JPH09130182A - Voice processing circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the voice processing circuit which obtains only required voice components without always excessively limiting the band of a voice input signal neither obtaining plural input signals and is easily realized. SOLUTION: A voice input signal 1 supplied to a cut-off frequency variable type low pass filter 4 has the high band component higher than a prescribed cut-off frequency eliminated and is supplied to an amplification degree variable type inverting amplifier 5 and not only is amplified with a prescribed amplification factor but also has the phase inverted by 180 deg. and is suppied to an adder 6 and is added to the voice input signal 1 (original signal), and the output of the adder 6 is supplied to an adder 7. The voice input signal 1 supplied to a cut-off frequency variable type low pass filter 11 has the low band component lower than a prescribed cut-off frequency eliminated and is supplied to an amplification degree variable type inverting amplifier 12 and not only is amplified with a prescribed amplification factor but also has the phase inverted by 180 deg. and is supplied to an adder 7 and is added to the signal supplied from the adder 6, and the result is outputted to a output terminal 16.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio processing circuit, and more particularly to an audio processing circuit that is mainly used in a camera-integrated VTR, a tape recorder or the like and is capable of obtaining only necessary audio components.

[0002]

2. Description of the Related Art Generally, it is known that the intelligibility of voice changes depending on the frequency range of a transmission system. FIG.
Is a graph showing the relationship between speech power and intelligibility by Fletcher's filter.

A broken line in FIG. 3 is a result of measuring the intelligibility of the voice signal when the voice signal is passed through a low-pass filter or a high-pass filter (LPF or HPF). And the speech intelligibility when receiving only frequency components below or above it.

As is clear from the HP broken line in FIG.
It can be seen that even if the low frequency (low-pitched sound portion) is cut out with F, the deterioration of the voice clarity is small. For example, even if the frequency is cut off at 1000 Hz or less, the clarity is as high as 85%, which is a level without any problem. However, as shown by the solid line in the figure, it can be seen that the energy level has dropped to 20%.

On the other hand, as is apparent from the broken line LP in FIG.
It can be seen that when the high frequency (high-pitched sound part) is cut out by the LPF, the intelligibility of the voice is relatively lowered. For example, 1
It can be seen that when the frequency is cut off at 000 Hz or more, the decrease in clarity is about 60%, but the decrease in energy level is 20%, which is not so low.

From the above, the sound energy is mostly 100
It is below 0Hz, and the clarity is improved (improved)
Therefore, it can be seen that a high frequency component is necessary for this.

By the way, FIG. 4 is a graph showing the relationship between the level of voice strength and intelligibility. It can be seen from FIG. 4 that the clarity becomes almost constant when the voice signal strength exceeds a certain value, and conversely decreases when the voice signal strength becomes too large. Further, the dB (decibel) value on the curve in the figure indicates the magnitude of noise, and it can be seen that if there is noise, the same intelligibility cannot be obtained unless the sound level is raised.

FIG. 5 is a graph showing the relationship between the level of the loudness of voice and the intelligibility. From FIG. 5, for example, if the noise level is 75 phons and the noise level is 65 phons, the intelligibility is 80%. You can see that you can't hear.

As described above, there is a generally known relationship (characteristic) between the voice signal and its intelligibility (intelligibility), and the present invention has been conventionally based on these characteristics. Some audio processing circuits have been designed. Next, an example of such a conventional voice processing circuit will be described.

FIG. 6 is a block diagram showing an example of a conventional voice processing circuit. Conventionally, in order to obtain a necessary voice component from a voice input signal, for example, a circuit as shown in FIG. 6A or 6B has been used.

The circuit shown in FIG. 6 (a) converts the audio signal 23 input from the input terminal 24 into a band limiting filter (BP).
F) 25 to output to the output terminal 26. However, in such a circuit, since the necessary voice component is obtained by using only the band limiting filter (BPF) 25, the band limiting filter (BPF) 25 has a sharp cutoff characteristic, A large amount of attenuation is required as its characteristic. If these requirements cannot be satisfied, there is little effect of obtaining the necessary voice component, and conversely, there is the problem that even the required band is attenuated.

On the other hand, the circuit shown in FIG. 6B is composed of a direct sound input microphone 21, an indirect sound input microphone 22, amplifiers 27 and 28, an adder 29, and an output terminal 30. Thus, two microphones for direct sound input and indirect sound input are used to obtain a voice input signal.

Then, the audio signal input to the direct sound input microphone 21 and converted into an electric signal is amplified by the amplifier 27 and output to the adder 29. The audio signal input to the indirect sound input microphone 22 and converted into an electric signal is inverted and amplified by the amplifier 28 and output to the adder 29. Then, the adder 29 receives an unnecessary voice signal (for example, noise such as private language or sound of an audience other than the lecturer) input from the amplifier 27 and picked up by the direct sound input microphone 21, and the amplifier 28. The audio signal picked up by the indirect sound input microphone 22 input more ideally (ideally, noise such as the private language of the audience not including the voice of the speaker or noise such as noise) is added by the signal inverted by 180 °, A target voice signal from which noises such as private words and noises of the audience other than the lecturer have been removed, that is, a voice signal including only the talk (voice) of the lecturer is output to the output terminal 30.

However, according to the method of adding a plurality of input signals, a plurality of input sources such as microphones are required, which complicates the apparatus and conditions for obtaining unnecessary voice components (for example, a place where the microphone is installed). (Problems such as) cannot be established (solved), problems such as attenuating necessary voice also occur, and there is a problem that the desired effect cannot be obtained. Further, the conventional circuit has a problem that the band control of the audio signal cannot be performed.

[0015]

As described above, in the conventional speech processing circuit using the band limiting filter (BPF), it is necessary to have a sharp cutoff characteristic and a large amount of attenuation as characteristics. Therefore, it is expensive and the band limiting filter (B
There was a problem that PF) required. Further, in the method of adding a plurality of audio input signals, since a plurality of input sources such as a microphone are required, the device becomes complicated, or when a condition for obtaining an unnecessary audio component cannot be established, There is a problem (defect) that a desired effect may not be obtained due to a problem such as a necessary sound being attenuated.

Therefore, in order to eliminate such a problem, the present invention does not always apply excessive band limitation to a voice input signal, and does not obtain a plurality of input signals (without necessity). An object of the present invention is to provide an audio processing circuit which can obtain only necessary audio components and can be easily realized.

[0017]

An audio processing circuit according to a first aspect of the present invention comprises input means for inputting an audio signal,
Low-pass filtering means capable of changing a cutoff frequency for extracting a low-frequency component from the audio signal, and first variable-amplification factor for inverting and amplifying a signal output from the low-pass filtering means. Detecting the signal level output from the inverting amplification means and the low-pass filtering means, and detecting the cutoff frequency of the low-pass filtering means and the amplification factor of the first inverting amplification means according to the detected signal level. A first signal level detection control means for controlling one or both of them, a first addition for adding a signal output from the first inverting amplification means, and a sound signal input from the input means. Means and means are provided.

According to a second aspect of the present invention, there is provided the audio processing circuit according to the first aspect, wherein the low-pass filtering means and the first inverting amplification means are integrally formed. Characterize.

According to the first or second aspect of the present invention, the band in which the voice input signal passes through the low-pass filtering means is set wider, and the original signal (voice input signal) is further set.
Since the amplification factor of the first inverting amplification means is made variable so that unnecessary low band components are attenuated when added to, it is possible to prevent excessive restriction of the low frequency band.

According to a third aspect of the present invention, there is provided an audio processing circuit, wherein input means for inputting an audio signal, high-pass filtering means for extracting a high-frequency component from the audio signal, the cut-off frequency being variable, are provided. Second inverting and amplifying means capable of changing the amplification factor for inverting and amplifying the signal output from the high-pass filtering means, and detecting the signal level output from the high-pass filtering means and detecting the detected signal Depending on the level
Output from the second signal level detection control means for controlling one or both of the cutoff frequency of the high-pass filtering means and the amplification factor of the second inverting amplification means, and output from the second inverting amplification means. It is characterized by further comprising second adding means for adding the generated signal and the audio signal inputted from the input means.

According to a fourth aspect of the present invention, in the voice processing circuit according to the third aspect, the high-pass filtering means and the second inverting amplification means are integrally formed. Characterize.

According to the third or fourth aspect of the present invention, the band in which the voice input signal passes through the high-pass filtering means is set wider, and the original signal (voice input signal) is further set.
Since the amplification factor of the second inverting amplification means is made variable so that the unnecessary high band component is attenuated when added to, it is possible to prevent excessive restriction of the high frequency band.

An audio processing circuit according to a fifth aspect of the present invention comprises an input means for inputting an audio signal, and a low-pass filtering means for extracting a low-frequency component from the audio signal, the low-pass filtering means capable of changing the cut-off frequency. First inverting amplification means capable of changing the amplification factor for inverting and amplifying the signal output from the low pass filtering means, and detecting and detecting the signal level output from the low pass filtering means. First signal level detection control means for controlling either one or both of the cutoff frequency of the low-pass filtering means and the amplification factor of the first inverting amplification means according to the signal level; and the first signal level detection control means.
First adding means for adding the signal output from the inverting amplifying means to the audio signal input from the input means,
High-pass filtering means for extracting a high-frequency component from the audio signal input from the input means, the cut-off frequency of which can be changed; and a signal output from the high-pass filtering means, which is inverted and amplified, and the amplification factor of which can be changed. The signal level output from the second inverting amplification means and the high-pass filtering means is detected, and the cutoff frequency of the high-pass filtering means and the second inverting amplification means of the second inverting amplification means are detected according to the detected signal level. A second signal level detection control means for controlling one or both of the amplification factors, a signal output from the second inverting amplification means, and a signal output from the first addition means are provided. And a second adding means for adding.

A sound processing circuit according to a sixth aspect of the present invention is the sound processing circuit according to the fifth aspect, wherein the low-pass filtering means and the first inverting amplifying means, and the high-pass filtering means and the It is characterized in that both or one of the second inverting amplification means is integrally formed.

According to the fifth or sixth aspect of the present invention, the band in which the voice input signal passes through the low-pass filtering means or the high-pass filtering means is set to be wider, and the original signal (voice Since the amplification factor of the first or second inverting amplification means is made variable so that unnecessary high band components or low band components are attenuated when added to the input signal), Excessive band limitation in the frequency band or either can be prevented.

According to a seventh aspect of the present invention, in the voice processing circuit according to the first, second, fifth or sixth aspect, the signal level detected by the first signal level detection control means is high. In this case, the cutoff frequency of the low-pass filtering means is set high, and when the signal level is low, the cutoff frequency of the low-pass filtering means is set low and the amplification factor of the first inverting amplification means is made small. It is characterized by setting.

An audio processing circuit according to an eighth aspect of the present invention is the audio processing circuit according to the third, fourth, fifth, sixth or seventh aspect, in which the signal level detected by the second signal level detection control means is detected. When the signal level is low, the cutoff frequency of the high-pass filtering means is set low, and when the signal level is low, the cutoff frequency of the high-pass filtering means is set high, and the amplification factor of the second inverting amplification means is set. Is set to be small.

According to the invention of claim 7 or 8, the band in which the voice input signal passes through the low-pass filtering means or the high-pass filtering means is set to a desired value, and the original signal is further set. Since it is possible to set the amplification factor of the first or second inverting amplification means to a desired value so that unnecessary high band component or low band component is attenuated when added to the (voice input signal). Since the band limitation in the low frequency band and / or the high frequency band can be set to a value desired by the user, for example, the sound (signal) from a specific sound source in noise is emphasized (pickup). You can take it out.

[0029]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a voice processing circuit of the present invention.

The audio processing circuit shown in FIG. 1 includes a low frequency band limiting unit 17 which limits (cuts off) low frequency components of a voice signal.
It is composed of two blocks of a high frequency band limiting unit 18 which limits (cuts off) a high frequency component of an audio signal.

Then, the low frequency band limiting section 17 receives the audio input signal 1 input from the input terminal 2 through the buffer amplifier 3.
Cut-off frequency variable low-pass filter (low-pass filtering means) 4 that passes the low-frequency components of the input signal, and a variable-amplification inverting amplifier that outputs a signal whose phase is inverted by 180 ° with respect to the input signal (first 1 inverting amplification means 5) and the cutoff frequency variable low-pass filter (low-pass filtering means)
4 cutoff frequency controlled by control signal 9 (variable)
In addition, a level detector (first signal level detection control means) 10 for controlling (changing) the amplification factor of the variable amplification type inverting amplifier (first inverting amplification means) 5 by a control signal 10, and an adder. 6 and.

Further, the high frequency limiter 18 is provided with the audio input signal 1 input from the input terminal 2 via the buffer amplifier 3.
A variable cut-off frequency type high-pass filter (high-pass filtering means) 11 that passes the high-frequency component of the input signal, and a variable-amplification inverting amplifier that outputs a signal whose phase is inverted by 180 ° with respect to the input signal (first (2) inverting amplification means 12) and the cutoff frequency variable type high-pass filter (high-pass filtering means) 11 control (variable) the cutoff frequency by the control signal 14, and the amplification degree variable inverting amplifier ( The level detector (second signal level detection control means) 13 that controls (varies) the amplification factor of the second inverting amplification means 12 by the control signal 15, the adder 7, and the output terminal 16.

Next, the operation of the audio processing circuit configured as described above will be described. The audio input signal 1 input to the input terminal 2 is passed through the buffer amplifier 3 installed so that the characteristics of both the low frequency band limiting unit 17 and the high frequency band limiting unit 18 do not change due to mutual interference. Cutoff frequency variable low pass filter (low pass filtering means) 4 that constitutes the low pass limiter 17, and cutoff frequency variable high pass filter (high pass filter) that constitutes the high pass limiter 18. ) 11 respectively.

The audio input signal 1 supplied to the variable cut-off frequency type low-pass filter 4 is supplied to the variable-amplification inverting amplifier 5 after removing the high-frequency components above a predetermined cut-off frequency. Then, the amplitude is amplified by a predetermined amplification factor, the phase is inverted by 180 °, the result is output to the adder 6, the voice input signal 1 (original signal) input from the input terminal 2 is added, and the result is added to the adder 7. Is output.

On the other hand, a part of the signal supplied from the cutoff frequency variable low pass filter 4 to the variable amplification type inverting amplifier 5 is sent to the level detector (first signal level detection control means) 8. It is supplied and level detection is performed.

When the amount of detection (signal strength) is large, the level detector 8 sets the cutoff frequency of the cutoff frequency variable low-pass filter 4 to a high value by the control signal 4 to attenuate the cutoff frequency. Control is performed so that the amplification factor of the variable amplification type inverting amplifier 5 is equal to the original signal by the control signal 10 and the phase is inverted by 180 ° (amplification factor: -1). To do. And
In the adder 6, the original signal input from the input terminal 2 is added, and unnecessary band components of the original signal are attenuated (pass band is narrow) and output to the adder 7.

When the detected amount (signal intensity) is small, the level detector 8 controls the cutoff frequency of the cutoff frequency variable low pass filter 4 by the control signal 14
By setting it lower, the attenuation band is narrowed,
The amplification factor of the variable amplification type inverting amplifier 5 is set smaller than that of the original signal by the control signal 15, and control is performed such that the phase is inverted by 180 ° (amplification factor: −1 to 0). Then, in the adder 6, the input terminal 2
The input signal is added to the input original signal, the unnecessary band component of the original signal is attenuated (the pass band is wide), and the added signal is output to the adder 7.

Similarly, the audio input signal 1 supplied to the variable cut-off frequency type low-pass filter 11 has its low-frequency component below a predetermined cut-off frequency removed, and then is fed to the variable amplification degree inverting amplifier 12. The signal is supplied, amplitude-amplified by a predetermined amplification factor, the phase is inverted by 180 ° and output to the adder 7, which is added to the signal supplied from the adder 6 and output to the output terminal 16.

On the other hand, a part of the signal supplied from the cutoff frequency variable low pass filter 11 to the variable gain inverting amplifier 12 is sent to the level detector (second signal level detection control means) 13. It is supplied and level detection is performed.

When the amount of detection (signal strength) is large, the level detector 8 sets the cutoff frequency of the variable cutoff frequency type low pass filter 11 to a low value so that the attenuation band is wide. , The amplification factor of the variable amplification type inverting amplifier 12 is controlled so that the signal is equal to the original signal and the phase is inverted by 180 ° (amplification factor: −1). Then, in the adder 7, the adder 6
The output signal is added to the output signal, the unnecessary band component of the output signal is attenuated, and output to the output terminal 16.

When the detected amount (signal intensity) is small, the level detector 8 sets the cutoff frequency of the cutoff frequency variable low-pass filter 11 to be high so that the attenuation band is narrowed. , The amplification factor of the variable amplification type inverting amplifier 12 is set smaller than that of the original signal, and control is performed such that the phase is inverted by 180 ° (amplification factor: −1 to 0). Then, in the adder 7, the signal supplied from the adder 6 is added, the unnecessary band component of the output signal is attenuated, and the result is output to the output terminal 16.

On the other hand, FIG. 2 is a graph showing frequency distributions of a voice input signal before and after the voice input signal passes through the voice processing circuit according to the present invention.

FIG. 2A shows the frequency distribution of the voice input signal 1 input from the input terminal 2,
It can be seen that the input signal band 31 extends over the entire input frequency range determined by the performance of the input device.

FIG. 2B shows the frequency distribution of the signal after the voice input signal 1 input from the input terminal 2 passes through the voice processing circuit of the present invention. It can be seen that the frequency band of the signal band 32 is narrowed (only the necessary frequency band is extracted). This makes it possible, for example, to remove various ambient noises, mechanical noises, and the like, and emphasize only the speaking voice of the speaker (in a state where it is easy to hear) to extract.

Further, as can be seen from FIG. 2 (b),
One of the characteristics of the audio processing circuit according to the present invention is that a signal having a low filter pass level, that is, a signal having a low signal pass through the filter has a small frequency spread (frequency band), and thus the signal level increases. It can be seen that the spread of frequency (frequency band) is increasing.

According to the characteristics of the voice processing circuit according to the present invention, the level of the voice signal of the speaker is generally smaller than the level of ambient noise (in the example of the talk of the speaker). Ambient noise (passes through to some extent), that is, the signal frequency band for a low-level signal is narrow, and the voice signal frequency band for a speaker's voice, that is, a high-level signal is wide, so that pickup (emphasis)
The desired sound (voice of the speaker in this example) can be obtained with high sound quality, and noise can be obtained with low sound quality.

According to the above-described embodiment of the present invention, it has been described that both the low frequency band limiting unit 17 and the high frequency band limiting unit 18 are used as the configuration of the audio processing circuit of the present invention, but the low frequency band limiting unit 17 is used. Of course, it may be an audio processing circuit configured by using only the high frequency band limiting unit 18. In this case,
The band will be limited for either the low band or the high band.

[0048]

As described above, according to the present invention, it becomes possible to appropriately and optimally limit the band of a voice input signal, and a necessary voice component can be provided without providing a plurality of voice input means. Only the (frequency band) can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a voice processing circuit of the present invention.

FIG. 2 is a graph showing frequency distributions of a signal before and after a voice input signal passes through a voice processing circuit according to the present invention.

FIG. 3 is a graph showing the relationship between the power of voice and the intelligibility by Fletcher's filter.

FIG. 4 is a graph showing the relationship between the level of voice strength and clear speech.

FIG. 5 is a graph showing the relationship between the level of voice volume and intelligibility.

FIG. 6 is a block diagram showing an example of a conventional audio processing circuit.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Audio input signal 2 ... Input terminal 3 ... Buffer amplifier 4 ... Cutoff frequency variable type low pass filter (low pass filter means) 5 ... Amplification degree variable inverting amplifier (first inverting amplifier means) 6, 7 ... Adder 8 ... Level detector (first signal level detection control means) 9, 10, 14, 15 ... Control signal 11 ... Cutoff frequency variable type high-pass filter (high-pass filtering means) 12 ... Amplification variable type Inversion amplifier (second inverting amplification means) 13 ... Level detector (second signal level detection control means) 16 ... Output terminal

Claims (8)

[Claims] 1. Input means for inputting an audio signal, low-pass filtering means for extracting a low-frequency component from the audio signal, the cut-off frequency being variable, and a signal output from the low-pass filtering means. First inverting amplification means for changing the amplification factor for inverting amplification, and detecting the signal level output from the low-pass filtering means, and cutting the low-pass filtering means according to the detected signal level. First signal level detection control means for controlling one or both of an off frequency and an amplification factor of the first inverting amplification means, a signal output from the first inverting amplification means, and the input means A voice processing circuit, comprising: a first adding means for adding the voice signal input from the. 2. The low-pass filtering means and the first inverting amplification means are integrally formed.
The voice processing circuit according to. 3. Input means for inputting an audio signal, high-pass filtering means capable of changing a cut-off frequency for extracting a high-frequency component from the audio signal, and a signal output from the high-pass filtering means. Second inverting amplification means, which can change the amplification factor for inverting amplification, detects the signal level output from the high-pass filtering means, and cuts the high-pass filtering means according to the detected signal level. Second signal level detection control means for controlling either or both of an off frequency and an amplification factor of the second inverting amplification means; a signal output from the second inverting amplification means; and the input An audio processing circuit comprising: a second adding means for adding the audio signal input from the means. 4. The high-pass filtering means and the second inverting amplification means are integrated with each other.
The voice processing circuit according to. 5. Input means for inputting an audio signal, low-pass filtering means capable of changing a cut-off frequency for extracting a low-frequency component from the audio signal, and a signal output from the low-pass filtering means. First inverting amplification means for inverting amplification, the amplification factor of which is changeable, and a signal level output from the low-pass filtering means is detected, and the low-pass filtering means of the low-pass filtering means is detected according to the detected signal level. A first signal level detection control means for controlling either or both of a cutoff frequency and an amplification factor of the first inverting amplification means; a signal output from the first inverting amplification means; First adding means for adding the audio signal input from the input means, high-pass filtering means for extracting a high-frequency component from the audio signal input by the input means, the cut-off frequency being variable, Inverting amplifying a signal output from the pass filter means,
Second inverting amplification means capable of changing the amplification factor, and detecting the signal level output from the high-pass filtering means, and depending on the detected signal level, the cut-off frequency of the high-pass filtering means and the first Second signal level detection control means for controlling one or both of the amplification factors of the second inverting amplification means, the signal output from the second inverting amplification means, and the first addition means. A voice processing circuit, comprising: a second adding means for adding the output signal. 6. The low-pass filtering means and the first inverting amplifying means, and the high-pass filtering means and the second inverting amplifying means, or one of them is integrally formed. The audio processing circuit according to claim 5. 7. The cutoff frequency of the low-pass filtering means is set high when the signal level detected by the first signal level detection control means is high, and the cutoff frequency of the low-pass filtering means is set high when the signal level is low. 7. The voice processing circuit according to claim 1, wherein the cutoff frequency of the low-pass filtering means is set low and the amplification factor of the first inverting amplification means is set small. 8. The cutoff frequency of the high-pass filtering means is set low when the signal level detected by the second signal level detection control means is high, and the cutoff frequency of the high-pass filtering means is set low when the signal level is low. 8. The cutoff frequency of the high-pass filtering means is set high and the amplification factor of the second inverting amplification means is set small.
The voice processing circuit according to.