CN103761964B - Automatic speech recognition circuit - Google Patents
Automatic speech recognition circuit Download PDFInfo
- Publication number
- CN103761964B CN103761964B CN201410019227.0A CN201410019227A CN103761964B CN 103761964 B CN103761964 B CN 103761964B CN 201410019227 A CN201410019227 A CN 201410019227A CN 103761964 B CN103761964 B CN 103761964B
- Authority
- CN
- China
- Prior art keywords
- field effect
- effect transistor
- circuit
- thyrite
- voltage
- 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
- Amplifiers (AREA)
Abstract
The invention discloses a kind of automatic speech recognition circuit, it comprises variable-gain amplification circuit, comparative voltage generative circuit and signal deteching circuit, variable-gain amplification circuit receives the voice signal of outside input and amplifies and export the voice signal after amplifying, comparative voltage generative circuit generates a reference voltage and is loaded on signal deteching circuit, signal deteching circuit detects the deviation of amplitude and the setting amplitude comparing the voice signal of output, when the amplitude of the voice signal that variable-gain amplification circuit exports departs from setting amplitude, signal deteching circuit feeds back the voltage of a change to variable-gain amplification circuit, and variable-gain amplification circuit adjusts the amplitude that it exports voice signal according to the voltage of this change, the amplitude of the voice signal exported to make variable-gain amplification circuit is identical with setting amplitude.Automatic speech recognition circuit of the present invention all can export the stable voice signal that an amplitude is identical with setting amplitude, improve the accuracy of identification to the voice signal received after receiving the voice signal of outside input.
Description
Technical field
The present invention relates to field of speech recognition, relate more specifically to a kind of automatic speech recognition circuit.
Background technology
When utilizing voice recognition chip to carry out speech recognition, be Protean as the position of the source of sound of signal source from the distance of voice recognition chip, and conventional speech recognition system exports the amplification that the voice signal of input is fixed multiple; Like this for adopting the speech recognition system of fixing enlargement factor just to cause as following problems: when source of sound position from speech recognition system very close to time, the voice signal amplitude detected due to voice recognition chip is very large, after the amplifier of fixing multiple, the slicing exporting voice signal will be caused; But when the position of source of sound is far from speech recognition system, because the amplitude of voice signal forms index decreased along with the increase of distance, the signal amplitude after making the amplifier of fixing enlargement factor amplify is still very little, thus is difficult to reach designing requirement.Therefore, above either way can be serious the accuracy of identification affecting speech recognition system.
Therefore, be necessary to provide a kind of speech recognition system of improvement to overcome above-mentioned defect.
Summary of the invention
The object of this invention is to provide a kind of automatic speech recognition circuit, this circuit carries out identifying processing to voice signal, and the voice signal within the scope of certain amplitude is by after automatic speech recognition circuit reception of the present invention, all can export an amplitude stable voice signal identical with setting amplitude, improve the accuracy of identification to the voice signal received.
The object of this invention is to provide a kind of automatic speech recognition circuit, for regulating the output amplitude of voice signal, it comprises variable-gain amplification circuit, comparative voltage generative circuit and signal deteching circuit, and described variable-gain amplification circuit comprises the first adjustable resistance electronic circuit, second adjustable resistance electronic circuit and amplifier, described first adjustable resistance electronic circuit has identical architectural feature with the second adjustable resistance electronic circuit, described first adjustable resistance electronic circuit is equivalent to the first thyrite and the second thyrite, one end of described first thyrite is connected with the output terminal of described amplifier, the other end is connected with the reverse input end of amplifier, the output terminal of described signal deteching circuit, external power source is all connected with the control end of described first thyrite and the second thyrite, one end of described second thyrite is connected with common mode voltage end, the other end is connected with the positive input of described amplifier, described second adjustable resistance electronic circuit receives the voice signal of outside input, and the voice signal received is inputted described amplifier, the voice signal of described amplifier to input amplifies and exports the voice signal after amplification, described comparative voltage generative circuit is connected with described signal deteching circuit, to generate a reference voltage and to be loaded on described signal deteching circuit, and described reference voltage sets according to the setting amplitude exporting voice signal, the input end of described signal deteching circuit is connected with the output terminal of amplifier, described signal deteching circuit detects the amplitude of the voice signal that described amplifier exports, and the deviation of the amplitude of the voice signal of more described output and setting amplitude, when the amplitude of the voice signal that described amplifier exports departs from setting amplitude, the output terminal of described signal deteching circuit feeds back the voltage of a change to described first adjustable resistance electronic circuit, change the control voltage of described first thyrite and the second thyrite, regulate the gain factor of described amplifier, the amplitude of the voice signal exported to make described amplifier is identical with setting amplitude.
Preferably, one end of first thyrite of the described second adjustable resistance electronic circuit of external sound signal input, the other end of the first thyrite of described second adjustable resistance electronic circuit is connected with the reverse input end of described amplifier, one end of second thyrite of the second adjustable resistance electronic circuit is connected with common mode voltage end, the other end is connected with the positive input of amplifier, and described first thyrite of the second adjustable resistance electronic circuit is all connected with external power source and common mode voltage end with the control end of the second thyrite.
Preferably, described first adjustable resistance electronic circuit comprises the first field effect transistor, the second field effect transistor, the 3rd field effect transistor and the 4th field effect transistor, described first field effect transistor is connected jointly with the grid of the 4th field effect transistor, and is equivalent to a control end of the first thyrite and the second thyrite; Described second field effect transistor is connected jointly with the grid of the 3rd field effect transistor, and is equivalent to another control end of the first thyrite and the second thyrite; Described first field effect transistor is connected jointly with the drain electrode of the second field effect transistor, and is equivalent to one end of the first thyrite; Described 3rd field effect transistor is connected jointly with the drain electrode of the 4th field effect transistor, and is equivalent to one end of the second thyrite; Described first field effect transistor is connected jointly with the source electrode of the second field effect transistor, and is equivalent to the other end of the first thyrite; Described 3rd field effect transistor is connected jointly with the source electrode of the 4th field effect transistor, and is equivalent to the other end of the second thyrite.
Preferably, described first thyrite of the first adjustable resistance electronic circuit and the control voltage of the second thyrite are the difference of supply voltage and described signal deteching circuit output end voltage, and described first thyrite of the second adjustable resistance electronic circuit and the control voltage of the second thyrite are the difference of supply voltage and common mode voltage.
Preferably, field effect transistor described in each has identical architectural feature, and is all operated in triode region.
Preferably, described signal deteching circuit comprises the 5th field effect transistor, 6th field effect transistor, 7th field effect transistor, 8th field effect transistor, 9th field effect transistor, 3rd resistance and electric capacity, the grid of described 5th field effect transistor is connected with the output terminal of described amplifier, and its source electrode is connected with described comparative voltage generative circuit, the drain electrode of described 5th field effect transistor and the drain electrode of described 6th field effect transistor, the grid of grid and the 7th field effect transistor connects jointly, described 6th field effect transistor, the source electrode of the 7th field effect transistor and the 8th field effect transistor is all connected with external power source, the drain electrode of described 7th field effect transistor, one end of 3rd resistance, one end of electric capacity and the grid of described 9th field effect transistor connect jointly, and the drain electrode of described 8th field effect transistor is connected with the drain electrode of described 9th field effect transistor and forms the output terminal of described signal deteching circuit, the source electrode of described 9th field effect transistor, the grid of the 8th field effect transistor, the other end of described 3rd resistance and the equal ground connection of the other end of described electric capacity.
Preferably, described comparative voltage generative circuit comprises the 4th resistance, the 5th resistance and voltage follower, described 4th resistance one end is connected with external power source, the other end is connected with described one end of 5th resistance and the positive input of described voltage follower, the other end ground connection of described 5th resistance, the reverse input end of described voltage follower is connected with its output terminal, and the output terminal of described voltage follower is connected with described signal deteching circuit.
Compared with prior art, automatic speech recognition circuit of the present invention, because described signal deteching circuit detects the amplitude of the voice signal that described amplifier exports, and the deviation of the amplitude of the voice signal of more described output and setting amplitude, when the amplitude of the voice signal that described amplifier exports departs from setting amplitude, described signal deteching circuit feeds back the voltage of a change to described first adjustable resistance electronic circuit, and described first adjustable resistance electronic circuit adjusts its impedance exported according to the voltage of this change, described amplifier is made to regulate it to export the amplitude of voice signal, thus the external sound signal within the scope of certain amplitude is by after automatic increasing regulating circuit reception of the present invention, after the amplification of described variable-gain amplification circuit and the regulating and controlling of described signal deteching circuit, all can export an amplitude stable voice signal identical with setting amplitude over time, improve the accuracy of identification to voice signal.
By following description also by reference to the accompanying drawings, the present invention will become more clear, and these accompanying drawings are for explaining the present invention.
Accompanying drawing explanation
Fig. 1 is the circuit structure diagram of automatic speech recognition circuit of the present invention.
Fig. 2 is the concrete structure figure of the first adjustable resistance electronic circuit of automatic speech recognition circuit of the present invention.
Embodiment
With reference now to accompanying drawing, describe embodiments of the invention, element numbers similar in accompanying drawing represents similar element.As mentioned above, the invention provides a kind of automatic speech recognition circuit, this circuit carries out identifying processing to voice signal, and the voice signal within the scope of certain amplitude is by after automatic speech recognition circuit reception of the present invention, all can export an amplitude stable voice signal identical with setting amplitude, improve the accuracy of identification to the voice signal received.
Please refer to Fig. 1, Fig. 1 is the circuit structure diagram of automatic speech recognition circuit of the present invention.Automatic speech recognition circuit of the present invention is for the output amplitude of the voice signal that regulates it to receive, and the setting amplitude of voice signal that setting automatic speech recognition circuit stability of the present invention exports is Vm.As shown in Figure 1, automatic speech recognition circuit of the present invention comprises variable-gain amplification circuit, comparative voltage generative circuit and signal deteching circuit.And described variable-gain amplification circuit comprises the first adjustable resistance electronic circuit, the second adjustable resistance electronic circuit and amplifier OP1.In a preferred embodiment of the invention, described first adjustable resistance electronic circuit has identical architectural feature with the second adjustable resistance electronic circuit, wherein, described first resistance electronic circuit is equivalent to the first thyrite R1 and the second thyrite R2, described first thyrite R1 one end V1 is connected with the output terminal of amplifier OP1, other end V3 is connected with the reverse input end of described amplifier OP1, thus described first thyrite R1 forms the feedback resistance of described amplifier OP1, its resistance value is by the gain factor of the described amplifier OP1 of impact, the output terminal of described signal deteching circuit is all connected with the control end of described first thyrite R1 and the second thyrite R2 with external power source VDD, wherein, the control end Vg2 of the output terminal output voltage Vg to described first thyrite R1 and the second thyrite R2 of described signal deteching circuit, external power source VDD is connected with the control end Vg1 of described first thyrite R1 and the second thyrite R2, and in a preferred embodiment of the invention, the difference of the output voltage Vg of described external power source VDD and signal deteching circuit forms the control voltage of described first thyrite R1 and the second thyrite R2, one end of described second thyrite R2 is connected with common mode voltage end VCM, and the other end is connected with the positive input of described amplifier OP1.Described second adjustable resistance electronic circuit receives the voice signal Vin of outside input, and the voice signal received is inputted described amplifier OP1, and the voice signal Vin of described amplifier OP1 to input amplifies and export the voice signal Vout after amplification.Described comparative voltage generative circuit is connected with described signal deteching circuit, is loaded on described signal deteching circuit to generate a reference voltage V c, and described reference voltage V c sets according to the setting amplitude Vm exporting voice signal, the input end of described signal deteching circuit is connected with the output terminal of described amplifier OP1, the voice signal Vout that described amplifier OP1 is exported inputs described signal deteching circuit, the output terminal of described signal deteching circuit is connected with the first thyrite R1 of described first adjustable resistance electronic circuit and the control end Vg2 of the second thyrite R2, to export its voltage Vg to described first thyrite R1 and the second thyrite R2, described signal deteching circuit detects the amplitude of the voice signal Vout that described amplifier OP1 exports and the deviation of setting amplitude Vm, when the amplitude of the voice signal Vout that described amplifier exports departs from setting amplitude Vm, described signal deteching circuit feeds back the voltage Vg of a change to described first adjustable resistance electronic circuit, and described first adjustable resistance electronic circuit adjusts its resistance value exported and adjusts the amplitude of the voice signal Vout that described amplifier OP1 exports according to the voltage of this change, the amplitude of the voice signal Vout exported to make described amplifier OP1 is identical with setting amplitude Vm.
Particularly, please combine again with reference to figure 2.In a preferred embodiment of the invention, described first adjustable resistance electronic circuit has identical architectural feature with the second adjustable resistance electronic circuit, that is to say that described second adjustable resistance electronic circuit is also equivalent to the first thyrite R1 ' and the second thyrite R2 '; External sound signal Vin inputs one end of described first thyrite R1 ', the other end of described first thyrite R1 ' is connected with the reverse input end of described amplifier OP1, one end of described second thyrite R2 ' is connected with common mode voltage end VCM, and the other end is connected with the positive input of amplifier OP1; And described first thyrite R1 ' is connected with external power source VDD with a control end Vg1 ' of the second thyrite R2 ', described first thyrite R1 ' is connected with common mode voltage end with another control end Vg2 ' of the second thyrite R2 '.Wherein, the concrete structure of described first adjustable resistance electronic circuit and the second adjustable resistance electronic circuit as shown in Figure 2; The architectural feature of described first adjustable resistance electronic circuit is only introduced at this, it comprises the first field effect transistor M1, the second field effect transistor M2, the 3rd field effect transistor M3 and the 4th field effect transistor M4, described first field effect transistor M1 is connected jointly with the grid of the 4th field effect transistor M4, and is equivalent to a control end Vg1 of the first thyrite R1 and the second thyrite R2; Described second field effect transistor M2 is connected jointly with the grid of the 3rd field effect transistor M3, and is equivalent to another control end Vg2 of the first thyrite R1 and the second thyrite R2; Described first field effect transistor M1 is connected jointly with the drain electrode of the second field effect transistor M2, and is equivalent to one end V1 of the first thyrite R1; Described 3rd field effect transistor M3 is connected jointly with the drain electrode of the 4th field effect transistor M4, and is equivalent to one end V2 of the second thyrite R2; Described first field effect transistor M1 is connected jointly with the source electrode of the second field effect transistor M2, and is equivalent to the other end V3 of the first thyrite R1; Described 3rd field effect transistor M3 is connected jointly with the source electrode of the 4th field effect transistor M4, and is equivalent to the other end V4 of the second thyrite R2; And field effect transistor described in each has identical architectural feature, be all operated in triode region (that is, the drain-source voltage of field effect transistor is less than the difference of gate source voltage and threshold voltage).ID1, ID2, ID3, ID4 shown in Fig. 2 is respectively the drain-source current flowing through described first field effect transistor M1, the second field effect transistor M2, the 3rd field effect transistor M3 and the 4th field effect transistor M4, and I1=ID1+ID3, I2=ID2+ID4, and in actual applications, four field effect transistor all should be operated in triode region, and namely the drain-source voltage of four field effect transistor is less than the difference of gate source voltage and threshold voltage.Wherein, the electric current flowing through each field effect transistor is respectively:
Wherein μ
n, c
oxwith V
tHNbe respectively the carrier mobility of field effect transistor, gate oxide unit-area capacitance amount and threshold voltage, W/L is that the breadth length ratio of each field effect transistor is (because each field-effect tube structure feature is all identical, therefore the breadth length ratio of M1 ~ M4 is also identical), V1-V4 corresponds to the magnitude of voltage of end points V1-V4, and (namely various middle V1 represents the voltage of end points V1 above, other is similar), Vg1 corresponds to the magnitude of voltage of end points Vg1, and Vg2 corresponds to the magnitude of voltage of end points Vg2.Can be derived by (1) ~ (4) formula and draw:
Be there is by the known first adjustable resistance electronic circuit of (5) formula the character of resistance, after manufacturing process is determined, μ
n, c
oxdetermine and constant, then the size of its equiva lent impedance is determined by the ratio and Vg1-Vg2 regulating breadth length ratio W/L.According to port network theory, the first adjustable resistance electronic circuit after conversion can be represented by the right figure in Fig. 2, namely connects the first thyrite R1 between port V1 and V3, connects the second thyrite R2 between V2 and V4, and the resistance of R1 and R2 equals (5) formula, namely
Described second other architectural feature of adjustable resistance electronic circuit and function are all identical with the adjustable electronic circuit of described first resistance, do not repeat them here.
Described signal deteching circuit comprises the 5th field effect transistor M5, the 6th field effect transistor M6, the 7th field effect transistor M7, the 8th field effect transistor M8, the 9th field effect transistor M9, the 3rd resistance R3 and electric capacity C1, the grid of described 5th field effect transistor M5 is connected with the output terminal of described amplifier OP1, its source electrode is connected with described comparative voltage generative circuit, thus the voltage of the output signal Vout of described amplifier OP1 inputs described signal deteching circuit by described 5th field effect transistor M5, the drain electrode of described 5th field effect transistor M5 and the drain electrode of described 6th field effect transistor M6, the grid of grid and the 7th field effect transistor M7 connects jointly, described 6th field effect transistor M6, the source electrode of the 7th field effect transistor M7 and the 8th field effect transistor M8 is all connected with external power source VDD, the drain electrode of described 7th field effect transistor M7, one end of 3rd resistance R3, one end of electric capacity C1 and the grid of described 9th field effect transistor M9 connect jointly, described 6th field effect transistor M6 and the 7th field effect transistor M7 forms mirror image circuit, thus by flow through described 5th field effect transistor M5 electric current I ds1 in proportion mirror image be the electric current I ds2 flowing through described 7th field effect transistor M7, the drain electrode of described 8th field effect transistor M8 is connected with the drain electrode of described 9th field effect transistor M9 and forms the output terminal of described signal deteching circuit, and output voltage Vg, and the change according to voltage Vout value changes by the magnitude of voltage of described voltage Vg, the source electrode of described 9th field effect transistor M9, the grid of the 8th field effect transistor M8, the other end of described 3rd resistance R3 and the equal ground connection of the other end of described electric capacity C1.
Described comparative voltage generative circuit comprises the 4th resistance R4, the 5th resistance R5 and voltage follower OP2, described 4th resistance R4 one end is connected with external power source VDD, the other end is connected with one end of described 5th resistance R5 and the positive input of described voltage follower OP2, the other end ground connection of described 5th resistance R5, namely described 4th resistance R4 and the 5th resistance R5 carries out dividing potential drop to supply voltage VDD, and obtains the voltage Vcomp of the positive input of described follower OP2; Thus when being set in circuit stability work according to circuit design requirements, export the setting amplitude Vm of voice signal, and according to this setting amplitude Vm, select the 4th of suitable resistance the resistance R4 and the 5th resistance R5 and make the voltage Vcomp of the positive input of described voltage follower OP2 may correspond to setting amplitude Vm in exporting voice signal; The reverse input end of described voltage follower OP2 is connected with its output terminal, and the output terminal of described voltage follower OP2 is connected with described signal deteching circuit, be connected with the source electrode of described 5th field effect transistor M5 particularly, the voltage Vc that described voltage follower OP2 is exported can be used as reference voltage for described signal deteching circuit reference, and Vc=Vcomp, thus described voltage Vc corresponds to setting amplitude Vm; In addition, from the circuit structure of automatic speech recognition circuit of the present invention, when circuit normally works, Vout=V
gs5+ V
c; And when circuit of the present invention is in steady-working state, when namely the amplitude of described variable-gain amplification circuit output voice signal Vout is setting amplitude Vm, Vout=V
m=V
gs5+ V
c, (V
gs5be the gate source voltage of the 5th field effect transistor M5), and in circuit design, make described 5th field effect transistor M5 just be operated in subthreshold value state (how to make field effect transistor be operated in subthreshold value state to be well known to those skilled in the art, no longer carefully state at this), thus when its conducting V
gs5≈ V
th5(V
th5be the threshold voltage of the 5th field effect transistor M5), then Vout=V
m=V
th5+ V
cthus the input voltage Vcomp(also i.e. Vc by setting described comparative voltage generative circuit) value, when the voice signal Vin inputted is when setting in adjustable extent, the amplitude of described output voice signal Vout can be made identical with setting amplitude Vm after adjustment after a while, to ensure the accuracy of identification to input speech signal.
Principle of work below in conjunction with describing automatic speech recognition circuit of the present invention with reference to figure 1 and Fig. 2:
The variable-gain amplification circuit of automatic speech recognition circuit of the present invention, because each thyrite architectural feature is identical, make R1=R2, R1 '=R2 ', thus ensure that the impedance being carried in described amplifier OP1 Direct/Reverse input end is equal, described amplifier OP1 imbalance in the course of the work can be reduced, to make the output of described amplifier OP1 more accurate; Described second adjustable resistance electronic circuit determines the input resistance of variable-gain amplification circuit, and its equivalent resistance R1 ' and R2 ' size are controlled by the difference of supply voltage VDD and common mode voltage VCM; First adjustable resistance electronic circuit determines the feedback resistance of described adjustable gain amplifying circuit, and its equivalent resistance size R1 and R2 is controlled by supply voltage VDD and signal deteching circuit output voltage Vg difference.Amplifier exports Vout:
Obtain in conjunction with (6) formula,
Wherein (W/L)
1be the breadth length ratio ratio of the first adjustable resistance electronic circuit four field effect transistor pipes, (W/L)
2be the breadth length ratio ratio of the second adjustable resistance electronic circuit four field effect transistor.In circuit design, when
after ratio is selected, the amplitude of the adjustable described output voice signal Vout of change of signal deteching circuit output voltage Vg.
The course of work below in conjunction with describing automatic speech recognition circuit of the present invention with reference to figure 1 and Fig. 2:
Automatic speech recognition circuit be to input speech signal Vin in certain limit through automatic gain process, make it output signal Vout amplitude and equal setting amplitude Vm.But when input speech signal Vin amplitude is very little, after variable gain amplifier, the amplitude exporting voice signal Vout is also very little, make its voltage also very little, now the 5th field effect transistor M5 does not open, and signal deteching circuit does not work, namely as Vout < V
cutoff+ V
c(V
cutoffbe the cut-in voltage of the 5th field effect transistor M5), signal deteching circuit does not work, also namely can not the amplitude of regulation output voice signal Vout to setting amplitude Vm; When automatic speech recognition circuit stability work of the present invention, also namely input speech signal is after automatic speech recognition circuit, and the amplitude exporting voice signal Vout is Vm, i.e. Vout=V
m=V
th5+ V
c(V
th5> V
cutoff).When exporting voice signal changes in amplitude, detecting feedback regulation by the signal deteching circuit of a period of time, making output voice signal Vout voltage meet Vout=V
th5+ V
c, finally reach steady-working state.
During automatic speech recognition circuit working, when external sound signal Vin is after variable-gain amplification circuit, if when the amplitude exporting voice signal Vout is greater than setting amplitude Vm, i.e. Vout>=V
m=V
th5+ V
c, make the electric current I ds1 flowing through described 5th field effect transistor M5 become large, meanwhile, the electric current flowing through described 6th field effect transistor M6 is also Ids1, the electric current of described 7th field effect transistor M7 the 6th field effect transistor M6 described in mirror image by a certain percentage, make the electric current I ds2 flowing through described 7th field effect transistor M7 also become large thereupon, now electric current I ds2 charges to electric capacity C1, 9th field effect transistor M9 grid voltage is raised, because the 8th field effect transistor M8 is PMOS, and grid connects ground voltage all the time, 8th field effect transistor M8 is in conducting state always, so the 9th field effect transistor M9 grid voltage raises, output voltage Vg voltage is reduced, reduced by the known Vout of (7) (8) formula, thus realize negative feedback process, thus make Vout finally be tending towards Vth2+Vc, so, after adjustment after a while, the amplitude of the voice signal Vout that the output terminal of described amplifier OP1 can be made to export falls after rise to setting amplitude Vm, make stable output signal.
In like manner, when external sound signal Vin is after variable gain amplifier, if when the amplitude exporting voice signal Vout is less than setting amplitude Vm, i.e. V
cutoff+ V
c≤ Vout≤V
m=V
th5+ V
c, the electric current I ds1 flowing through described 5th field effect transistor M5 is diminished, and meanwhile, the electric current flowing through described 6th field effect transistor M6 is also Ids1, the electric current of described 7th field effect transistor M7 the 6th field effect transistor M6 described in mirror image by a certain percentage, the electric current I ds2 flowing through described 7th field effect transistor M7 is also diminished thereupon, now electric capacity C1 is discharged by described 3rd resistance R3, 9th field effect transistor M9 grid voltage is reduced, because the 8th field effect transistor M8 is PMOS, and grid connects ground voltage all the time, 8th field effect transistor M8 is in conducting state always, so the 9th field effect transistor M9 grid voltage reduces, output voltage Vg voltage is raised, become large by the known Vout of (7) (8) formula, thus realize negative feedback process, thus make Vout finally be tending towards Vth2+Vc, so, after adjustment after a while, the amplitude of the voice signal Vout that the output terminal of described amplifier OP1 can be made to export rises to the amplitude Vm of setting, makes stable output signal.
More than in conjunction with most preferred embodiment, invention has been described, but the present invention is not limited to the embodiment of above announcement, and should contain various carry out according to essence of the present invention amendment, equivalent combinations.
Claims (7)
1. an automatic speech recognition circuit, for regulating the output amplitude of voice signal, it is characterized in that, comprise variable-gain amplification circuit, comparative voltage generative circuit and signal deteching circuit, and described variable-gain amplification circuit comprises the first adjustable resistance electronic circuit, second adjustable resistance electronic circuit and amplifier, described first adjustable resistance electronic circuit has identical architectural feature with the second adjustable resistance electronic circuit, described first adjustable resistance electronic circuit is equivalent to the first thyrite and the second thyrite, one end of described first thyrite is connected with the output terminal of described amplifier, the other end is connected with the reverse input end of amplifier, the output terminal of described signal deteching circuit, external power source is all connected with the control end of described first thyrite and the second thyrite, one end of described second thyrite is connected with common mode voltage end, the other end is connected with the positive input of described amplifier, described second adjustable resistance electronic circuit receives the voice signal of outside input, and the voice signal received is inputted described amplifier, the voice signal of described amplifier to input amplifies and exports the voice signal after amplification, described comparative voltage generative circuit is connected with described signal deteching circuit, to generate a reference voltage and to be loaded on described signal deteching circuit, and described reference voltage sets according to the setting amplitude exporting voice signal, the input end of described signal deteching circuit is connected with the output terminal of amplifier, described signal deteching circuit detects the amplitude of the voice signal that described amplifier exports, and the deviation of the amplitude of the voice signal of more described output and setting amplitude, when the amplitude of the voice signal that described amplifier exports departs from setting amplitude, the output terminal of described signal deteching circuit feeds back the voltage of a change to described first adjustable resistance electronic circuit, change the control voltage of described first thyrite and the second thyrite, regulate the gain factor of described amplifier, the amplitude of the voice signal exported to make described amplifier is identical with setting amplitude.
2. automatic speech recognition circuit as claimed in claim 1, it is characterized in that, one end of first thyrite of the described second adjustable resistance electronic circuit of external sound signal input, the other end of the first thyrite of described second adjustable resistance electronic circuit is connected with the reverse input end of described amplifier, one end of second thyrite of the second adjustable resistance electronic circuit is connected with common mode voltage end, the other end is connected with the positive input of amplifier, and described first thyrite of the second adjustable resistance electronic circuit is all connected with external power source and common mode voltage end with the control end of the second thyrite.
3. automatic speech recognition circuit as claimed in claim 2, it is characterized in that, described first adjustable resistance electronic circuit comprises the first field effect transistor, the second field effect transistor, the 3rd field effect transistor and the 4th field effect transistor, described first field effect transistor is connected jointly with the grid of the 4th field effect transistor, and is equivalent to a control end of the first thyrite and the second thyrite; Described second field effect transistor is connected jointly with the grid of the 3rd field effect transistor, and is equivalent to another control end of the first thyrite and the second thyrite; Described first field effect transistor is connected jointly with the drain electrode of the second field effect transistor, and is equivalent to one end of the first thyrite; Described 3rd field effect transistor is connected jointly with the drain electrode of the 4th field effect transistor, and is equivalent to one end of the second thyrite; Described first field effect transistor is connected jointly with the source electrode of the second field effect transistor, and is equivalent to the other end of the first thyrite; Described 3rd field effect transistor is connected jointly with the source electrode of the 4th field effect transistor, and is equivalent to the other end of the second thyrite.
4. automatic speech recognition circuit as claimed in claim 3, it is characterized in that, described first thyrite of the first adjustable resistance electronic circuit and the control voltage of the second thyrite are the difference of supply voltage and described signal deteching circuit output end voltage, and described first thyrite of the second adjustable resistance electronic circuit and the control voltage of the second thyrite are the difference of supply voltage and common mode voltage.
5. automatic speech recognition circuit as claimed in claim 4, it is characterized in that, field effect transistor described in each has identical architectural feature, and is all operated in triode region.
6. automatic speech recognition circuit as claimed in claim 1, it is characterized in that, described signal deteching circuit comprises the 5th field effect transistor, 6th field effect transistor, 7th field effect transistor, 8th field effect transistor, 9th field effect transistor, 3rd resistance and electric capacity, the grid of described 5th field effect transistor is connected with the output terminal of described amplifier, and its source electrode is connected with described comparative voltage generative circuit, the drain electrode of described 5th field effect transistor and the drain electrode of described 6th field effect transistor, the grid of grid and the 7th field effect transistor connects jointly, described 6th field effect transistor, the source electrode of the 7th field effect transistor and the 8th field effect transistor is all connected with external power source, the drain electrode of described 7th field effect transistor, one end of 3rd resistance, one end of electric capacity and the grid of described 9th field effect transistor connect jointly, and the drain electrode of described 8th field effect transistor is connected with the drain electrode of described 9th field effect transistor and forms the output terminal of described signal deteching circuit, the source electrode of described 9th field effect transistor, the grid of the 8th field effect transistor, the other end of described 3rd resistance and the equal ground connection of the other end of described electric capacity.
7. automatic speech recognition circuit as claimed in claim 1, it is characterized in that, described comparative voltage generative circuit comprises the 4th resistance, the 5th resistance and voltage follower, described 4th resistance one end is connected with external power source, the other end is connected with described one end of 5th resistance and the positive input of described voltage follower, the other end ground connection of described 5th resistance, the reverse input end of described voltage follower is connected with its output terminal, and the output terminal of described voltage follower is connected with described signal deteching circuit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410019227.0A CN103761964B (en) | 2014-01-16 | 2014-01-16 | Automatic speech recognition circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410019227.0A CN103761964B (en) | 2014-01-16 | 2014-01-16 | Automatic speech recognition circuit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103761964A CN103761964A (en) | 2014-04-30 |
| CN103761964B true CN103761964B (en) | 2016-04-13 |
Family
ID=50529192
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410019227.0A Active CN103761964B (en) | 2014-01-16 | 2014-01-16 | Automatic speech recognition circuit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103761964B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111490739B (en) * | 2019-01-25 | 2023-09-26 | 杭州海康威视数字技术股份有限公司 | Device and equipment for amplifying audio signal |
| CN112929790B (en) * | 2021-01-20 | 2022-02-01 | 四川湖山电器股份有限公司 | Single-transmitting double-receiving infrared microphone volume self-adaptive circuit |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201078843Y (en) * | 2007-06-05 | 2008-06-25 | 广州信锐电子科技有限公司 | Sine wave signal fixed amplitude and fixed voltage converting circuit with high precision, broad band and low aberration |
| CN101399908A (en) * | 2008-10-13 | 2009-04-01 | 深圳创维－Rgb电子有限公司 | Circuit for automatically regulating amplitude of video output signal of television |
| CN102201792A (en) * | 2010-03-25 | 2011-09-28 | 上海沙丘微电子有限公司 | Automatic gain control circuit of audio power amplifier |
| CN202334444U (en) * | 2011-10-31 | 2012-07-11 | 四川和芯微电子股份有限公司 | Adjustable-gain audio power amplifying circuit |
| CN203746419U (en) * | 2014-01-16 | 2014-07-30 | 四川和芯微电子股份有限公司 | Automatic voice identification circuit |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100547840B1 (en) * | 2003-06-04 | 2006-01-31 | 삼성전자주식회사 | Automatic Gain Control with Fast Settling Time |
-
2014
- 2014-01-16 CN CN201410019227.0A patent/CN103761964B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN201078843Y (en) * | 2007-06-05 | 2008-06-25 | 广州信锐电子科技有限公司 | Sine wave signal fixed amplitude and fixed voltage converting circuit with high precision, broad band and low aberration |
| CN101399908A (en) * | 2008-10-13 | 2009-04-01 | 深圳创维－Rgb电子有限公司 | Circuit for automatically regulating amplitude of video output signal of television |
| CN102201792A (en) * | 2010-03-25 | 2011-09-28 | 上海沙丘微电子有限公司 | Automatic gain control circuit of audio power amplifier |
| CN202334444U (en) * | 2011-10-31 | 2012-07-11 | 四川和芯微电子股份有限公司 | Adjustable-gain audio power amplifying circuit |
| CN203746419U (en) * | 2014-01-16 | 2014-07-30 | 四川和芯微电子股份有限公司 | Automatic voice identification circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103761964A (en) | 2014-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102722207B (en) | Low dropout regulator (LDO) | |
| CN106357227B (en) | Power amplifying module | |
| CN101794159B (en) | Band-gap reference voltage source of high power supply voltage rejection ratio | |
| CN111474975B (en) | Output current sampling circuit of LDO (low dropout regulator) and sampling precision adjusting method | |
| CN103034275A (en) | Low noise voltage regulator and method with fast settling and low-power consumption | |
| CN111176358B (en) | Low-power-consumption low-dropout linear voltage regulator | |
| WO2022179195A1 (en) | Bias circuit and amplifier | |
| KR20150111301A (en) | Voltage regulator | |
| CN107102680A (en) | A kind of low noise low pressure difference linear voltage regulator | |
| CN103916098A (en) | Programmable gain amplifier with high gain precision | |
| CN101594119B (en) | Method for designing weak signal reading circuit of variable gain amplifier | |
| CN105183061A (en) | Voltage buffer circuit | |
| CN106059512A (en) | Novel low-complexity broadband variable gain amplifier | |
| CN103761964B (en) | Automatic speech recognition circuit | |
| CN106026954B (en) | Operational Amplifier Frequency Compensation circuit | |
| CN103163926B (en) | High-accuracy low drop-out voltage regulator | |
| CN107422774A (en) | LDO on a kind of piece of low pressure fast transient response | |
| CN107066024A (en) | A kind of low power consumption high-precision non-bandgap reference voltage source | |
| KR20200110410A (en) | Inverting pseudo fully differential amplifier with common mode feedback control circuit | |
| CN203747761U (en) | Automatic gain adjustment circuit | |
| CN107154785B (en) | Control circuit, power amplification circuit and method | |
| CN203746419U (en) | Automatic voice identification circuit | |
| CN101833346A (en) | Low dropout regulator with enhanced precision and power supply rejection rate | |
| CN204652316U (en) | Low voltage difference amplifier, error amplifier and amplifying circuit | |
| CN102314187B (en) | Direct-current voltage proportion output circuit and control method thereof |
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 |