KR20080014455A - Audio amplification device and method of amplifying audio signal - Google Patents

Abstract

A method and an apparatus for amplifying audio signals are provided to suppress a howling phenomenon dynamically by using a howling monitor. A reference signal generator(200) generates a reference signal in a time domain and outputs a first audio signal to the outside through a speaker. A microphone(210) converts the audio signal to a first analog electric signal. An ADC(Analog Digital Converter)(220) converts the first analog electric signal to a first digital signal. A first FFT(510) converts the first digital signal to a first frequency domain signal. A first coefficient extractor(610) extracts first magnitude coefficients for respective frequency fields from the first frequency domain signal. A second FFT(Fast Fourier Transform)(520) converts the second digital signal to a second frequency domain signal. A digital amplifier(710) amplifies the second frequency domain signal by using a gain coefficient which is an inverted version of the first magnitude coefficient. An adder(720) adds the second frequency domain signals for respective frequency fields to one another. An inverse FFT(730) converts the second frequency domain signals to a third digital signal of a time domain. A DAC(Digital to Analog Converter)(230) converts the third digital signal to the second analog electric signal. A speaker(290) outputs the second analog electric signal to the outside as the second audio signal.

Description

Audio amplification device and method of amplifying audio signal

1A and 1B are block diagrams schematically showing a conventional audio amplifying apparatus each having a howling cancellation function.

2 is a block diagram showing the configuration of an audio amplifying apparatus according to an embodiment of the present invention.

FIG. 3 is a flowchart illustrating a signal processing method of the audio amplifier shown in FIG. 2.

FIG. 4 is a conceptual diagram illustrating an operation process performed to cancel a howling in the audio amplifying apparatus shown in FIG. 2.

5 is a block diagram showing the configuration of an audio amplifying apparatus according to another embodiment of the present invention.

FIG. 6 is a flowchart illustrating a signal processing method of the audio amplifier shown in FIG. 5.

FIG. 7 is a conceptual diagram illustrating an operation process performed to cancel a howling in the audio amplifying apparatus shown in FIG. 5.

The present invention relates to an audio amplification apparatus and an audio amplification method, and more particularly, to an audio amplification apparatus and an audio amplification method for canceling a howling by signal processing on a frequency domain.

In an audio amplifying apparatus using a microphone and a speaker at the same time, a howling is generated when the sound output from the speaker is reflected by a wall or the like in the room and inputted back into the microphone. It is known that the cause of the howling is a resonance or oscillation caused by a repetitive positive feedback loop generated when a signal output from a speaker is input through a microphone again. Since the howling not only provides annoying noise but also damages other useful audio signals, conventional audio amplification devices have a device for canceling them.

1A and 1B are block diagrams schematically showing conventional audio amplifiers 9a and 9b each having a howling cancellation function.

The audio amplifying apparatus 9a shown in FIG. 1A includes a microphone amplifier 2a and a speaker having a frequency filtering characteristic for removing high frequency components of an audio signal input through the microphone 1 and amplifying the audio signals of the remaining frequency components. 8) the analog amplifier 3 for amplifying the sound output to the. This is howling cancellation by filtering schemes that invariably attenuate or cut-off the signal strength for certain frequency components, especially high-frequency audio signals, which can cause howling by a forward feedback loop. Provide the function.

The audio amplifier 9b shown in FIG. 1B is delayed by an output signal of an analog-digital converter 4 and an oscillator 6 between the microphone amplifier 2b and the analog amplifier 3. It consists of a delay circuit 5 and a digital-analog converter 7 which provide a time interval. The delay circuit 5 phase shifts the audio signal input from the analog-digital converter 4 and then transfers the phase-shifted audio signal to the digital-analog converter 7. It is proposed that the human hearing is not sensitive to phase shift, and is a proposed technique, which provides a howling cancellation function by phase shifting a signal input to the positive feedback loop.

However, the audio amplifying apparatus 9a of FIG. 1A has a problem in that sound quality is impaired by permanently losing an audio signal of a high frequency component, and has no effect on howling generated by other frequency components except high frequency. In addition, the audio amplification apparatus 9b of FIG. 1B delays the audio signal of the entire band and causes a chorus phenomenon to deteriorate the sound quality, and requires resources that require extremely complex operations and high memory capacity due to continuous signal processing. There is a resource budget.

Therefore, the technical problem to be achieved by the present invention, the problem of the conventional filtering method that time-varying or blocking the audio signal of a specific frequency causing the ringing of the audio signal input to the positive feedback loop and the deterioration of sound quality and full-band The present invention provides an audio amplifying apparatus that cancels a howling that can improve the problem of the phase shifting method having a resource burden due to continuous signal processing on an audio signal.

In addition, another technical problem to be achieved by the present invention is the problem of the conventional filtering scheme that time-varying or blocking the audio signal of a specific frequency causing the ringing of the audio signal input to the positive feedback loop and the degradation of sound quality and full-band It is an object of the present invention to provide an audio amplification method for canceling a howling that can improve a problem of a phase shifting method having a resource burden due to continuous signal processing on an audio signal.

According to an aspect of the present invention, there is provided an audio amplifying apparatus, including: a reference signal generator generating a reference signal having a reference intensity in a time domain and outputting a first audio signal to an external environment through a speaker; A microphone for converting an audio signal including the first audio signal reflected from the external environment into a first analog electrical signal; An analog-to-digital converter for converting the first analog electrical signal output from the microphone into a first digital signal; A first fast Fourier transform unit for converting the first digital signal into a first frequency domain signal in a frequency domain; A first coefficient extractor configured to extract first intensity coefficients for various frequency fields from the first frequency domain signal; A second fast Fourier transform unit for receiving a new second digital signal output through the microphone and the analog-digital converter and converting the second digital signal into a second frequency domain signal in a frequency domain; A digital amplifier for amplifying the second frequency domain signal for each frequency field using the inverse of the first intensity coefficient as a gain factor; An adder configured to add the second frequency domain signals amplified for each frequency field; An inverse fast Fourier transform unit for converting the added second frequency domain signal into a third digital signal in a time domain; A digital-analog converter that converts the third digital signal into a second analog electrical signal; And the speaker outputting the second analog electrical signal as a second audio signal to the external environment.

According to the audio amplifying apparatus of the present invention, by generating the first gain coefficient according to an external environment, it is possible to improve the sound quality degradation problem of the conventional filtering method that time-invariably cancels a fixed specific frequency audio signal. In addition, by using an arithmetic process based on a fast Fourier transform algorithm, which is an approximate method, the resource burden of the audio amplifier of the conventional phase shift method can be improved. In addition, the audio amplifying apparatus of the present invention may provide an audio amplifying apparatus capable of erasing a howling phenomenon in real time or dynamically by including a howling monitoring unit.

In addition, the audio amplification method according to an embodiment of the present invention for achieving the another technical problem, generates a reference signal having a reference intensity in the time domain to output the first audio signal to the external environment through the speaker, and again Converts an audio signal including the first audio signal echoed from the external environment into a first analog electrical signal. Thereafter, converting the first analog electrical signal into a first digital signal, and converting the first digital signal into a first frequency domain signal having a plurality of frequency fields in the frequency domain, thereby converting the first analog electrical signal from the first frequency domain signal. Extracting first intensity coefficients for each frequency field to generate first gain coefficients consisting of the inverse of the first intensity coefficients.

Thereafter, a new audio signal input to the microphone is converted into a second digital signal to obtain a second frequency domain signal having a plurality of frequency fields from the second digital signal in the frequency domain. The second frequency domain signal is amplified for each frequency field by the first gain factor. The second frequency domain signal amplified for each frequency field is added and converted, and the second frequency domain signal is converted into a third digital signal in the time domain. The third digital signal is converted into a second analog electrical signal, and is output to the external environment as a second audio signal through the speaker.

For this reason, in the audio amplification method of the present invention, it is possible to flexibly block the positive feedback loop of a specific frequency field that varies according to the external environment by the first gain coefficient generated differently according to the external environment, and approximate by a fast Fourier transform algorithm. In this way, the howling phenomenon can be easily eliminated while minimizing degradation of sound quality.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The embodiments of the present invention are provided to more fully explain the present invention to those skilled in the art, and the following examples can be modified in various other forms, and the scope of the present invention is It is not limited to the Example. Like numbers refer to like elements in the figures.

2 is a block diagram illustrating a configuration of an audio amplifying apparatus 1000 according to an embodiment of the present invention, and FIG. 3 is a flowchart illustrating a signal processing method of the audio amplifying apparatus 1000 illustrated in FIG. 2. 4 is a conceptual diagram illustrating an arithmetic process performed to cancel a howling in the audio amplifying apparatus shown in FIG. 2.

2 to 4, the reference signal generator 200 generates a reference signal having a reference strength in the time domain and outputs the first audio signal 10 to the external environment through the speaker 290 ( S10). The reference signal refers to a signal having a substantially constant reference intensity in an audible frequency band, that is, a frequency band within a range of 15 Hz to 20 kHz, that is, a signal or a flat signal having a substantially constant frequency-specific intensity. For example, an impulse signal in the time domain may be used as the reference signal. The external environment may be a closed place where the audio amplifying apparatus of the present invention is used, for example, a room, a vehicle, a movie theater, or the like.

The reference signal generator 200 may be connected to, for example, an input of the digital-analog converter 230. In this case, the reference signal output from the reference signal generator 200 is a digital signal having a constant intensity in an audible frequency band, and the digital reference signal is input to the digital-analog converter 230 and then the analog amplification unit. Through the 27 and the speaker 290, it is output as the first audio signal 10 to the external environment. In this case, the output first audio signal 10 may have substantially the same characteristics as that of the reference signal having a constant intensity in the audio frequency band, that is, a flat characteristic.

Thereafter, the first audio signal 10R reflected from the external environment through the speaker 290 may be converted into a first analog electric signal through the microphone 210 (S20). Although not shown, a microphone amplifier for amplifying the first analog electrical signal may be installed between the output of the microphone 210 and the input of the analog-to-digital converter 220.

The first analog electrical signal output from the microphone 210 may be converted into the first digital signal 30 by the analog-digital converter 220 (S30). The analog-to-digital converter 220 may convert the first analog electrical signal into the first digital signal 30 at a level of 22 kHz, which is the sampling rate of the FM radio sound, and 44.1 kHz, which is the sampling rate of the CD sound quality.

The first fast Fourier transform unit 510 receives the first digital signal 30 and converts the first digital signal 30 into the first frequency domain signal 40 in the frequency domain (S40). The first fast Fourier transform unit 510 may convert the first digital signal 30 into a first frequency domain signal 40 having a plurality of frequency fields in the frequency domain by a fast Fourier transform algorithm.

Thereafter, the first coefficient extractor 610 extracts the first intensity coefficients 50 for each frequency field from the first frequency domain signal 40 generated by the first fast Fourier transform unit 510. First gain coefficients 55 including the inverse of the first intensity coefficients 50 are generated (S50). In this case, the first intensity factor 50 means the strength of the first frequency domain signal 40 corresponding to each frequency field.

For example, the first digital signal 30 is generated at 22 kHz, which is a sampling rate of FM radio sound quality, and the first fast Fourier transform unit 510 performs the first digital signal 30 on 1024 frequency fields. In the case of the fast Fourier transform, a first frequency domain signal 40 having 1024 frequency fields is generated. At this time, the size of each frequency field is 21.5 Hz. The first coefficient extractor 610 extracts 1024 first intensity coefficients 50 from the first frequency domain signal 40, and obtains 1024 first gain coefficients 55 from the first frequency domain signal 40. The generated first gain coefficient may be stored in a memory 700 such as a register.

The reference signal generator 200 may repeatedly generate the reference signal until the distribution of the first gain coefficients 55 obtained by the first coefficient extractor 610 converges to a predetermined error range. Accordingly, the step S10 of outputting the first audio signal 10 to the step S50 of generating the first gain coefficients 50 for each frequency field may be repeated (× n).

The audio signal newly input to the microphone 210, for example, the voice signal S, is converted into an analog electric signal. The analog electrical signal of the speech signal S is then subjected to analog-to-digital converter 220 at 22 kHz, which is the same sampling rate as the first audio signal 10 described above, e.g., the sampling rate of the FM radio sound quality. The sample is sampled at the level and converted into the second digital signal 60 (S60).

The second fast Fourier transform unit 520 receives the second digital signal 60 and converts the second digital signal 60 into a second frequency domain signal 70 having a plurality of frequency fields in the frequency domain. (S70). The second fast Fourier transform unit 520 also converts the second digital signal 60 into the second frequency domain so that the second fast Fourier transform 520 has the same number as the first frequency domain signal 40, for example, 1024 frequency fields. Can be converted into a signal 70.

The digital amplifier 710 amplifies the second frequency domain signal 70 for each corresponding frequency field by the first gain factor 55 (S80). For example, in the case of the fast Fourier transform for the first digital signal 30 and the second digital signal 60 sampled at 22 kHz as described above to have 1024 frequency fields, respectively, N is equal to 1024. The first frequency domain signal 40 and the second frequency domain signal 70 having the size of each frequency field are about 21.5 Hz. The intensity coefficients α ₁ , α ₂ , α ₃ ,..., Α _i , ..., α ₁₀₂₂ , α ₁₀₂₃ , α ₁₀₂₄ corresponding to the corresponding frequency field of the second frequency domain signal 70 and the first A multiplication operation (80) is performed for each gain field (G ₁ , G ₂ , G ₃ , ..., G _i , ..., G ₁₀₂₂ , G ₁₀₂₃ , G ₁₀₂₄ ) for each frequency field, and for each corresponding frequency field. The second frequency domain signal 70 can be amplified to have new intensity coefficients Y ₁ , Y ₂ , Y ₃ ,..., Y _i ,..., Y ₁₀₂₂ , Y ₁₀₂₃ , Y ₁₀₂₄ .

Subsequently, the adder 720 performs an addition operation 90 on the second frequency domain signal 70 amplified for each frequency field (S90). Subsequently, the inverse fast Fourier transform unit 730 converts the added second frequency domain signal into a third digital signal 100 in the time domain (S100). The third digital signal 100 is converted into a second analog electrical signal by a digital-to-analog converter 110 (S120). Next, the second analog electrical signal is converted into the second audio signal 130 through the speaker 290 and output to the external environment (S130).

At this time, the audio amplifying apparatus 1000 of the present invention is connected between the output of the digital-analog converter 230 and the input of the speaker 290, the analog amplifying unit for amplifying the second analog electrical signal 270 may be further included. The second audio signal 130 output through the speaker 290 may have a relatively flat intensity distribution over the entire frequency to suppress occurrence of the howling phenomenon.

In addition, the audio amplifying apparatus 1000 according to the embodiment of the present invention is a second digital signal 60 output from the analog-to-digital converter 220, for example, the voice signal S input to the microphone 210. And a hybrid unit 300 which combines the digital signals output from the recording medium 400 for storing the multimedia data including the digital signal and the audio digital data to the second fast Fourier transform unit 520. can do. Thereby, the audio amplifying apparatus 1000 can be used as a karaoke apparatus having a howling cancellation function.

In an embodiment of the present invention, the first gain coefficients 50 reflect the internal amplification characteristics of the audio amplifying apparatus as well as the environmental characteristics of the external environment of the audio amplifying apparatus 1000, for example, a room, a vehicle, a movie theater, and the like. will be. The first gain coefficients 50 input the first audio signal 10 which is output from the external environment to be reflected back to the microphone 210 and the first audio signal 10R which is reflected back to the speaker 290 again. In the loop formed when outputted through, the first audio signal 10R reflected in the audio frequency band corresponds to a gain factor that can be amplified to have a flat characteristic. As a result, the first gain coefficients 50 may suppress formation of the positive feedback loop corresponding to the frequency field generating the howling phenomenon for each frequency field by reflecting the external environment.

Although the first gain coefficients 50 are extracted from the reference signal, the howling may be generated by applying the first gain coefficients 50 to the voice signal S input through the microphone 210. It is possible to selectively attenuate a signal of a specific frequency field. Thus, the howling phenomenon can be prevented by finally outputting the third digital signal 100 through which the signal of a specific frequency field in which the howling can be generated is selectively weakened through the speaker 290.

When the audio amplifying apparatus according to the embodiment of the present invention is used as a karaoke machine, since the relative positions of the microphone and the speaker may change according to the movement of the singer, the first gain factor 55 may be modified in consideration of external environmental factors. There is a need. Hereinafter, an audio amplifying apparatus and an audio amplifying method according to another embodiment of the present invention capable of monitoring the howling phenomenon in real time and modifying the first gain coefficient 55 to cancel the howling phenomenon will be described in detail.

FIG. 5 is a block diagram showing the configuration of an audio amplifying apparatus 2000 according to another embodiment of the present invention, and FIG. 6 is a flowchart showing a signal processing method of the audio amplifying apparatus 2000 shown in FIG. In addition, FIG. 7 is a conceptual diagram illustrating an arithmetic process performed to cancel a howling in the audio amplifying apparatus 2000 shown in FIG. 5.

5 to 7, the audio amplifying apparatus 2000 according to the present embodiment further includes a howling monitoring unit 640 for monitoring and erasing a howling phenomenon in real time. The howling monitoring unit 640 of the present invention includes a third fast Fourier transform unit 530, a second coefficient extractor 620, and a coefficient corrector 630. Hereinafter, the configuration and function of the howling monitoring unit 640 will be described in detail with respect to other components.

In FIG. 3, after outputting the second audio signal 130 to the external environment (S130), the third audio signal 130 including the second audio signal 130R and the new voice signal S reflected from the external environment ( 140 is input to the microphone 210, and is converted into a third analog electrical signal (S140). Although not shown, a microphone amplifier for amplifying the third analog electrical signal may be installed between the output of the microphone 210 and the input of the analog-to-digital converter 220.

The third analog electrical signal output from the microphone 210 may be converted into the fourth digital signal 150 by the analog-digital converter 220 (S150). As described above, the fourth digital signal 150 may be converted from the third analog electrical signal to a level of 22 kHz, which is the sampling rate of the FM radio sound quality, to 44.1 kHz, which is the sampling rate of the CD sound quality.

Thereafter, the third fast Fourier transform unit 530 receives the fourth digital signal 150 and converts the fourth digital signal 150 into the third frequency domain signal 160 in the frequency domain (S160). The third fast Fourier transform unit 530 also uses the fast Fourier transform algorithm to convert the fourth digital signal 150 into a third frequency such that it has the same number as the first frequency domain signal 40, for example, 1024 frequency fields. May be converted into a domain signal 160.

Thereafter, the second coefficient extractor 620 extracts the second intensity coefficients 170 for each frequency field from the third frequency domain signal 160 generated by the third fast Fourier transform unit 530 (S170). ). The coefficient corrector 630 determines whether at least one of the second intensity coefficients 170 exceeds a predetermined threshold value (S180). When at least one of the second intensity coefficients 170 exceeds the threshold value, it may be determined that a howling phenomenon has occurred. The threshold value may be selected in consideration of an appropriate noise level according to an external environment in which the audio amplifier 2000 is used.

The coefficient corrector 630 modifies the first gain factor by the inverse of the second intensity factor of the frequency field exceeding the threshold when one or more second intensity factor 170 exceeds the threshold ( S190). The digital amplifier 710 selects the second frequency domain signal 70 of the corresponding frequency field having the second intensity factor 170 exceeding the threshold value by the modified first gain coefficient 175 for each frequency field. Amplify by (S200).

For example, when the fourth digital signal 150 and the second digital signal 60 sampled at 22 kHz are fast Fourier-transformed for 1024 frequency fields, respectively, the third frequency domain having 1024 frequency fields, respectively. Signal 160 and second frequency domain signal 70 are generated. At this time, the size of the frequency field is about 21.5 Hz. Intensity coefficients α ₁ , α ₂ , α ₃ , ..., α _i , ..., α ₁₀₂₂ , α ₁₀₂₃ , α ₁₀₂₄ corresponding to the corresponding frequency field of the second frequency domain signal 70 and the third Among the second intensity coefficients 170 extracted from the frequency domain signal 160, the first gain coefficients G ₁ , G ₂ , G ₃ , modified by the inverse of the intensity coefficient X _i ′ exceeding the threshold value. The multiplication operation 190 is performed for each frequency field with respect to ..., G _i ', ..., G ₁₀₂₂ , G ₁₀₂₃ , G ₁₀₂₄ ).

As a result, the third digital signal 100m having new intensity coefficients Y ₁ , Y ₂ , Y ₃ , ..., Y _i ', ..., Y ₁₀₂₂ , Y ₁₀₂₃ , Y _{1024 for} each corresponding frequency field. Can be obtained. The howling phenomenon can be suppressed in real time because the strength factor of the corresponding frequency field having the Xi 'intensity factor that can cause the howling phenomenon to exceed the threshold is reduced by G _i '.

Subsequently, the adder 720 performs an add operation 90 on the second frequency domain signal 70 amplified for each frequency field (S90), and performs the add operation by the inverse fast Fourier transform unit 730. 90) the second frequency domain signal is converted into the third digital signal 100m in the time domain (S100). The third digital signal 100m is converted into a second analog electrical signal by the digital-analog converter 110 (S120). Next, the second analog electrical signal may be amplified by the analog amplifying unit 270, converted into the second audio signal 130 through the speaker 290, and then output to the external environment (S130). In this case, the second audio signal 130 may have a relatively flat intensity distribution over the entire frequency.

In the present embodiment, the arithmetic process (S170) of selectively amplifying the second frequency domain signal 70 of the corresponding frequency field by the modified first gain factor 175 reduces the sound quality of the second digital signal 60. In order to prevent this, considering that the maximum audible frequency of the human being is 20 kHz, it is preferably performed for 20 kHz to 200 kHz, which is difficult to distinguish from hearing. Accordingly, the modified first gain factor 175 is restored to the first gain factor 55 before the correction.

As described above, according to the audio amplifying apparatus and the audio amplifying method according to the present embodiment, the new third audio signal 140 including the echoed second audio signal 130R and the voice signal S input through the microphone 210. In addition to monitoring the presence or absence of a howling in real time, and when a howling is detected, only the signal in the frequency field that generates the howling is amplified and reduced to minimize the deterioration of sound quality while eliminating the howling. can do. In addition, it is possible to prevent the continuous deterioration of sound quality by performing the above-described signal processing only during a time difficult to distinguish human hearing.

In addition, the audio amplifying apparatus 2000 according to another embodiment of the present invention also has a second digital signal 60 output from the analog-to-digital converter 220, for example, a voice signal S input to the microphone 210. A hybrid unit 300 which combines the digital signal output from the recording medium 400 storing the digital signal including the digital signal and the audio data including the digital signal) and provides it to the second fast Fourier transform unit 520. It may further include. As a result, when the audio amplifying apparatus 2000 according to the present embodiment is applied to a karaoke apparatus in which a howling phenomenon occurs frequently, the howling phenomenon that may occur in the karaoke apparatus may be canceled in real time or dynamically.

In the audio amplification apparatus according to the embodiments of the present invention described above, the fast Fourier transform units 510, 520, 530, coefficient extractors 610, 620, coefficient corrector 630, memory 700, Any one or a combination of the digital amplifier 710, the adder 720, and the inverse fast Fourier transform 730 may include an output of the analog-to-digital converter 220 and a digital-to-analog converter 230. It may be implemented as a single digital signal processor 800 connected between the input units. In this case, the audio amplifying apparatus may further include a microcomputer for controlling the digital signal processing apparatus 800.

In addition, the first, second, and third fast Fourier transform units 510, 520, and 530 are functionally divided, and to those skilled in the art, they may have separate software or hardware configurations, and each may have a single software or hardware. It is apparent that the configuration 500 may have an enemy configuration 500.

The present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications, and alterations are possible within the scope without departing from the technical spirit of the present invention, which are common in the art. It will be apparent to those who have knowledge.

Therefore, according to the audio amplifying apparatus of the present invention, by generating the first gain coefficient according to the external environment, it is possible to improve the sound quality degradation problem of the conventional filtering method that time-invariably cancels the fixed specific frequency audio signal. . In addition, the calculation process according to the present invention by the fast Fourier transform algorithm is an approximate method, which is less complicated, and can improve the resource budget of a conventional phase conversion type audio amplifier. In addition, the audio amplifying apparatus of the present invention may provide an audio amplifying apparatus capable of erasing a howling phenomenon in real time or dynamically by including a howling monitoring unit.

According to the audio amplification method of the present invention, it is possible to flexibly block the positive feedback loop of a specific frequency field that varies depending on the external environment by the first gain coefficient generated differently according to the external environment. In addition, the audio amplification method of the present invention not only can easily remove the howling phenomenon by using a fast Fourier transform algorithm, which is an approximate method, but also includes a howling monitoring step to eliminate the howling phenomenon in real time or dynamically. Amplification methods can be provided.

Claims (24)

A reference signal generator for generating a reference signal having a reference intensity in the time domain and outputting a first audio signal to an external environment through a speaker; A microphone for converting an audio signal including the first audio signal reflected from the external environment into a first analog electrical signal; An analog-to-digital converter for converting the first analog electrical signal output from the microphone into a first digital signal; A first fast Fourier transform unit for converting the first digital signal into a first frequency domain signal in a frequency domain; A first coefficient extractor configured to extract first intensity coefficients for various frequency fields from the first frequency domain signal; A second fast Fourier transform unit for receiving a new second digital signal output through the microphone and the analog-digital converter and converting the second digital signal into a second frequency domain signal in a frequency domain; A digital amplifier for amplifying the second frequency domain signal for each frequency field using the inverse of the first intensity coefficient as a gain factor; An adder configured to add the second frequency domain signals amplified for each frequency field; An inverse fast Fourier transform unit for converting the added second frequency domain signal into a third digital signal in a time domain; A digital-analog converter that converts the third digital signal into a second analog electrical signal; And And the speaker outputting the second analog electrical signal as a second audio signal to the external environment. The method of claim 1, And the reference signal is a digital signal having a constant intensity in an audible frequency band. The method of claim 1, And the reference signal generator repeatedly generating the first audio signal until a distribution of first intensity coefficients for each frequency field extracted by the first coefficient extractor converges to a predetermined error range. The method of claim 1, And the first audio signal is an impulse signal. The method of claim 1, And the analog-to-digital converter converts the first digital signal and the second digital signal at the same sampling rate. The method of claim 1, And a storage unit for storing the first intensity coefficient for each frequency field. The method of claim 1, A third fast Fourier transform for converting a third analog electrical signal converted from the frequency domain to a third frequency domain signal while a third audio signal comprising the second audio signal passes through the microphone and the analog-to-digital converter part; A second coefficient extracting unit extracting second intensity coefficients for each frequency field from the third frequency domain signal; A coefficient corrector for correcting the first gain coefficient of the frequency field when at least one of the second intensity coefficients exceeds a predetermined threshold, the inverse of the second intensity coefficient of the corresponding frequency field exceeding the threshold value; To further include howling monitoring unit, And the digital amplifying unit selectively amplifies the second frequency domain signal of the corresponding frequency field having a second intensity coefficient exceeding the threshold value by the modified first gain coefficient for each frequency field. The method according to claim 1 or 7, And the second frequency domain signal and / or the third frequency domain signal are converted for the same number of frequency fields as the first frequency domain signal. The method of claim 7, wherein And in the digital amplifier, the second frequency domain signal is selectively amplified for 20 kHz to 200 kHz, and the modified first gain coefficient is restored to the first gain coefficient before correction. The method according to claim 1 or 7, And a hybrid unit which combines the digital signals output from the recording medium storing the second digital signal and the multimedia data including the audio digital data to each other and provides them to the second fast Fourier transform unit. The method according to claim 1 or 7, And a microphone amplifier connected between the output of the microphone and the input of the analog-to-digital converter to amplify the first analog electrical signal. The method according to claim 1 or 7, And an analog amplifier connected between an output of the digital-analog converter and an input of the speaker to amplify the second analog electrical signal. The method according to claim 1 or 7, The first fast Fourier transform unit, the second fast Fourier transform unit, the third fast Fourier transform unit, the first coefficient extractor, the second coefficient extractor, the coefficient corrector, the digital amplifier, and the addition At least one or a combination of an inverse fast Fourier transform unit and a combination thereof is connected to an output of the analog-to-digital converter and is coupled to an input of the digital-to-analog converter. Audio amplification device implemented in. The method of claim 13, And a microcomputer for controlling the digital signal processor. Generating a reference signal having a reference intensity in the time domain and outputting a first audio signal to an external environment through a speaker; Converting an audio signal comprising the first audio signal echoed from the external environment into a first analog electrical signal; Converting the first analog electrical signal to a first digital signal; A first fast Fourier transform step of converting the first digital signal into a first frequency domain signal having a plurality of frequency fields in the frequency domain; Extracting first intensity coefficients for each frequency field from the first frequency domain signal to generate first gain coefficients consisting of an inverse of the first intensity coefficients; Converting a new audio signal input to the microphone into a second digital signal; A second fast Fourier transform step of converting the second digital signal into a second frequency domain signal having a plurality of frequency fields in the frequency domain; Amplifying the second frequency domain signal for each corresponding frequency field by the first gain factor; Performing an addition operation on the second frequency domain signal amplified for each frequency field; An inverse fast Fourier transform step of converting the added second frequency domain signal into a third digital signal in a time domain; Converting the third digital signal to a second analog electrical signal; And And outputting the second analog electrical signal to the external environment through the speaker as a second audio signal. The method of claim 15, And after converting the first analog electrical signal through the microphone, amplifying the first analog electrical signal. The method according to claim 15 or 16, After converting the third digital signal to a second analog electrical signal, further comprising amplifying the second analog electrical signal. The method of claim 15, And said reference signal is a digital signal having a constant intensity in an audible frequency band. The method of claim 15, And the reference signal is an impulse signal. The method of claim 15, Outputting the first audio signal to generating first gain coefficients consisting of the reciprocal of the first intensity coefficients until the first gain coefficients converge to a predetermined error range. The method of claim 15, And the first digital signal and the second digital signal are converted at the same sampling rate. The method of claim 15, After outputting the second audio signal to the external environment, Converting a third audio signal comprising the second audio signal echoed from the external environment into a third analog electrical signal; Converting the third analog electrical signal to a fourth digital signal; A third fast Fourier transforming step of converting the fourth digital signal from a frequency domain into a third frequency domain signal; Extracting a second intensity coefficient for each frequency field from the third frequency domain signal; Determining whether the second intensity factor exceeds a predetermined threshold value; And If the second intensity factor exceeds a predetermined threshold value, a howling monitoring step comprising modifying the first gain factor of the frequency field by the inverse of the second intensity factor of the corresponding frequency field exceeding the threshold value; More, In the amplifying the second frequency domain signal for each frequency field, the second frequency domain of the corresponding frequency field having a second intensity factor exceeding the threshold value by the modified first gain coefficient for each frequency field. An audio amplification method for selectively amplifying a signal; The method of claim 22, And selectively amplifying the second frequency domain signal is performed for 20 Hz to 200 Hz, and the modified first gain coefficient is restored to the first gain coefficient before the correction. The method of claim 15 or 22, Combining the digital signals output from the recording medium storing the second digital signal and the multimedia data including the audio digital data with each other and providing them to the second fast Fourier transform step.

Images