CN106297767B - Voice acquisition method and system based on speech recognition - Google Patents
Voice acquisition method and system based on speech recognition Download PDFInfo
- Publication number
- CN106297767B CN106297767B CN201610679482.7A CN201610679482A CN106297767B CN 106297767 B CN106297767 B CN 106297767B CN 201610679482 A CN201610679482 A CN 201610679482A CN 106297767 B CN106297767 B CN 106297767B
- Authority
- CN
- China
- Prior art keywords
- voice signal
- analog voice
- past
- enlargement ratio
- present day
- 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
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Processing of the speech or voice signal to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
Abstract
The present invention provides a kind of voice acquisition method and system based on speech recognition.Wherein method includes: acquisition analog voice signal；According to collected present day analog voice signal and in the past analog voice signal calculates the enlargement ratio that amplifies to present day analog voice signal, obtains current enlargement ratio；Present day analog voice signal is amplified according to current enlargement ratio, the present day analog voice signal amplified；The present day analog voice signal of amplification is subjected to analog-to-digital conversion, obtains Contemporary Digital voice signal, and using Contemporary Digital voice signal as the input signal of speech recognition.Its current enlargement ratio amplified to present day analog voice signal is real-time change, a relatively good current enlargement ratio can be provided for present day analog voice signal, present day analog voice signal after making enhanced processing is not in distorted signals or the inadequate situation of precision, one good signal input basis can be provided for speech recognition, improve the discrimination of speech recognition.
Description
Technical field
The present invention relates to voice collecting fields, more particularly to a kind of voice acquisition method based on speech recognition and are
System.
Background technique
Present speech recognition is more and more mature, and people also are untiringly making great efforts to seek that phonetic recognization rate can be improved
Method.One important link of speech recognition is exactly the acquisition of voice signal, and collected analog signal is passed through mould by chip
Number converter is converted into digital signal, the input as speech recognition algorithm.Therefore the acquisition of voice signal is phonetic recognization rate
One important parameter of height, a good base could be provided to speech recognition algorithm by only obtaining good digital signal
Plinth, to improve the discrimination of speech recognition.But traditional speech signal collection technology is providing signal for speech recognition algorithm
The amplification for usually carrying out phase same multiplying when input to the voice signal of acquisition, it cannot be language that the signal input provided is relatively rough
Sound identification provides good digital signal input, causes the discrimination of speech recognition not high.
Summary of the invention
In consideration of it, the problem that it is necessary to cause the discrimination of speech recognition not high for traditional voice Signal Collection Technology,
The voice acquisition method and system based on speech recognition of a kind of discrimination can be improved speech recognition are provided.
To reach goal of the invention, a kind of voice acquisition method based on speech recognition is provided, which comprises
Analog voice signal is acquired, wherein the analog voice signal includes present day analog voice signal and in the past simulates language
Sound signal；
It is calculated according to the collected present day analog voice signal and the in the past analog voice signal to described current
The enlargement ratio that analog voice signal amplifies obtains current enlargement ratio；
The present day analog voice signal is amplified according to the current enlargement ratio, the present day analog amplified
Voice signal；
The present day analog voice signal of the amplification is subjected to analog-to-digital conversion, obtains Contemporary Digital voice signal, and by institute
State input signal of the Contemporary Digital voice signal as speech recognition.
It is described according to collected present day analog voice signal and in the past analog voice signal in one of the embodiments,
And preset algorithm acquires and includes: with the step of current enlargement ratio amplified to the present day analog voice signal
The present day analog voice letter is obtained according to the present day analog voice signal and the in the past analog voice signal
Number average value；
It is obtained according to the maximum analog voice signal in the present day analog voice signal and the in the past analog voice signal
Take the outstanding value for indicating the optimal amplification effect of present day analog voice signal；
The current enlargement ratio is obtained according to the average value and the outstanding value.
According to collected present day analog voice signal and in the past, analog voice signal is obtained in one of the embodiments,
The step of obtaining the current enlargement ratio amplified to the present day analog voice signal include:
It obtains in the past ideal with the in the past analog voice signal of the neighbouring predetermined number of the present day analog voice signal
Enlargement ratio；
The in the past ideal enlargement ratio and corresponding with each in the past ideal enlargement ratio according to predetermined number
The enlargement ratio factor obtain the current enlargement ratio.
It is described in one of the embodiments, to obtain with the neighbouring predetermined number of the present day analog voice signal in the past
The step of in the past ideal enlargement ratio of analog voice signal includes:
The historical simulation voice signal acquired before each in the past analog voice signal is obtained, according to each described past
When analog voice signal and the corresponding historical simulation voice signal of each in the past analog signal obtain it is each it is described in the past
The in the past average value of analog voice signal；
According to the maximum history mould in each in the past analog voice signal and corresponding historical simulation voice signal
Quasi- voice signal obtains the in the past outstanding value for indicating each in the past optimal amplification effect of analog voice signal；
It is obtained according to the corresponding in the past average value of each in the past analog voice signal and in the past outstanding value each described
The in the past corresponding in the past ideal enlargement ratio of analog voice signal.
The in the past analog voice signal is closer to the present day analog voice signal, institute in one of the embodiments,
It is bigger to state specific gravity shared by the corresponding enlargement ratio factor of in the past ideal enlargement ratio of in the past analog voice signal；
The sum of corresponding enlargement ratio factor of the in the past ideal enlargement ratio of predetermined number meets default measurement value.
The present invention also provides a kind of speech collecting system based on speech recognition, the system comprises:
Acquisition module, for acquiring analog voice signal, wherein the analog voice signal includes present day analog voice letter
Number and in the past analog voice signal；
Obtain module, for according to collected present day analog voice signal and in the past analog voice signal calculating to described
The enlargement ratio that present day analog voice signal amplifies obtains current enlargement ratio；
Amplification module is obtained for being amplified according to the current enlargement ratio to the present day analog voice signal
The present day analog voice signal of amplification；
Conversion module obtains Contemporary Digital language for the present day analog voice signal of the amplification to be carried out analog-to-digital conversion
Sound signal, and using the Contemporary Digital voice signal as the input signal of speech recognition.
The acquisition module includes: in one of the embodiments,
Average value acquiring unit, for being obtained according to the present day analog voice signal and the in the past analog voice signal
The average value of the present day analog voice signal；
Outstanding value acquiring unit, for according in the present day analog voice signal and the in the past analog voice signal
Maximum analog voice signal obtains the outstanding value for indicating the optimal amplification effect of present day analog voice signal；
Enlargement ratio obtaining unit, for obtaining the current enlargement ratio according to the average value and the outstanding value.
The acquisition module includes: in one of the embodiments,
First acquisition unit, for obtaining and the in the past simulation language of the neighbouring predetermined number of the present day analog voice signal
The in the past ideal enlargement ratio of sound signal；
Second acquisition unit, for according to predetermined number in the past ideal enlargement ratio and with it is each it is described in the past
The corresponding enlargement ratio factor of ideal enlargement ratio obtains the current enlargement ratio.
The first acquisition unit includes: in one of the embodiments,
In the past average value obtains subelement, for obtaining the history mould acquired before each in the past analog voice signal
Quasi- voice signal, according to each in the past analog voice signal and each in the past corresponding historical simulation of analog signal
Voice signal obtains the in the past average value of each in the past analog voice signal；
In the past outstanding value obtains subelement, for according to each in the past analog voice signal and corresponding history mould
Maximum historical simulation voice signal in quasi- voice signal, which obtains, indicates the optimal amplification effect of each in the past analog voice signal
The in the past outstanding value of fruit；
In the past ideal enlargement ratio obtains subelement, corresponding in the past average according to each in the past analog voice signal
Value and in the past outstanding value obtain the corresponding in the past ideal enlargement ratio of each in the past analog voice signal.
The in the past analog voice signal is closer to the present day analog voice signal, institute in one of the embodiments,
It is bigger to state specific gravity shared by the corresponding enlargement ratio factor of in the past ideal enlargement ratio of in the past analog voice signal；
The sum of corresponding enlargement ratio factor of the in the past ideal enlargement ratio of predetermined number meets default measurement value.
The present invention also provides a kind of speech collecting system based on speech recognition, the system comprises:
Speech signal collection device, for acquiring analog voice signal, wherein the analog voice signal includes present day analog
Voice signal and in the past analog voice signal；
Multiplying power arithmetic unit, for according to collected present day analog voice signal and in the past analog voice signal calculating to institute
The enlargement ratio that present day analog voice signal amplifies is stated, current enlargement ratio is obtained；
Analogue amplifier is connect, for receiving the voice with the speech signal collection device and the multiplying power arithmetic unit
The present day analog voice signal of signal picker acquisition, and the current times magnification provided according to the multiplying power arithmetic unit
Rate amplifies the present day analog voice signal, the present day analog voice signal amplified；
Analog-digital converter is connect with the analogue amplifier, applies also for connecting with speech recognition arithmetic unit, for institute
The present day analog voice signal for stating amplification carries out analog-to-digital conversion, obtains Contemporary Digital voice signal, and transport to the speech recognition
It calculates device and inputs the Contemporary Digital voice signal.
The beneficial effect comprise that
Above-mentioned voice acquisition method and system based on speech recognition, amplifies present day analog voice signal current
Enlargement ratio is real-time change, current enlargement ratio or adjusted in real time according to the size of present day analog voice signal
To or be to be calculated with the corresponding in the past ideal enlargement ratio of in the past analog voice signal, therefore can be present day analog
Voice signal provides a relatively good current enlargement ratio, and the present day analog voice signal after making enhanced processing is not in put
Big excessive and distorted signals or the inadequate situation of the too small precision of amplification, and then a good signal can be provided for speech recognition
Input basis, improves the discrimination of speech recognition.
Detailed description of the invention
Fig. 1 is the flow diagram of the voice acquisition method based on speech recognition in one embodiment；
Fig. 2 is the flow diagram of the voice acquisition method based on speech recognition in another embodiment；
Fig. 3 is the flow diagram of the voice acquisition method based on speech recognition in another embodiment；
Fig. 4 is the modular structure schematic diagram of the speech collecting system based on speech recognition in one embodiment；
Fig. 5 is the electrical block diagram of the speech collecting system based on speech recognition in one embodiment；
Fig. 6 is the schematic diagram of original input signal in one embodiment；
Fig. 7 is the schematic diagram of the output signal in one embodiment；
Fig. 8 is the signal of the output signal in another embodiment.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, right with reference to the accompanying drawings and embodiments
The present invention is based on the voice acquisition methods of speech recognition and system to be further elaborated.It should be appreciated that described herein
Specific examples are only used to explain the present invention, is not intended to limit the present invention.
In one embodiment, as shown in Figure 1, providing a kind of voice acquisition method based on speech recognition, this method
The following steps are included:
S100 acquires analog voice signal, and wherein analog voice signal includes present day analog voice signal and in the past simulates
Voice signal.
S200, according to collected present day analog voice signal and in the past analog voice signal calculate to present day analog voice
The enlargement ratio that signal amplifies obtains current enlargement ratio；
S300 amplifies present day analog voice signal according to current enlargement ratio, the present day analog language amplified
Sound signal.
The present day analog voice signal of amplification is carried out analog-to-digital conversion, obtains Contemporary Digital voice signal, and will work as by S400
Input signal of the preceding audio digital signals as speech recognition.
After voice acquisition method in the present embodiment collects analog voice signal, by collected analog voice signal
It stores, collected analog voice signal is in the past analog voice signal before in this way, current collected analog voice
Signal is present day analog voice signal, when amplifying to present day analog voice signal, no longer using traditional to all simulations
Voice signal is all amplified using identical enlargement ratio, but according to present day analog voice signal and in the past analog voice letter
Number be calculated one can according to the size of the present day analog voice signal and real-time current enlargement ratio of dynamic change, Jin Ergen
Present day analog voice signal is amplified according to current enlargement ratio, the present day analog voice signal amplified, then to putting
Big present day analog voice signal carries out analog-to-digital conversion, obtains Contemporary Digital voice signal, only converts analog voice signal
For that could be identified by speech recognition algorithm after audio digital signals, using the current audio digital signals as the defeated of speech recognition
Enter signal, speech recognition is carried out according to the input signal, the discrimination of speech recognition can be effectively improved.Since it is every time to working as
It is all real-time change that front simulation voice signal, which amplifies the current enlargement ratio used when processing, current enlargement ratio or
It is to adjust in real time according to the size of present day analog voice signal or be corresponding in the past ideal in the past analog voice signal
Enlargement ratio and be calculated, a relatively good current enlargement ratio can be provided for present day analog voice signal, can be preferably
Processing is amplified to present day analog voice signal, the present day analog voice signal after making enhanced processing is not in amplified
Big and distorted signals or the inadequate situation of the too small precision of amplification, amplified present day analog voice signal can express its institute very well
The voice signal of number, the number are converted to after the amplified present day analog voice signal carries out analog-to-digital conversion comprising information
Input signal of the voice signal as speech recognition can provide a good signal input basis for speech recognition, improve
The discrimination of speech recognition.
In one embodiment, referring to fig. 2, step S200 includes:
S200a obtains the flat of present day analog voice signal according to present day analog voice signal and in the past analog voice signal
Mean value.
S200b, according to present day analog voice signal and in the past the maximum analog voice signal in analog voice signal obtains
Indicate the outstanding value of the optimal amplification effect of present day analog voice signal.
S200c obtains current enlargement ratio according to average value and outstanding value.
In the present embodiment for obtain with the size of present day analog voice signal and the one of the current enlargement ratio of dynamic change
A specific embodiment.For speech recognition algorithm, if analog signal be amplified it is excessive, may distortion phenomenon,
Such as: the voltage greater than 3.3V can be regarded as 1023 this digital signal by speech recognition algorithm, and if analog signal quilt
The multiple of amplification is inadequate, then the digital signal be converted to also can very little, cause precision not high, thus provide one it is good current
Enlargement ratio is the basis for obtaining good digital signal.A kind of available good current amplification is provided in the present embodiment
The algorithm of multiplying power: design present day analog voice signal outstanding value x and average value y, average value be present day analog voice signal with
The average value of the in the past analog voice signal acquired before, it is preferred that average value be present day analog voice signal with acquire before
In the past analog voice signal arithmetic mean of instantaneous value, outstanding value indicates that present day analog voice signal is optimal the number of amplification effect
Value, it is generally recognized that when the average value y of the present day analog voice signal of acquisition is intended to outstanding value x, it is believed that collected voice letter
It is number more perfect.Preferably, in one embodiment, outstanding value is 3/4ths of the value of maximum historical simulation voice signal.It is logical
Often think 3/4ths when the value of the average value and maximum historical simulation voice signal of collected present day analog voice signal
When close, collected present day analog voice signal is best, can directly use, and no longer needs to amplify, that is, amplification
Multiplying power is 1.Operation for simplicity is usually used as outstanding value for 3/4ths of the value of maximum historical simulation voice signal, this
It is worth the amplification effect of reflection present day analog voice signal that can be relatively good.Present day analog language is obtained according to average value and outstanding value
The current enlargement ratio z=z* (x/y) of sound signal, in this way when collected present day analog voice signal is very big, currently
The corresponding average value of analog voice signal will be bigger than outstanding value, then the value of x/y is exactly to reduce amplification less than 1, z* (x/y)
Multiplying power, when collected present day analog voice signal very little, the corresponding average value of present day analog voice signal will compare
Outstanding value is small, then the value of x/y is greater than 1, z* (x/y) is exactly to increase enlargement ratio at this time, to reach automatic adjustment
The effect of current enlargement ratio improves the discrimination of speech recognition to provide a good signal input for speech recognition.
The embodiment has relatively good application in recording field, and the voice signal human ear dealt using the above method is sounded more
It is good, it can be effectively reduced the probability of sonic boom.
Referring to Fig. 6 and Fig. 7, Fig. 6 is the schematic diagram that present day analog voice signal is collected in one embodiment, and Fig. 7 is to adopt
With the schematic diagram for the output signal that the method in the present embodiment obtains, it can be seen from the figure that using the method in the present embodiment
It can be good at amplifying original input signal.
In one embodiment, include: referring to Fig. 3, step S200
S210 is obtained in the past ideal with the in the past analog voice signal of the neighbouring predetermined number of present day analog voice signal
Enlargement ratio.
S220, according to the in the past ideal enlargement ratio of predetermined number and with each in the past ideal enlargement ratio is corresponding puts
The big multiplying power factor obtains current enlargement ratio.
The embodiment of current enlargement ratio is obtained using the above-mentioned average value according to collected voice signal and outstanding value
Although having preferable regulating effect to collected suddenly big or suddenly small voice signal, due to changing present day analog voice letter in real time
Number enlargement ratio to will lead to data unnatural, many deleterious effects are had to speech recognition.Speech recognition is different from human ear,
Speech recognition is the discriminance analysis to digital signal, and size, tone of digital signal etc. can all have an impact to it, so part
It zooms in or out sound instead and will affect the effect of speech recognition.Therefore, the preset algorithm in the present embodiment are as follows: using with it is current
The in the past ideal enlargement ratio of the in the past analog voice signal of the neighbouring predetermined number of analog voice signal calculates current times magnification
Rate, the current times magnification obtained by the in the past analog voice signal thus according to the predetermined number before present day analog voice signal
Rate considers the in the past analog voice signal in the past a period of time, be for the global enlargement ratio carried out
Adjust, rather than part zoom in or out voice signal, so that enlargement ratio can not only be automatically adjusted by reaching, but also do not have office
The adverse effect of voice signal bring is amplified in portion, so that the analog voice signal by amplification output is more natural.Referring to Fig. 6 and
Fig. 8, Fig. 6 are the schematic diagram for collecting present day analog voice signal in one embodiment, and Fig. 8 is using the side in the present embodiment
The schematic diagram for the output signal that method obtains, it can be seen from the figure that aobvious using the output signal that the method in the present embodiment obtains
So more smooth, the analog voice signal acquired is more natural.
Wherein, it is worth noting that, in the past analog voice signal is closer to present day analog voice signal, in the past analog voice
Specific gravity shared by the corresponding enlargement ratio factor of in the past ideal enlargement ratio of signal is bigger.The in the past ideal amplification of predetermined number
The sum of corresponding enlargement ratio factor of multiplying power meets default measurement value.That is the current enlargement ratio of present day analog voice signal be by
The corresponding in the past ideal enlargement ratio decision of the in the past analog voice signal of front predetermined number, and closer to present day analog language
Influence of the corresponding in the past ideal enlargement ratio of the in the past analog voice signal of sound signal to current enlargement ratio is bigger, so both
It can reflect influence of the overall situation to current enlargement ratio in a period of time, and can obtain preferably can be to present day analog voice signal
The current enlargement ratio amplified keeps the analog signal of output more natural, further improves good input for speech recognition
Signal, to further increase the discrimination of speech recognition.
Wherein, it should be noted that default measurement value indicates the in the past ideal enlargement ratio of predetermined number on the whole to working as
The metric level of the influence of preceding enlargement ratio, this metric level can be arbitrary value, obtain currently putting for the metric level
After big multiplying power, corresponding adjustment is done according to this current measurements rank, present day analog voice signal is put to obtain
Big current enlargement ratio.Operation for simplicity, this default measurement value is preferably 1.
In one embodiment, step S210 includes:
S210a obtains the historical simulation voice signal acquired before each in the past analog voice signal, according to each described
In the past analog voice signal and the corresponding historical simulation voice signal of each in the past analog signal, which obtain, each in the past simulates language
The in the past average value of sound signal.
S210b, according to the maximum history in each in the past analog voice signal and corresponding historical simulation voice signal
Analog voice signal obtains the in the past outstanding value for indicating each optimal amplification effect of in the past analog voice signal.
S210c, according to the corresponding in the past average value of each in the past analog voice signal and in the past outstanding value obtains each past
When the corresponding in the past ideal enlargement ratio of analog voice signal.
The step is to obtain the specific implementation step of the corresponding in the past ideal enlargement ratio of each in the past analog voice signal.
The dynamic adjustment effect for each in the past analog voice signal of reflection that each in the past ideal enlargement ratio can be relatively good, it is comprehensive
The influence to current enlargement ratio of multiple in the past ideal enlargement ratios, can reduce partial enlargement or reduces voice signal to language
The adverse effect of sound identification.
In one embodiment, predetermined number is 10.Quantity below in conjunction in the past ideal enlargement ratio is 10
One specific embodiment is described in detail:
It obtains and the in the past corresponding in the past ideal amplification of analog voice signal of present day analog voice signal neighbouring first 10
Multiplying power, respectively z1, z2, z3, z4, z5, z6, z7, z8, z9, z10 obtain the corresponding amplification of each in the past ideal enlargement ratio
The multiplying power factor, wherein the enlargement ratio factor can be preset based on experience value, and be stored in corresponding memory module
In, when use, calls directly, and is also possible to dynamic change, and rule change meets closer apart from present day analog voice signal
The corresponding enlargement ratio factor specific gravity of in the past analog voice signal it is bigger, and the sum of each enlargement ratio factor meets predetermined amount
The rule of angle value.According to 10, in the past in the past the corresponding enlargement ratio factor of ideal enlargement ratio obtains ideal enlargement ratio grade 10
To current enlargement ratio.Preferably, in one embodiment, the current enlargement ratio z of present day analog voice signal are as follows: z=a1*
z1+a2*z2+a3*z3+a4*z4+a5*z5+a6*z6+a7*z7+a8*z8+a9*z9+a10*z10；Wherein, a1 >=a2 >=a3 >=
a4≥a5≥a6≥a7≥a8≥a9≥a10；A1+a2+a3+a4+a5+a6+a7+a8+a9+a10=1.
Preferably, in a specific embodiment, a1=0.2, a2=0.18, a31=0.16, a4=0.14, a5=
0.10, a6=0.08, a71=0.06, a8=0.04, a91=0.02, a10=0.02.
It is worth noting that obtaining current amplification in the acquisition modes and previous embodiment of each in the past ideal enlargement ratio
The mode of multiplying power is identical, is all to be obtained by obtaining average value and outstanding value, details are not described herein again.
In one embodiment, include: storage analog voice signal after step S100, transfer mould when facilitating subsequent calculating
Quasi- voice signal.
In one embodiment, include: the current enlargement ratio of storage after step S200, facilitate and calculate next current amplification
It is used when multiplying power as in the past ideal enlargement ratio.
In one embodiment, include: storage Contemporary Digital voice signal after step S400, facilitate carry out speech recognition
It calls and reads.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random Access
Memory, RAM) etc..
In one embodiment, as shown in figure 4, additionally providing a kind of speech collecting system based on speech recognition, this is
System includes: acquisition module 100, and for acquiring analog voice signal, wherein analog voice signal includes present day analog voice signal
In the past analog voice signal.Module 200 is obtained, for according to collected present day analog voice signal and in the past analog voice
Signal calculates the enlargement ratio amplified to present day analog voice signal, obtains current enlargement ratio.Amplification module 300 is used
In being amplified according to current enlargement ratio to the present day analog voice signal, the present day analog voice signal amplified.
Conversion module 400, the present day analog voice signal for that will amplify carry out analog-to-digital conversion, obtain Contemporary Digital voice signal, and
Using Contemporary Digital voice signal as the input signal of speech recognition.
The current times magnification of processing is amplified in speech collecting system in the present embodiment to present day analog voice signal
Rate is real-time change, current enlargement ratio or be to adjust to obtain in real time according to the size of present day analog voice signal,
It is to be calculated with the corresponding in the past ideal enlargement ratio of in the past analog voice signal, therefore can believe for present day analog voice
Number provide a relatively good current enlargement ratio, the present day analog voice signal after making enhanced processing be not in amplification it is excessive
And distorted signals or the inadequate situation of the too small precision of amplification, amplified present day analog voice signal can express very well it and be wrapped
Containing information, and then a good signal input basis can be provided for speech recognition, improve the discrimination of speech recognition.
In one embodiment, obtaining module 200 includes: average value acquiring unit 200a, for according to present day analog language
Sound signal and in the past analog voice signal obtain the average value of the present day analog voice signal.Outstanding value acquiring unit 200b,
It indicates to work as being obtained according to the maximum analog voice signal in present day analog voice signal and the in the past analog voice signal
The outstanding value of the optimal amplification effect of front simulation voice signal.Enlargement ratio obtaining unit 200c, for according to average value and outstanding
Value obtains current enlargement ratio.
In one embodiment, obtaining module 200 includes: first acquisition unit 210, for obtaining and the present day analog
The in the past ideal enlargement ratio of the in the past analog voice signal of the neighbouring predetermined number of voice signal.Second acquisition unit 220 is used
According to the in the past ideal enlargement ratio of predetermined number and the enlargement ratio factor corresponding with each in the past ideal enlargement ratio
Obtain current enlargement ratio.
In one embodiment, first acquisition unit 210 include: in the past average value obtain subelement 210a, for obtaining
The historical simulation voice signal acquired before each in the past analog voice signal, according to each in the past analog voice signal and respectively
The corresponding historical simulation voice signal of a in the past analog signal obtains the in the past average value of each in the past analog voice signal.In the past
Outstanding value obtains subelement 210b, for according in each in the past analog voice signal and corresponding historical simulation voice signal
Maximum historical simulation voice signal obtain the in the past outstanding value for indicating each optimal amplification effect of in the past analog voice signal.It is past
When ideal enlargement ratio obtain subelement 210c, according to the corresponding in the past average value of each in the past analog voice signal and in the past excellent
Show value obtains the corresponding in the past ideal enlargement ratio of each in the past analog voice signal.
In one embodiment, in the past analog voice signal is described in the past to simulate language closer to present day analog voice signal
Specific gravity shared by the corresponding enlargement ratio factor of in the past ideal enlargement ratio of sound signal is bigger.The in the past ideal of predetermined number is put
The sum of corresponding enlargement ratio factor of big multiplying power meets default measurement value.
In one embodiment, further includes: memory module 500, for store analog voice signal, current enlargement ratio and
Contemporary Digital voice signal.
In one embodiment, the present invention also provides a kind of speech collecting system based on speech recognition, which includes:
Speech signal collection device 10, for acquiring analog voice signal, wherein the analog voice signal includes present day analog voice letter
Number and in the past analog voice signal.Multiplying power arithmetic unit 20, for according to collected present day analog voice signal and in the past simulating
Voice signal calculates the enlargement ratio amplified to the present day analog voice signal, obtains current enlargement ratio.Simulation is put
Big device 30, connect with speech signal collection device and the multiplying power arithmetic unit, for receiving the current of speech signal collection device acquisition
Analog voice signal, and present day analog voice signal is amplified according to the current enlargement ratio that multiplying power arithmetic unit 20 provides,
The present day analog voice signal amplified.Analog-digital converter 40, connect with analogue amplifier, applies also for transporting with speech recognition
It calculates device 60 to connect, for carrying out analog-to-digital conversion to the present day analog voice signal of amplification, obtains Contemporary Digital voice signal, and to
Speech recognition arithmetic unit 60 inputs Contemporary Digital voice signal.
The present embodiment can provide the hardware for the speech collecting system that good signal inputs to realize for speech recognition algorithm
Realization device can provide the current enlargement ratio an of dynamic change, after making enhanced processing for present day analog voice signal
Present day analog voice signal be not in the excessive and distorted signals of amplification or the inadequate situation of the too small precision of amplification, can be language
Sound identification provides a good signal input basis, improves the discrimination of speech recognition.
Preferably, in one embodiment, speech signal collection device 10 is MIC (Microphone, microphone) collector,
With preferable recording effect.
In one embodiment, further includes: digital signal processor 50 connects with analog-digital converter 40 and multiplying power arithmetic unit 20
It connects, applies also for connecting with speech recognition arithmetic unit 60, for storing analog voice signal, current enlargement ratio and Contemporary Digital
Voice signal, and the Contemporary Digital voice signal of storage is input to speech recognition arithmetic unit 60.
Since the principle that this system solves the problems, such as is similar to a kind of aforementioned voice acquisition method based on speech recognition,
The implementation of the system may refer to the implementation of preceding method, and overlaps will not be repeated.
Each technical characteristic of embodiment described above can be combined arbitrarily, for simplicity of description, not to above-mentioned reality
It applies all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, all should be considered as described in this specification.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to protection of the invention
Range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (7)
1. a kind of voice acquisition method based on speech recognition, which is characterized in that the described method includes:
Analog voice signal is acquired, wherein the analog voice signal includes present day analog voice signal and in the past analog voice is believed
Number；
It is calculated according to the collected present day analog voice signal and the in the past analog voice signal to the present day analog
The enlargement ratio that voice signal amplifies obtains current enlargement ratio；
The present day analog voice signal is amplified according to the current enlargement ratio, the present day analog voice amplified
Signal；
The present day analog voice signal of the amplification is subjected to analog-to-digital conversion, obtains Contemporary Digital voice signal, and work as by described in
Input signal of the preceding audio digital signals as speech recognition；
It is described to be calculated according to the collected present day analog voice signal and the in the past analog voice signal to described current
The enlargement ratio that analog voice signal amplifies, the step of obtaining current enlargement ratio include:
The present day analog voice signal is obtained according to the present day analog voice signal and the in the past analog voice signal
Average value；It is obtained according to the maximum analog voice signal in the present day analog voice signal and the in the past analog voice signal
Indicate the outstanding value of the optimal amplification effect of present day analog voice signal；Institute is obtained according to the average value and the outstanding value
State current enlargement ratio；
Alternatively,
It obtains and amplifies with the in the past ideal of the in the past analog voice signal of the neighbouring predetermined number of the present day analog voice signal
Multiplying power；According to predetermined number in the past ideal enlargement ratio and with each described in the past ideal enlargement ratio is corresponding puts
The big multiplying power factor obtains the current enlargement ratio.
2. the voice acquisition method according to claim 1 based on speech recognition, which is characterized in that it is described acquisition with it is described
The step of in the past ideal enlargement ratio of the in the past analog voice signal of the neighbouring predetermined number of present day analog voice signal includes:
The historical simulation voice signal acquired before each in the past analog voice signal is obtained, according to each in the past mould
Quasi- voice signal and the corresponding historical simulation voice signal of each in the past analog signal obtain each described in the past simulate
The in the past average value of voice signal；
According to the maximum historical simulation language in each in the past analog voice signal and corresponding historical simulation voice signal
Sound signal obtains the in the past outstanding value for indicating each in the past optimal amplification effect of analog voice signal；
According to the corresponding in the past average value of each in the past analog voice signal and in the past outstanding value obtain it is each it is described in the past
The corresponding in the past ideal enlargement ratio of analog voice signal.
3. the voice acquisition method according to claim 2 based on speech recognition, which is characterized in that described in the past to simulate language
For sound signal closer to the present day analog voice signal, the in the past ideal enlargement ratio of the in the past analog voice signal is corresponding
Specific gravity shared by the enlargement ratio factor is bigger；
The sum of corresponding enlargement ratio factor of the in the past ideal enlargement ratio of predetermined number meets default measurement value.
4. a kind of speech collecting system based on speech recognition, which is characterized in that the system comprises:
Acquisition module, for acquiring analog voice signal, wherein the analog voice signal include present day analog voice signal and
In the past analog voice signal；
Module is obtained, for calculating according to collected present day analog voice signal and in the past analog voice signal to described current
The enlargement ratio that analog voice signal amplifies obtains current enlargement ratio；
Amplification module is amplified for being amplified according to the current enlargement ratio to the present day analog voice signal
Present day analog voice signal；
Conversion module obtains Contemporary Digital voice letter for the present day analog voice signal of the amplification to be carried out analog-to-digital conversion
Number, and using the Contemporary Digital voice signal as the input signal of speech recognition；
The acquisition module includes average value acquiring unit, outstanding value acquiring unit and enlargement ratio obtaining unit；It is described average
It is worth acquiring unit, for obtaining the present day analog according to the present day analog voice signal and the in the past analog voice signal
The average value of voice signal；The outstanding value acquiring unit, for according to the present day analog voice signal and the in the past mould
Maximum analog voice signal in quasi- voice signal, which obtains, indicates the outstanding of the optimal amplification effect of present day analog voice signal
Value；The enlargement ratio obtaining unit, for obtaining the current enlargement ratio according to the average value and the outstanding value；
Alternatively,
The acquisition module includes first acquisition unit and second acquisition unit；The first acquisition unit, for acquisition and institute
State the in the past ideal enlargement ratio of the in the past analog voice signal of the neighbouring predetermined number of present day analog voice signal；Described second
Acquiring unit, for according to predetermined number in the past ideal enlargement ratio and with each in the past ideal enlargement ratio
The corresponding enlargement ratio factor obtains the current enlargement ratio.
5. the speech collecting system according to claim 4 based on speech recognition, which is characterized in that described first obtains list
Member includes:
In the past average value obtains subelement, for obtaining the historical simulation language acquired before each in the past analog voice signal
Sound signal, according to each in the past analog voice signal and each in the past corresponding historical simulation voice of analog signal
The in the past average value of each in the past analog voice signal of signal acquisition；
In the past outstanding value obtains subelement, for according to each in the past analog voice signal and corresponding historical simulation language
Maximum historical simulation voice signal in sound signal, which obtains, indicates each in the past optimal amplification effect of analog voice signal
In the past outstanding value；
In the past ideal enlargement ratio obtains subelement, according to the corresponding in the past average value of each in the past analog voice signal and
In the past outstanding value obtains the corresponding in the past ideal enlargement ratio of each in the past analog voice signal.
6. the speech collecting system according to claim 5 based on speech recognition, which is characterized in that described in the past to simulate language
For sound signal closer to the present day analog voice signal, the in the past ideal enlargement ratio of the in the past analog voice signal is corresponding
Specific gravity shared by the enlargement ratio factor is bigger；
The sum of corresponding enlargement ratio factor of the in the past ideal enlargement ratio of predetermined number meets default measurement value.
7. a kind of speech collecting system based on speech recognition, which is characterized in that the system comprises:
Speech signal collection device, for acquiring analog voice signal, wherein the analog voice signal includes present day analog voice
Signal and in the past analog voice signal；
Multiplying power arithmetic unit, for being worked as according to collected present day analog voice signal and in the past analog voice signal calculating to described
The enlargement ratio that front simulation voice signal amplifies obtains current enlargement ratio；
Analogue amplifier is connect, for receiving the voice signal with the speech signal collection device and the multiplying power arithmetic unit
The present day analog voice signal of collector acquisition, and the current enlargement ratio pair provided according to the multiplying power arithmetic unit
The present day analog voice signal amplifies, the present day analog voice signal amplified；
Analog-digital converter is connect with the analogue amplifier, applies also for connecting with speech recognition arithmetic unit, for putting to described
Big present day analog voice signal carries out analog-to-digital conversion, obtains Contemporary Digital voice signal, and to the speech recognition arithmetic unit
Input the Contemporary Digital voice signal；
The multiplying power arithmetic unit is also used to:
The present day analog voice signal is obtained according to the present day analog voice signal and the in the past analog voice signal
Average value；It is obtained according to the maximum analog voice signal in the present day analog voice signal and the in the past analog voice signal
Indicate the outstanding value of the optimal amplification effect of present day analog voice signal；Institute is obtained according to the average value and the outstanding value
State current enlargement ratio；
Alternatively,
It obtains and amplifies with the in the past ideal of the in the past analog voice signal of the neighbouring predetermined number of the present day analog voice signal
Multiplying power；According to predetermined number in the past ideal enlargement ratio and with each described in the past ideal enlargement ratio is corresponding puts
The big multiplying power factor obtains the current enlargement ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610679482.7A CN106297767B (en) | 2016-08-16 | 2016-08-16 | Voice acquisition method and system based on speech recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610679482.7A CN106297767B (en) | 2016-08-16 | 2016-08-16 | Voice acquisition method and system based on speech recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106297767A CN106297767A (en) | 2017-01-04 |
| CN106297767B true CN106297767B (en) | 2019-11-12 |
Family
ID=57679505
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610679482.7A Active CN106297767B (en) | 2016-08-16 | 2016-08-16 | Voice acquisition method and system based on speech recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106297767B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1700603A (en) * | 2004-12-31 | 2005-11-23 | 北京中星微电子有限公司 | Apparatus and method for digitalizing analog signal |
| CN101004673A (en) * | 2005-09-20 | 2007-07-25 | 三星电子株式会社 | Apparatus to convert analog signal of array microphone into digital signal and computer system including the same |
| CN101315770A (en) * | 2008-05-27 | 2008-12-03 | 北京承芯卓越科技有限公司 | System on speech recognition piece and voice recognition method using the same |
| CN101454973A (en) * | 2006-05-30 | 2009-06-10 | 冲电气工业株式会社 | Automatic gain controller |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04367899A (en) * | 1991-06-14 | 1992-12-21 | Ricoh Co Ltd | Agc control system of voice recognition device |
| JPH11194797A (en) * | 1997-12-26 | 1999-07-21 | Kyocera Corp | Speech recognition operating device |
-
2016
- 2016-08-16 CN CN201610679482.7A patent/CN106297767B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1700603A (en) * | 2004-12-31 | 2005-11-23 | 北京中星微电子有限公司 | Apparatus and method for digitalizing analog signal |
| CN101004673A (en) * | 2005-09-20 | 2007-07-25 | 三星电子株式会社 | Apparatus to convert analog signal of array microphone into digital signal and computer system including the same |
| CN101454973A (en) * | 2006-05-30 | 2009-06-10 | 冲电气工业株式会社 | Automatic gain controller |
| CN101315770A (en) * | 2008-05-27 | 2008-12-03 | 北京承芯卓越科技有限公司 | System on speech recognition piece and voice recognition method using the same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106297767A (en) | 2017-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106782584A (en) | Audio signal processing apparatus, method and electronic equipment | |
| JP4150798B2 (en) | Digital filtering method, digital filter device, digital filter program, and computer-readable recording medium | |
| CN206349145U (en) | Audio signal processing apparatus | |
| CN105847611B (en) | Echo time delay detection method, echo cancellation chip and terminal equipment | |
| CN107919133A (en) | For the speech-enhancement system and sound enhancement method of destination object | |
| CN106164845A (en) | Based on the dynamic audio frequency horizontal adjustment paid close attention to | |
| CN109121057A (en) | A kind of method and its system of intelligence hearing aid | |
| CN104883437B (en) | The method and system of speech analysis adjustment reminding sound volume based on environment | |
| CN108235181B (en) | Method for noise reduction in an audio processing apparatus | |
| CN107734126A (en) | voice adjusting method, device, terminal and storage medium | |
| CN102164203A (en) | Information processing device and method and program | |
| CN108696648A (en) | A kind of method, apparatus, equipment and the storage medium of Short Time Speech signal processing | |
| CN106448696A (en) | Adaptive high-pass filtering speech noise reduction method based on background noise estimation | |
| CN107369441A (en) | Noise-eliminating method, device and the terminal of voice signal | |
| CN110534125A (en) | A kind of real-time voice enhancing system and method inhibiting competitive noise | |
| CN111276150B (en) | Intelligent voice-to-text and simultaneous interpretation system based on microphone array | |
| CN103168479B (en) | Anti-singing device, sonifer, singing suppressing method and integrated circuit | |
| CN106297767B (en) | Voice acquisition method and system based on speech recognition | |
| US20100195839A1 (en) | Method and hearing device for tuning a hearing aid from recorded data | |
| CN110309284B (en) | Automatic answer method and device based on Bayesian network reasoning | |
| CN109326298B (en) | Game voice chat volume self-adaptive adjusting method | |
| WO2023006107A1 (en) | Automatic gain control method and apparatus for voice interaction system, and system | |
| CN114550729A (en) | Cry detection model training method and device, electronic equipment and storage medium | |
| US9355648B2 (en) | Voice input/output device, method and programme for preventing howling | |
| JP3514714B2 (en) | Sound collection method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| CB02 | Change of applicant information |
Address after: 519085 Guangdong city of Zhuhai province Jida West Road No. 107 Building 9 Building (1-4) Applicant after: Zhuhai jelee Polytron Technologies Inc Address before: 519085 Guangdong city of Zhuhai province Jida West Road No. 107 Building 9 Building Applicant before: Zhuhai Jieli Technology Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP02 | Change in the address of a patent holder | ||
| CP02 | Change in the address of a patent holder |
Address after: 519000 No. 333, Kexing Road, Xiangzhou District, Zhuhai City, Guangdong Province Patentee after: ZHUHAI JIELI TECHNOLOGY Co.,Ltd. Address before: Floor 1-107, building 904, ShiJiHua Road, Zhuhai City, Guangdong Province Patentee before: ZHUHAI JIELI TECHNOLOGY Co.,Ltd. |