1450739A (en) * | 2003-04-21 | 2003-10-22 | 徐忠义 | Phoneme background noise inhibitor |
| CN2682532Y (en) * | 2003-09-16 | 2005-03-02 | 陈修志 | Convenient anti-noise speech recognition electronic installation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101079261A (en) | 2007-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101079261B (en) | Voice extraction circuit for suppressing environment noise | |
| CN101996630B (en) | Automatic sound recognition based on binary time frequency unit | |
| US9800734B2 (en) | Echo cancellation | |
| CN102572646A (en) | Denoising method and device implemented in state of listening to music by using headset | |
| CN101207663A (en) | Internet communication device and method for controlling noise thereof | |
| CN102938254A (en) | Voice signal enhancement system and method | |
| Song et al. | Locally optimum detection of signals in a generalized observation model: The known signal case | |
| CN112037806B (en) | Method for detecting wind noise and equipment for detecting wind noise | |
| CN1668058A (en) | Recursive least square difference based subband echo canceller | |
| CN108074581B (en) | Control system for human-computer interaction intelligent terminal | |
| CN106251856A (en) | A kind of environment noise based on mobile terminal eliminates system and method | |
| CN101635873A (en) | Adaptive long-term prediction filter for adaptive whitening | |
| CN102165708A (en) | Signal processing method, signal processing device, and signal processing program | |
| EP0887797A3 (en) | Method, equipment and recording device for suppressing pulsed interference in analogue audio and/or video signals | |
| Ahgren | Acoustic echo cancellation and doubletalk detection using estimated loudspeaker impulse responses | |
| CN103312911B (en) | Data processing method and electronic terminal | |
| CN112349291A (en) | Sound amplification system and method based on AI noise reduction model | |
| CN102778622A (en) | Noise-contained small-signal detecting circuit based on non-saturation stochastic resonance mathematical model | |
| CN108109631A (en) | A kind of small size dual microphone voice collecting noise reduction module and its noise-reduction method | |
| CN103124386A (en) | De-noising, echo-eliminating and acute directional microphone for long-distance speech | |
| CN201956058U (en) | Speed signal processing circuit, device and television | |
| CN205693744U (en) | There is the communicator of speaker decrease of noise functions | |
| Yeom et al. | DSP implementation of sound source localization with gain control | |
| Du et al. | An energy-efficient spike encoding circuit for speech edge detection | |
| Sällberg et al. | Speech enhancement implementations in the digital, analog, and hybrid domain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |