JP2005269543A - Bass enhancing circuit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a bass enhancing circuit, capable of conducting optimal bass enhancement according to the type of the input voice signal. <P>SOLUTION: The bass enhancing circuit generates harmonics of the even-numbered orders for low audio band components with a full-wave rectifying circuit, and emphasizes bass using the harmonics obtained. The bass enhancing circuit comprises a features amount detecting means for detecting the amount of features in a voice signal passing a first band-pass filter, and a gain adjusting means for adjusting the gain of a first amplifier and the gain of a second amplifier, based on the amount of features detected by the features amount detection means. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a bass emphasis circuit that can enhance bass feeling when a small speaker or headphones with low bass reproduction capability is used.

[A] Description of Prior Art When listening to a sound consisting of signals 2f, 3f, 4f,..., Which are harmonics of a signal having a reference frequency f, the frequency of the reference frequency f is not included even if it is not included. An auditory psychological phenomenon is known in which the sound of f is felt as being reproduced.
There are various methods for generating harmonics. As the simplest method, there is known a method of generating even-order harmonics by full-wave rectifying a signal of a reference frequency.

  By the way, the bass reproduction capability is restricted mainly by the volume of the speaker. Therefore, a small speaker used in a television or the like cannot obtain a sufficient bass reproduction capability. As a method for solving this problem, there is a method for causing a human to sense a reference frequency signal that cannot be reproduced by a speaker due to an auditory psychological phenomenon caused by a harmonic signal. In Japanese Patent Laid-Open No. 08-237800, an even-order harmonic of the low-frequency component is generated by full-wave rectifying only the low-frequency component below the speaker reproduction limit in the input audio signal. A technique is disclosed in which a low-frequency feeling is enhanced by amplifying a second-order harmonic of the harmonics and adding it to an input audio signal.

  However, this conventional technique has the following problems, so that even if bass can be recognized, a natural sound cannot be obtained.

(1) Even in a loudspeaker used in a normal large TV, the reproduction lower limit frequency is about 150 to 180 Hz. For this reason, harmonics with respect to the harmonics of the instrument sound are also generated, but when the harmonics of the instrument sound is reproduced, the sound becomes unnatural. Moreover, when the harmonic of a low frequency is produced | generated, harmonics will become a dissonance. When harmonics with respect to a wide-band reference sound are generated, various harmonic components may be mixed and the bass part may not be recognized.
(2) When only harmonics are amplified and added to the input audio signal in music or the like, the sound becomes unnatural. In addition, the envelope characteristics of the harmonics generated by full-wave rectification are fixed, and natural sound cannot be obtained.

  In view of this, the present inventor has developed a bass emphasis circuit that can solve the above-mentioned problems, obtain a natural harmonic with coherency, and obtain a natural reproduced sound.

[B] Description of bass enhancement circuit already developed by the present inventor The bass enhancement circuit already developed by the present inventor will be described. This bass emphasis circuit is not known at the time of filing of the present application and is not a conventional technique.

[B1] Description of the first bass enhancement circuit already developed by the present inventor [B1-1] Description of the configuration of the bass enhancement circuit FIG. 1 shows the configuration of the first bass enhancement circuit already developed by the present inventor. Is shown.

  An audio signal composed of an L channel signal (Lch in) and an R channel signal (Rch in) is input to the bass enhancement circuit. The L channel signal is amplified by the amplifier 1L. The L channel signal amplified by the amplifier 1L is sent to the adder 3L via the delay circuit 2L. The L channel signal amplified by the amplifier 1L is also sent to the adder 11. The R channel signal is amplified by the amplifier 1R. The R channel signal amplified by the amplifier 1R is sent to the adder 3R via the delay circuit 2R. The R channel signal amplified by the amplifier 1R is also sent to the adder 11.

  The adder 11 adds the L channel signal amplified by the amplifier 1L and the R channel signal amplified by the amplifier 1R. The output of the adder 11 is sent to the full-wave rectifier circuit 13 through a bandpass filter (BPF1) 12 that passes only a signal in a predetermined frequency band. The characteristics and functions of the bandpass filter (BPF1) 12 will be described later. The full-wave rectifier circuit 13 generates an even-order harmonic signal of the signal that has passed through the bandpass filter (BPF1) 12. The harmonic signal generated by the full-wave rectifier circuit 13 is sent to a low-pass filter (LPF1) 14 for cutting a high-frequency component.

  If full-wave rectification is performed on an arbitrary signal (sin θ), even-order harmonics are generated as shown in the following equation (1).

abs [sinθ] = 2 / π + 4 / π × {(1/3) × sin2θ− (1/15) × sin4θ + (1/35) × sin6θ…}
... (1)

  In the above formula (1), abs [A] means that the absolute value of A is taken.

  The signal output from the low-pass filter (LPF1) 14 is sent to the adder 3L via the amplifier 15L and is sent to the adder 3R via the amplifier 15R.

  In the adder 3L, the output signal (original sound signal) of the delay circuit 2L and the output signal (harmonic signal) of the amplifier 15L are added. The output signal of the adder 3L is sent to a high pass filter (HPF) 4L for cutting a low frequency component. The characteristics and functions of the high pass filter (HPF) 4L will be described later. The signal that has passed through the high-pass filter (HPF) 4L is sent to the adder 8L via the delay circuit 5L. The signal that has passed through the high-pass filter (HPF) 4L is sent to the adder 8L via a low-pass filter (LPF2) 6L for cutting off high-frequency components and an amplifier 7L. The characteristics and functions of the low-pass filter (LPF2) 6L will be described later. In the adder 8L, the output signal of the delay circuit 5L and the output signal of the amplifier 7L are added and output to an L channel speaker (not shown) as an L channel output signal (Lch out).

  Similarly, in the adder 3R, the output signal (original sound signal) of the delay circuit 2R and the output signal (harmonic signal) of the amplifier 15R are added. The output signal of the adder 3R is sent to a high pass filter (HPF) 4R for cutting a low frequency component. The signal that has passed through the high-pass filter (HPF) 4R is sent to the adder 8R via the delay circuit 5R. The signal that has passed through the high-pass filter (HPF) 4R is sent to the adder 8R via a low-pass filter (LPF2) 6R for cutting off high-frequency components and an amplifier 7R. In the adder 8R, the output signal of the delay circuit 5R and the output signal of the amplifier 7R are added and output to an R channel speaker (not shown) as an R channel output signal (Rch out).

[B1-2] Explanation of each filter

[B1-2-1] Description of Bandpass Filter (BPF1) 12 The bandpass filter (BPF1) 12 is a filter that passes only a signal in a predetermined frequency band from a combined signal of an L channel signal and an R channel signal. is there.

  By the way, the reproduction lower limit frequency of the speaker is often 100 Hz to 200 Hz. Of the harmonic signals generated by the full-wave rectifier circuit 13, the upper limit frequency of the harmonics used for bass emphasis is about 500 Hz to 600 Hz. The critical bandwidth of the signal in the frequency band from the reproduction lower limit frequency (100 Hz to 200 Hz) of the speaker to the upper limit frequency (500 Hz to 600 Hz) of the harmonic to be used for bass enhancement is about 80 to 100 Hz. It is known that dissonance occurs when two sounds having a frequency difference less than the critical bandwidth are heard. Also, the smaller the frequency difference, the stronger the incoherence and the more harsh sound you can hear.

  Therefore, in this embodiment, the lower limit frequency of the bandpass filter (BPF1) 12 is set to the critical bandwidth of the signal in the frequency band from the lower limit frequency of reproduction of the speaker to the upper limit frequency of the harmonic to be used for bass enhancement. It is set to 45 Hz, which is about 1/2. By setting the lower limit frequency of the bandpass filter (BPF1) 12 in this way, the frequency interval between the harmonic signals of the adjacent orders among the even harmonic signals generated by the full-wave rectifier circuit 13 is the critical value. It becomes 90 Hz or more corresponding to the bandwidth, and no dissonance is generated.

  In general, musical instrument sounds originally contain harmonics. Therefore, if the upper limit frequency of the bandpass filter (BPF1) 12 is set high, the full-wave rectifier circuit 13 causes the reference wave of the instrument sound and its harmonics to be set. Second harmonics are generated for both waves, and unnatural sound is reproduced. Therefore, in this embodiment, a value not more than twice the lower limit frequency of the bandpass filter (BPF1) 12 is set as the upper limit frequency of the bandpass filter (BPF1) 12. Specifically, the upper limit frequency of the band pass filter (BPF 1) 12 is set to 90 Hz corresponding to twice the lower limit frequency 45 Hz of the band pass filter (BPF 1) 12.

[B1-2-2] Description of Low-Pass Filter (LPF1) 14 The low-pass filter (LPF1) 14 is provided for cutting high-frequency components of the harmonic signal generated by the full-wave rectifier circuit 13. ing. The cut-off frequency of the low-pass filter (LPF) 14 is set to be equal to or lower than the cut-off frequency of the low-pass filters (LPF2) 6L and 6R.

[B1-2-3] Description of High-Pass Filters (HPF) 4L and 4R The high-pass filters (HPF) 4L and 4R are speakers to be used among signals obtained by adding a low-frequency emphasis harmonic signal to the original sound signal. It is provided to cut bass signals in a band that cannot be reproduced by The high-pass filters (HPF) 4L and 4R also have a function of suppressing signal saturation due to amplification of the amplifiers 7L and 7R.

[B1-2-4] Description of Low-Pass Filters (LPF2) 6L and 6R The cutoff frequency of the low-pass filters (LPF2) 6L and 6R is set to a value larger than the cutoff frequency of the low-pass filter (LPF1) 14. .

  The low pass filters (LPF2) 6L and 6R cut high frequency components from the signals that have passed through the high pass filters (HPF) 4L and 4R. The signals that have passed through the low-pass filters (LPF2) 6L and 6R are amplified by the amplifiers 7L and 7R, and then combined with the signals that have passed through the high-pass filters (HPF) 4L and 4R.

  In other words, among the synthesized signal of the original sound signal and the harmonic signal for bass emphasis, the signal in the frequency band determined by the low pass filters (LPF2) 6L and 6R is amplified, and the amplified signal (enhanced signal) is the original sound. The signal is synthesized. In this way, rather than amplifying only the harmonic signal for bass emphasis, the predetermined band component of the synthesized signal of the original sound signal and the harmonic signal for bass emphasis is amplified. Compared to a case where only the harmonic signal is amplified, a sense of incongruity due to the harmonic is suppressed, and a powerful deep bass feeling can be enhanced as a ripple effect.

  However, if higher harmonics than necessary are emphasized, the sound becomes harsh, so the nth order and (n + 2) th order with respect to the reference frequency of 90 Hz, which is the upper limit frequency of the bandpass filter (BPF1) 12. The cut-off frequencies of the low-pass filters (LPF2) 6L and 6R are set so that two harmonics can pass. Let n be the order corresponding to the harmonic having the lowest frequency in the band equal to or higher than the reproduction lower limit frequency of the speaker among the even-order harmonics with respect to the reference frequency of 90 Hz. For a 45 Hz signal, a maximum of four harmonics pass through the low-pass filters (LPF2) 6L and 6R.

  For example, when a speaker that cannot reproduce a signal of 200 Hz or less is used, the harmonics for the signal with the upper limit frequency of 90 Hz that passes through the bandpass filter (BPF1) 12 are 180 Hz, 360 Hz, 540 Hz,. If there are fourth and sixth harmonics, a reference sound of 90 Hz can be recognized, so the cutoff frequencies of the low pass filters (LPF2) 6L and 6R are 540 Hz.

  In this case, the harmonics of the signal having the lower limit frequency of 45 Hz of the band pass filter (BPF1) 12 are 90 Hz, 180 Hz, 270 Hz, 360 Hz, 450 Hz, 540 Hz,... The 12th-order harmonics pass through the low-pass filters (LPF2) 6L and 6R.

  Table 1 shows the characteristics of the various filters with respect to the type of speaker used (reproduction lower limit frequency). In Table 1, BPF1 is the lower limit frequency and upper limit frequency of the bandpass filter 12, LPF1 is the cutoff frequency of the lowpass filter 13, HPF is the cutoff frequency of the highpass filters 4L and 4R, and LPF2 is the lowpass filters 6L and 6R. The cut-off frequency is shown respectively.

  According to the first bass emphasis circuit, the bandpass filter (BPF1) 12 sets the frequency band of the reference signal from which harmonics are generated to the specific band as described above. And natural playback sound can be obtained. In addition, it is possible to prevent the generation of the second harmonic with respect to both the reference wave of the instrument sound and its harmonics, and a natural reproduced sound can be obtained.

  Further, after synthesizing the original sound signal and the low-frequency emphasis harmonic signal, the signal in the frequency band determined by the low-pass filters (LPF2) 6L and 6R is amplified, and the amplified signal (enhanced signal) is converted into the original sound signal. In other words, since the harmonics are emphasized together with the original sound, the sense of incongruity due to the harmonics can be suppressed. As a result, the recognition of the reference sound (bass) can be strengthened.

[B2] Description of Second Bass Emphasis Circuit Developed by the Inventor FIG. 2 shows a configuration of a second bass emphasis circuit already developed by the present inventor.

  The second bass enhancement circuit is different from the first bass enhancement circuit in the following points.

(1) The high-pass filters (HPF) 4L and 4R are arranged not before the adders 3L and 3R but before the adders 3L and 3R. Accordingly, instead of the low-pass filter (LPF) 14 provided at the subsequent stage of the full-wave rectifier circuit 13, a bandpass having both functions of the low-pass filter (LPF1) 14 and the high-pass filters (HPF) 4L and 4R. The filter (BPF2) 16 was used.

(2) In order to further alleviate the uncomfortable feeling caused by emphasizing only a specific band of the original sound, the low-pass filter (LPF3) is connected in parallel with the series circuit of the low-pass filter (LPF2) 6L (6R) and the amplifier 7L (7R). ) A series circuit of 9L (9R) and amplifier 10L (10R) is provided. The characteristics of the low-pass filter (LPF2) 6L (6R) and the low-pass filter (LPF3) 9L (9R) in this case are set as shown in Table 2, for example. Further, the gain of the amplifier 7L (7R) that amplifies the signal that has passed through the low-pass filter (LPF2) 6L (6R) is G1, and the amplifier 10L (10R) that amplifies the signal that has passed through the low-pass filter (LPF3) 9L (9R). When the gain is G2, G1> G2 is set.

Japanese Patent Laid-Open No. 08-237800

  The above-described bass emphasis circuit developed by the present applicant emphasizes bass by high frequency synthesis and bass amplification.

  By the way, the high frequency has an effect of recognizing a specific difference frequency, and the bass amplification has an effect of enhancing the overall low tone. Therefore, it can be said that the former is effective for musical instrument sounds and the latter is effective for explosion sounds. For this reason, regardless of the type (characteristics) of the low frequency band in the input audio signal, the optimum effect cannot be obtained if the processing is fixed. In addition, the signal level may exceed the effective bit length by performing high frequency synthesis and bass amplification.

  It is an object of the present invention to provide a bass emphasis circuit that can perform optimum bass emphasis according to the type of input audio signal.

  It is another object of the present invention to provide a bass emphasis circuit that can perform optimum bass emphasis according to the type of input audio signal and suppress overflow.

  According to the first aspect of the present invention, a first adder that synthesizes input audio signals of a plurality of channels, and a first signal that passes only a signal in a predetermined frequency band among the audio signals synthesized by the first adder. Of the harmonic signals generated by the band-pass filter, the harmonic signal generated by the full-wave rectifier circuit, the full-wave rectifier circuit that generates even-order harmonic signals of the audio signal that has passed through the first band-pass filter, is used for bass enhancement A low-pass filter for cutting a signal in a high frequency band exceeding the upper limit frequency of a harmonic to be tried, a first amplifier provided for each channel and amplifying the harmonic signal that has passed through the low-pass filter, and each channel The harmonic signal provided by each and amplified by the first amplifier corresponding to the channel is converted into an input audio signal corresponding to the channel. A plurality of second adders to be added, a plurality of which is provided for each channel and cuts a signal in a low frequency band that cannot be reproduced by a speaker to be used from an output signal of the second adder corresponding to the channel. A plurality of second amplifiers that are provided for each channel and that amplify a low-frequency band signal among signals that have passed through the high-pass filter corresponding to the channel, and are provided for each channel, and In a bass emphasis circuit including a plurality of third adders that synthesize a signal that has passed through a high-pass filter corresponding to a channel and an output signal of a second amplifier corresponding to the channel, the first band-pass filter includes: A feature amount detecting means for detecting a feature amount of a passing voice signal, and a feature amount detected by the feature amount detecting means. Zui it, characterized in that it comprises a gain adjusting means for adjusting the gain of the gain and the second amplifier of the first amplifier.

  According to a second aspect of the present invention, a plurality of first amplifiers that respectively amplify a plurality of channels of input audio signals, a first adder that combines the audio signals of each channel amplified by the first amplifiers, The first band-pass filter that passes only a signal in a predetermined frequency band among the audio signals synthesized by the adder, and all the harmonic signals of the even order of the audio signal that has passed through the first band-pass filter Low-pass filter for cutting high-frequency band signals exceeding the upper limit frequency of harmonics to be used for bass emphasis among harmonic signals generated by wave rectifiers and full-wave rectifiers, for each channel And a second amplifier that amplifies the harmonic signal that has passed through the low-pass filter, and is provided for each channel. A plurality of second adders for adding each of the harmonic signals amplified by the corresponding second amplifier to the audio signal amplified by the first amplifier corresponding to the channel; A plurality of high-pass filters that cut low-frequency band signals that cannot be reproduced by the speakers used from the output signal of the second adder corresponding to the channel, provided for each channel, and passed through the high-pass filter corresponding to the channel A plurality of third amplifiers for amplifying a signal in a low frequency band among the processed signals, and a third amplifier corresponding to the channel provided for each channel and passing through a high-pass filter corresponding to the channel In the bass emphasis circuit comprising a plurality of third adders for synthesizing the output signals of An overflow determination means for determining whether or not the addition result of the corresponding third adder exceeds the number of effective bits of the output signal, and a feature amount of the audio signal passing through the first bandpass filter. The gain of the first amplifier, the gain of the second amplifier, and the gain of the third amplifier are determined based on the feature quantity detection means to be detected, the discrimination result of the overflow discrimination means and the feature quantity detected by the feature quantity detection means. Gain adjusting means for adjusting is provided.

  According to a third aspect of the present invention, a first adder that synthesizes input audio signals of a plurality of channels, and a first signal that passes only a signal in a predetermined frequency band among the audio signals synthesized by the first adder. Of the harmonic signals generated by the band-pass filter, the harmonic signal generated by the full-wave rectifier circuit, the full-wave rectifier circuit that generates even-order harmonic signals of the audio signal that has passed through the first band-pass filter, is used for bass enhancement A second band-pass filter that cuts a signal in a high frequency band exceeding an upper limit frequency of a harmonic to be tried and cuts a signal in a low frequency band that cannot be reproduced by a speaker to be used; A first amplifier for amplifying the harmonic signal that has passed through the two band-pass filters, provided for each channel and corresponding to that channel; Among the audio signals, a high-pass filter that cuts a signal in a low frequency band that cannot be reproduced by a speaker to be used, and a signal that is provided for each channel and that has passed through the high-pass filter corresponding to that channel, corresponds to the first corresponding to that channel. A plurality of second adders for adding the harmonic signals amplified by the amplifiers of the first and second amplifiers provided for each channel, and amplifying a signal in a low frequency band among the output signals of the second adders corresponding to the channels. A plurality of second amplifiers, and a plurality of second amplifiers that are provided for each channel and synthesize the output signal of the second amplifier corresponding to the channel with the output signal of the second adder corresponding to the channel. In the bass emphasis circuit having the adder 3, a feature for detecting the feature amount of the audio signal passing through the first bandpass filter Amount detecting means, and on the basis of the characteristic amount detected by the feature quantity detecting means, characterized in that it comprises a gain adjusting means for adjusting the gain of the gain and the second amplifier of the first amplifier.

  According to a fourth aspect of the present invention, a plurality of first amplifiers for amplifying a plurality of channels of input sound signals, a first adder for combining the sound signals of each channel amplified by the first amplifiers, The first band-pass filter that passes only a signal in a predetermined frequency band among the audio signals synthesized by the adder, and all the harmonic signals of the even order of the audio signal that has passed through the first band-pass filter Of the harmonic signals generated by the wave rectifier circuit and full-wave rectifier circuit, the signal in the high frequency band exceeding the upper limit frequency of the harmonics to be used for emphasizing the bass is cut and cannot be reproduced by the speakers used. A second bandpass filter that cuts a signal in a low frequency band, a harmonic signal provided for each channel and passed through the second bandpass filter A second amplifier for amplifying each of the signals, and a high pass for cutting a signal in a low frequency band that cannot be reproduced by a speaker to be used from among the audio signals provided for each channel and amplified by the first amplifier corresponding to the channel. A plurality of second adders for adding a harmonic signal amplified by a second amplifier corresponding to the channel to a signal that is provided for each channel and that has passed through a high-pass filter corresponding to the channel; A plurality of second amplifiers for amplifying a low frequency band signal among the output signals of the second adder corresponding to the channel, and provided for each channel, and provided for each channel. The output signal of the second amplifier corresponding to the channel is synthesized with the output signal of the corresponding second adder. In a bass emphasis circuit including a plurality of third adders, an overflow determination is provided for each channel and determines whether or not the addition result of the corresponding third adder exceeds the number of effective bits of the output signal. Means, a feature amount detection means for detecting a feature amount of an audio signal passing through the first bandpass filter, and a first amplifier based on the discrimination result of the overflow discrimination means and the feature amount detected by the feature amount detection means And gain adjustment means for adjusting the gain of the second amplifier and the gain of the third amplifier.

  According to a fifth aspect of the present invention, in the first to fourth aspects of the present invention, the feature amount detection means includes a frequency component having a frequency that is contained most in the signal component that has passed through the first bandpass filter. It is characterized in that the ratio or difference between the spectrum and the average frequency spectrum of the entire signal that has passed through the first bandpass filter is detected as a feature quantity.

  According to the present invention, optimum bass enhancement can be performed according to the type of input audio signal.

  Further, according to the present invention, it is possible to perform the optimum bass emphasis according to the type of the input audio signal and to suppress the overflow.

  Hereinafter, embodiments of the present invention will be described with reference to FIGS.

[1] Description of Configuration of Bass Emphasis Circuit FIG. 3 shows a configuration of the bass enhancement circuit in the first embodiment. In FIG. 3, the same components as those in FIG.

  The bass emphasis circuit of FIG. 3 differs from the bass emphasis circuit of FIG. 1 in the following points.

(1) A music feature quantity detection unit 21 for detecting a music feature quantity based on a signal passing through the bandpass filter (BPF1) 12 is provided.
(2) The gain adjusting unit 22 for adjusting the gain G1 of the amplifiers 15L and 15R and the gain G2 of the amplifiers 7L and 7R based on the music feature amount detected by the music feature amount detecting unit 21 is provided. .

  When the signal passing through the bandpass filter (BPF1) 12, that is, the reference signal for generating a high frequency is limited to a specific frequency, the effect of the reference signal recognition by the harmonics is great. On the other hand, when a reference signal exists in the entire band such as explosion sound, it is difficult to recognize the reference sound by using harmonics. Therefore, according to the output characteristics of the bandpass filter (BPF1) 12, the ratio of harmonics is increased in the former case, and the bass boost amount by the lowpass filters (LPF2) 6L and 6R is set large in the latter case. .

[2] Description of Operation of Music Feature Quantity Detection Unit 21 The music feature quantity detection unit 21 matches the upper limit frequency of the signal passing through the band pass filter (BPF 1) 12 every predetermined time, and the band pass filter (BPF 1). The output signal of the bandpass filter (BPF1) 12 is converted into frequency spectrum data by performing FFT after downsampling the 12 output signals. Then, the average value of the spectrum in the frequency band of the signal passing through the bandpass filter (BPF1) 12 (the average frequency spectrum of the entire signal that has passed through the bandpass filter 12) and the maximum value (the signal component that has passed through the bandpass filter 12). The frequency spectrum of the frequency component contained in the largest amount) is calculated, and the ratio H (= maximum value / average value) is calculated. Then, the calculated ratio H is given to the gain adjusting unit 22. Note that the difference between the maximum value and the average value may be calculated and given to the gain adjusting unit 22.

[3] Description of Operation of Gain Adjustment Unit 22 The gain adjustment unit 22 calculates gains G1 and G2 based on the following equation (2) every time the ratio H is given from the music feature amount detection unit 21. The gains of the amplifiers 15L and 15R are set to the calculated gain G1, and the gains of the amplifiers 7L and 7R are set to the calculated gain G2.

G1 = (G1_orig + ΔG1_h)
G2 = (G2_orig + ΔG2_h)
ΔG1_h = A × H
ΔG2_h = B / H (2)

  In the above equation (1), G1_orig is an initial value of the gain G1, and G2_orig is an initial value of the gain G2. A and B are arbitrary coefficients.

  When the ratio H of the maximum value to the average value of the spectrum of the frequency band of the signal passing through the band pass filter (BPF1) 12 is large, that is, when the reference signal for generating a high frequency is limited to a specific frequency. The gain G1 of the amplifiers 15L and 15R increases, and the gain G2 of the amplifiers 7L and 7R decreases.

  On the other hand, when the ratio H of the maximum value to the average value of the spectrum of the frequency band of the signal passing through the bandpass filter (BPF1) 12 is small, that is, when the reference signal exists in the entire band such as explosion sound, The gain G1 of the amplifiers 15L and 15R decreases, and the gain G2 of the amplifiers 7L and 7R increases.

[1] Description of Configuration of Bass Emphasis Circuit FIG. 4 shows a configuration of a bass enhancement circuit in the second embodiment. In FIG. 4, the same components as those in FIG.

  The bass emphasis circuit of FIG. 4 is different from the bass emphasis circuit of FIG. 1 in the following points.

(1) An overflow detector 31L is provided at the subsequent stage of the adder 8L, and an overflow detector 31R is provided at the subsequent stage of the adder 8R. (2) A signal passing through the bandpass filter (BPF1) 12 The music feature quantity detection unit 21 for detecting the music feature quantity is provided based on the above.
(3) Based on the detection signals of the overflow detection units 31L and 31R and the music feature quantity detected by the music feature quantity detection unit 21, the gain G0 of the amplifiers 1L and 1R, the gain G1 of the amplifiers 15L and 15R, and the amplifier 7L A gain adjusting unit 22 for adjusting the gain G2 of 7R is provided.

  When the gains G1 and G2 are controlled as in the first embodiment, the final output gain varies depending on the contents of the input signal, making it difficult to grasp the maximum output value, and the output signals of the adders 8L and 8R are effective. The bit length may be exceeded. Therefore, in the second embodiment, the final output signal level is reflected in gain control.

[2] Description of Operation of Music Feature Quantity Detection Unit 21 The music feature quantity detection unit 21 is the same as in the first embodiment, and the frequency band of the signal passing through the bandpass filter (BPF1) 12 every predetermined time. A ratio H (= maximum value / average value) of the maximum value to the average value of the spectrum is calculated and given to the gain adjustment unit 22.

[3] Description of Overflow Detection Units 31L and 31R For example, when the overflow detection units 31L and 31R output an output signal with 16-bit accuracy, the addition result of the adders 8L and 8R is the maximum value represented by 16 bits. Is exceeded, the overflow ratio over is calculated based on the following equation (3) and given to the gain adjusting unit 22. If the addition result (ADD) of the adders 8L and 8R is equal to or less than the maximum value (0x7FFF) represented by 16 bits, the overflow detection units 31L and 31R set the value of over to 0.

  over = (| ADD | −0x7FFF) / 0x7FFF (3)

[4] Description of Operation of Gain Adjustment Unit 22 The gain adjustment unit 22 performs gain adjustment processing as will be described later every time the ratio H is given from the music feature amount detection unit 21.

  A basic concept of gain adjustment processing performed by the gain adjustment unit 22 will be described. Basically, similarly to the first embodiment, the gain G1 of the amplifiers 15L and 15R and the gain G2 of the amplifiers 7L and 7R are controlled based on the music feature quantity detected by the music feature quantity detection unit 21. When over exceeds 0, the gains G1 and G2 are decreased. When over exceeds a predetermined value γ predetermined value greater than 0 (for example, “2”), the gain G0 of the amplifiers 1L and 1R is decreased. Note that the gain G0 is not further increased once it is lowered due to overflow. However, when the power is turned on after the power is turned off, the gains G0, G1, G2 are returned to their initial values.

  FIG. 5 shows a gain adjustment processing procedure by the gain adjustment unit 22.

The gain adjustment unit 22 is initialized when the power is turned on (step S1). That is, the gains G0, G1, and G2 are set to initial values G0_orig, G1_orig, and G2_orig, respectively. In addition to ΔG1_h and ΔG2_h expressed by the above equation (1), initial values are set for the correction value ΔG0 _{t for} the gain G0, the correction value ΔG1 _{t for} the gain G1, and the correction value ΔG2 _{t for} the gain G2, respectively. The initial values of these variables are as follows:

ΔG1_h = ΔG2_h = 0,
ΔG0 _t = ΔG1 _t = ΔG2 _t = 1

  Note that the constants α1, α2, β1, β2, γ, and n used for the gain adjustment processing are set as follows, for example.

α1 = α2 = 1.0,
β1 = β2 = 0.0,
γ = 2.0, n = 0.25

  Next, when the ratio H is calculated by the music feature amount detection unit 21 and given to the gain adjustment unit 22 (step S2), the gain adjustment unit 22 calculates ΔG1_h and ΔG2_h based on the above equation (2). (Step S3).

  Next, it is determined whether or not the larger over (hereinafter simply referred to as “over”) of the overs given from the overflow detection units 31L and 31R is larger than γ (step S4). If over is less than or equal to γ, it is determined whether or not over is greater than 0 (step S6). If over is 0 or less, the process proceeds to step S8, and gains G0, G1, and G2 are calculated and set. Then, the process returns to step S2.

If it is determined in step S4 that over is greater than γ, a process for calculating a correction value ΔG0 _t for the gain G0 is performed (step S5). Then, the process proceeds to step S6. When it is determined in step S6 that over is greater than 0, a process for calculating the correction value ΔG1 _t of the gain G1 and the correction value ΔG2 _t of the gain G2 is performed (step S7). Then, the process proceeds to step S8.

  The processing contents of step S5, step S7 and step S8 will be described in detail.

In step S5, after calculating the correction value ΔG0 _t of the gain G0 based on the following equation (4), γ is set as the over value (over = γ).

ΔG0 _t = ΔG0 _t−1 × γ / over (4)

ΔG0 _t−1 is the previously calculated correction value ΔG0 _t .

In step S7, correction values ΔG1 _t and ΔG2 _t for the gains G1 and G2 are calculated based on the following equations (5) and (6).

ΔG1 _t = ΔG1 _t−1 × {α1 / (1 + over + β1)} ⁿ (5)
ΔG2 _t = ΔG2 _t−1 × {α2 / (1 + over + β2)} ⁿ (6)

ΔG1 _t−1 and ΔG2 _t−1 are the previously calculated correction values ΔG1 _t and ΔG2 _t .

  In step S8, first, gains G0, G1, and G2 are calculated based on the following equations (7), (8), and (9).

G0 = G0_orig + ΔG0t (7)
G1 = (G1_orig + ΔG1_h) × ΔG1 _t (8)
G2 = (G2_orig + ΔG2_h) × ΔG2 _t (9)

Then, the latest calculated correction values ΔG0 _t , ΔG1 _t , ΔG2 _t are set to the previously calculated correction values ΔG0 _t−1 , ΔG1 _t−1 , ΔG2 _t−1 . Thereafter, the process returns to step S2.

  In the present invention, as shown in FIG. 6 or FIG. 7, high-pass filters (HPF) 4L and 4R are arranged not before the adders 3L and 3R but before the adders 3L and 3R. Instead of the low pass filter (LPF) 14 provided at the subsequent stage of the rectifier circuit 13, a band pass filter (BPF 2) 16 having both functions of the low pass filter (LPF 1) 14 and the high pass filters (HPF) 4L, 4R is provided. The present invention can also be applied to the bass emphasis circuit used.

It is a block diagram which shows the structure of the 1st bass emphasis circuit which this inventor has already developed. It is a block diagram which shows the structure of the 2nd bass emphasis circuit which this inventor has already developed. It is a block diagram which shows the structure of the bass emphasis circuit which is 1st Example. It is a block diagram which shows the structure of the bass emphasis circuit which is 2nd Example. It is a flowchart which shows the gain adjustment processing procedure by the gain adjustment part 22 of FIG. It is a block diagram which shows another Example. It is a block diagram which shows other Example.

Explanation of symbols

1L, 1R First amplifier 2L, 2R Delay circuit 3L, 3R Adder 4L, 4R High pass filter (HPF)
5L, 5R delay circuit 6L, 6R Low-pass filter (LPF2)
7L, 7R Third amplifier 8L, 8R Adder 11 Adder 12 Band pass filter (BPF1)
13 Full-wave rectifier circuit 14 Low-pass filter (LPF1)
15L, 15R Second amplifier 16 Band pass filter (BPF2)
21 Music feature detection unit 22 Gain adjustment unit 31L, 31R Overflow detection unit

Claims (5)

A first adder that synthesizes input audio signals of a plurality of channels, a first bandpass filter that passes only a signal in a predetermined frequency band among the audio signals synthesized by the first adder, and a first bandpass The full-wave rectifier circuit that generates even-order harmonic signals of the audio signal that has passed through the filter. Of the harmonic signals generated by the full-wave rectifier circuit, the upper limit frequency of the harmonics to be used for bass emphasis. A low-pass filter for cutting a signal in a high frequency band exceeding, a first amplifier that is provided for each channel and that amplifies a harmonic signal that has passed through the low-pass filter, provided for each channel, and for each channel A plurality of second signals, each of which adds the harmonic signal amplified by the corresponding first amplifier to the input audio signal corresponding to the channel. Adders, provided for each channel, and provided for each channel, a plurality of high-pass filters for cutting low frequency band signals that cannot be reproduced by the speakers used from the output signal of the second adder corresponding to that channel And a plurality of second amplifiers that amplify a signal in a low frequency band among signals that have passed through a high-pass filter corresponding to the channel, and a high-pass filter that is provided for each channel and that corresponds to the channel In a bass emphasis circuit comprising a plurality of third adders for synthesizing the output signal of the second amplifier corresponding to the channel with the signal,
Feature quantity detection means for detecting a feature quantity of an audio signal passing through the first bandpass filter, and gain of the first amplifier and gain of the second amplifier based on the feature quantity detected by the feature quantity detection means Gain adjusting means for adjusting
A bass enhancement circuit characterized by comprising: A plurality of first amplifiers that respectively amplify a plurality of channels of input audio signals, a first adder that combines the audio signals of each channel amplified by each first amplifier, and an audio signal synthesized by the first adder Among them, a first band-pass filter that passes only a signal in a predetermined frequency band, a full-wave rectifier circuit that generates an even-order harmonic signal of an audio signal that has passed through the first band-pass filter, and a full-wave rectifier circuit A low-pass filter for cutting high-frequency band signals that exceed the upper limit frequency of the harmonics to be used for bass emphasis among the generated harmonic signals, provided for each channel and passing through the low-pass filter A second amplifier for amplifying the harmonic signals respectively provided by the second amplifier provided for each channel and corresponding to the channel. A plurality of second adders for adding the amplified harmonic signal to the audio signal amplified by the first amplifier corresponding to the channel, provided for each channel, and corresponding to the channel. A plurality of high-pass filters that cut a low-frequency band signal that cannot be reproduced by the speaker used from the output signal of the adder of No. 2, and a low-frequency signal that is provided for each channel and that has passed through the high-pass filter corresponding to that channel The output signal of the third amplifier corresponding to the channel is synthesized with the plurality of third amplifiers that amplify the band signal and the signal that is provided for each channel and that has passed through the high-pass filter corresponding to the channel. In a bass emphasis circuit including a plurality of third adders,
An overflow discriminating means provided for each channel, for discriminating whether or not the addition result of the corresponding third adder exceeds the number of effective bits of the output signal;
A feature amount detecting means for detecting a feature amount of an audio signal passing through the first bandpass filter, and a gain of the first amplifier based on the determination result of the overflow determining means and the feature amount detected by the feature amount detecting means Gain adjusting means for adjusting the gain of the second amplifier and the gain of the third amplifier;
A bass enhancement circuit characterized by comprising: A first adder that synthesizes input audio signals of a plurality of channels, a first bandpass filter that passes only a signal in a predetermined frequency band among the audio signals synthesized by the first adder, and a first bandpass The full-wave rectifier circuit that generates even-order harmonic signals of the audio signal that has passed through the filter. Of the harmonic signals generated by the full-wave rectifier circuit, the upper limit frequency of the harmonics to be used for bass emphasis. A second band-pass filter that cuts a signal in a high frequency band that exceeds and cuts a signal in a low-frequency band that cannot be reproduced by a speaker to be used; a harmonic that is provided for each channel and that has passed through the second band-pass filter A first amplifier for amplifying each of the wave signals, provided for each channel and used among the input audio signals corresponding to the channel; A high-pass filter that cuts a signal in a low frequency band that cannot be reproduced by a speaker, a harmonic that is provided for each channel and that has passed through the high-pass filter corresponding to that channel is amplified by the first amplifier corresponding to that channel. A plurality of second adders for adding wave signals, a plurality of second amplifiers provided for each channel and amplifying signals in the low frequency band among the output signals of the second adders corresponding to the channels And a plurality of third adders that are provided for each channel and synthesize the output signal of the second amplifier corresponding to the channel with the output signal of the second adder corresponding to the channel. In the bass enhancement circuit
Feature quantity detection means for detecting a feature quantity of an audio signal passing through the first bandpass filter, and gain of the first amplifier and gain of the second amplifier based on the feature quantity detected by the feature quantity detection means Gain adjusting means for adjusting
A bass enhancement circuit characterized by comprising: A plurality of first amplifiers that respectively amplify a plurality of channels of input audio signals, a first adder that combines the audio signals of each channel amplified by each first amplifier, and an audio signal synthesized by the first adder Among them, a first band-pass filter that passes only a signal in a predetermined frequency band, a full-wave rectifier circuit that generates an even-order harmonic signal of an audio signal that has passed through the first band-pass filter, and a full-wave rectifier circuit Cut out the high frequency band signal that exceeds the upper limit frequency of the harmonic to be used for low frequency emphasis among the generated harmonic signals, and cut the low frequency band signal that cannot be reproduced by the speakers used. 2 band-pass filters, each of which is provided for each channel and amplifies the second harmonic signal that has passed through the second band-pass filter. Amplifier, provided for each channel, and provided for each channel, a high-pass filter that cuts a low frequency band signal that cannot be reproduced by the speaker to be used, among audio signals amplified by the first amplifier corresponding to that channel. A plurality of second adders for adding the harmonic signal amplified by the second amplifier corresponding to the channel to the signal passed through the high-pass filter corresponding to the channel, and provided for each channel, And a plurality of second amplifiers for amplifying signals in the low frequency band among the output signals of the second adder corresponding to the channel, and a second adder provided for each channel and corresponding to the channel A plurality of third adders for synthesizing the output signal of the second amplifier corresponding to the channel with the output signal of In and has bass enhancement circuit,
An overflow discriminating means provided for each channel, for discriminating whether or not the addition result of the corresponding third adder exceeds the number of effective bits of the output signal;
A feature amount detecting means for detecting a feature amount of an audio signal passing through the first bandpass filter, and a gain of the first amplifier based on the determination result of the overflow determining means and the feature amount detected by the feature amount detecting means Gain adjusting means for adjusting the gain of the second amplifier and the gain of the third amplifier;
A bass enhancement circuit characterized by comprising: The feature amount detection means includes a frequency spectrum of a frequency component that is contained most in the signal component that has passed through the first bandpass filter, and an average frequency spectrum of the entire signal that has passed through the first bandpass filter. 5. The bass enhancement circuit according to claim 1, wherein the ratio or difference is detected as a feature amount.

Images