JP3605706B2 - Sound signal reproducing method and apparatus - Google Patents

Description

[0001]
FIELD OF THE INVENTION
The present invention relates to a method and an apparatus for reproducing an audio signal of various audio devices, and more particularly to a method and an apparatus for improving reproduced sound quality in a high-frequency range.
[0002]
[Prior art]
The sound signal includes a sound source recorded in analog on a medium such as a compact cassette or a record, a compact disc (CD), a mini disc (MD), a digital compact cassette (DCC), a digital audio tape (DAT), a laser disc ( LD) and other media that are digitally recorded, as well as those using radio waves as the media such as FM broadcasts, BSs and CSs. These acoustic signals are transmitted through an amplifier or the like to an electroacoustic device such as a speaker or earphone. Reproduction is performed through the conversion element. Some electronic musical instruments, such as electronic musical instruments, convert sound information input through input means such as a keyboard into a digital signal, amplify the signal, and reproduce the digital signal through an electroacoustic transducer.
[0003]
2. Description of the Related Art Conventionally, how to reproduce reproduced sound faithfully to the original sound has been a major issue of acoustic signal reproduction technology. The human audible frequency band is said to be approximately 20 Hz to 20 kHz, and the frequency characteristics of the sound reproducing device have been standardized and designed accordingly. However, in the conventional analog audio equipment, there is a problem that designing an amplifier or an electroacoustic transducer becomes difficult and expensive when trying to widen a reproduction frequency band in order to faithfully reproduce an original sound. In addition, there may be a case where a limitation is imposed on the characteristics of the recording medium itself such as a tape or a limitation on a frequency assigned to a radio wave.
[0004]
On the other hand, digital audio equipment, which is the mainstream of sound reproduction technology in recent years, records and reproduces audio signals with high fidelity even if the medium is used repeatedly and there is no deterioration in the audio information even in the transmission path of the audio signals. The advantage is that it can be played.
[0005]
When digital audio equipment such as CDs began to appear, these digital audio equipment were standardized to sufficiently cover the audible frequency band of humans and the dynamic range of music, and the sound quality of the studio reached home as it was. With the advent of digital audio equipment, the fidelity in reproducing audio signals seemed to be satisfactory to a very high level. However, in digital audio information recording, the frequency band and dynamic range are limited by the sampling frequency and the number of quantization bits, so the audio information after digitization must be discarded from the original audio information. Become. This is essentially different from an analog recording in which the original information remains even if it is hidden by noise or accompanied by distortion.
[0006]
Such a lack of an audio signal due to an essential property of a digital device leaves a complaint from the viewpoint of reproducing an audio signal with high fidelity. Subsequent research has shown that it is necessary to further improve the resolution at low levels, and that recording at higher frequencies improves sound quality. Attempts have also been made to record information using high bits and high sampling frequencies. In addition, a 1-bit high-speed transmission such as 1.96 MHz, which is superior in principle to low-level linearity, is also being studied. In any case, in addition to such problems in digital devices, the problems of analog devices described above after D / A conversion are also added. At present, it is undeniable that it is still insufficient.
[0007]
As described above, the background of the recognition that the current sound reproducing device is inadequately auditory is that human auditory cells not only generate sound signals in a range called an audible frequency band but also a high frequency of 90 kHz. It has recently started to be said that people will react. That is, it has been considered that a person can feel a sound in a region exceeding 20 kHz, which is a so-called upper limit of the audible frequency band.
[0008]
Now, taking a digital audio device as an example, the high-frequency characteristics of the digital audio device are limited by the sampling frequency. For example, the sampling frequency of a CD is 44.1 kHz, and the upper limit on the high frequency side that can be reproduced at this sampling frequency is 20 kHz, which is approximately 1/2 of that. Accordingly, the spectrum of the reproduced audio signal is truncated at the boundary of 20 kHz. However, as described above, since a person can also sense a sound in a high-frequency range of 20 kHz or more, even if a signal in which such a high-frequency region is missing is reproduced, the sound is faithful to the original sound signal and the human hearing is perceived. It can be said that a sound signal that satisfies cannot be reproduced.
[0009]
FIG. 1 is a graph showing the reproduction state of the above-mentioned CD. FIG. 1A shows a spectrum of a natural music signal that includes a frequency component exceeding 20 kHz. On the other hand, FIG. 1B shows a spectrum of a reproduced sound of an audio signal digitally stored in a CD, and the frequency characteristic is sharply reduced at a boundary of 20 kHz. That is, it can be said that the spectrum does not substantially have a frequency band exceeding 20 kHz. Conventionally, there has been an expectation that reproducing this truncated portion will provide better reproduced sound quality. However, it is impossible or extremely difficult to restore the information once discarded by digitization.
[0010]
In addition, in analog audio equipment, even if the high frequency range is not cut off unlike digital equipment, the reproduction characteristics in a band exceeding 20 kHz are substantially attenuated, so that the original sound cannot be sufficiently reproduced. Have difficulty. Therefore, in order to obtain a reproduced sound that satisfies the sense of hearing, it can be said that analog devices need to supplement the treble range in some way as in digital devices.
[0011]
Conventionally, means for generating and adding a high-frequency signal having a frequency component exceeding a cut-off frequency of a digital device in order to improve the reproduction sound quality of a high-frequency range is disclosed in Japanese Patent Application Laid-Open No. 2-68773, entitled "Audio Signal Reproduction Apparatus". Is disclosed.
[0012]
This conventional device adds an energy component based on the knowledge that humans are affected by a frequency band exceeding the audible band as an energy component rather than discriminating it as a musical tone. It suggests adding noise or the like that can be easily obtained. The treble signal to be added does not necessarily have to be the treble tone signal of the original sound signal, and is equivalent to the truncated band or the band in which the reproduction characteristics are rapidly attenuated. It is suggested in the prior art that a reasonable result is obtained even when a noise signal having a spectrum is added.
[0013]
[Problems to be solved by the invention]
However, as described above, it is said that human auditory cells also respond to 90 kHz as described above. For a region exceeding a frequency range called a so-called audible frequency band, a spectrum of a signal component of a high-frequency range added or It has been found by the inventor's research that an auditory difference is caused by the additional method. There are various types of noise, and the type of noise to be added and the difference in the method of addition have a great effect on the reproduced sound quality.
[0014]
Further, it is considered necessary to add an energy component closely related to the audio information in the audible band in order to obtain a more natural reproduced sound quality.
[0015]
Furthermore, for example, in the case of playing sounds of various musical instruments, the sound quality of the musical instrument is formed from the distribution of harmonic components of the respective orders, and the sound quality specific to the musical instrument is formed. include. In addition, hitting, rubbing, blowing, and similar sounds that occur when playing a musical instrument also include a very large number of components exceeding the audible band. Therefore, if only the harmonic component is simply added, the manner in which the harmonic component is included differs depending on the type of musical instrument, and the balance of the reproduced sound quality in the high frequency range may be lost.
[0016]
Therefore, the present invention reduces or eliminates the frequency characteristic spectrum when reproducing an audio signal, in order to eliminate the unnaturalness that the spectrum of the frequency characteristic is extremely reduced in a high frequency range (analog device) or is missing (digital device). By adding a signal component that approximates the spectral component to the original audio signal, it is possible to improve the reproduction sound of the audio signal in audio equipment, especially the reproduction sound quality in the high frequency range, and reproduce the audio signal that is comfortable for the human ear. It is an object of the present invention to provide a method and apparatus for reproducing an acoustic signal.
[0017]
[Means for Solving the Problems]
In order to solve the above-mentioned problem, in the first invention of the present application, a noise component having a spectrum similar to a reduced or missing spectrum component at the time of reproduction is added to the original audio signal, and the type of the added noise component and a means for adding In order to obtain excellent playback sound quality by giving a device to the device.
[0018]
That is, according to the first invention of the present application, as a result of analyzing an acoustic signal, many high-frequency signal components are generated by factors such as hitting, rubbing, and blowing an object, and the nature of these sounds is random noise. Therefore, it is characterized in that a random or noise component similar to random noise is used as a signal to be added to the original audio signal.
[0019]
In the present invention, the signal to be added to the original sound signal is not particularly limited as long as it is a signal containing each frequency component at random, but it can be generally said that the most easily obtained signal is noise. Therefore, various types of noise generators and noise generators are used as random noise signal sources, and from the output of this signal source, the upper frequency range upper limit of the reproduction frequency band or the higher frequency range upper limit of the audible frequency band is exceeded. The output is selected using a filter means or the like, and this is added to the original sound signal. However, it is needless to say that the term “random” does not necessarily mean that the signal is randomly distributed in a strict sense, but may be various artificially generated signals called pseudo-random. No.
[0020]
Note that the audible frequency band referred to in this specification refers to a frequency band that can be heard as sound by human ears measured using a sine wave, and its value is approximately 16 Hz to 20 kHz. The upper limit of the audible frequency band measured using a sine wave is about 20 kHz. However, as described above, according to recent research, it can be said that in the spectrum of a music signal, a person can feel up to a frequency as high as about 90 kHz. This has been confirmed in experiments performed by the present inventors.
[0021]
The second invention of the present application is directed to a reproduction of an audio signal, in which a harmonic component of the audio signal is a higher harmonic range upper limit of a reproduction frequency band of the acoustic signal or a higher harmonic range upper limit of an audible frequency band. The reproduction is performed by adding a noise component having a spectrum in a frequency band exceeding the upper limit of the treble range of the reproduction frequency band or the upper limit of the treble range of the audible frequency band to the original acoustic signal. In addition, a noise component having a spectrum in a high frequency range of the reproduction frequency band or a frequency band exceeding the high frequency range of the audible frequency band selected by a filter from a noise generator or the like is used as a noise component to be added. , And "random" may be a pseudo-random signal, as in the first invention.
[0022]
As described above, according to the second aspect of the present invention, the original sound is replaced with the sound signal of the band which is conventionally truncated (digital reproduction device) or extremely attenuated (analog reproduction device) or exceeds the upper limit of the audible band. By adding both the harmonic component of the signal and the random noise component obtained from another signal source to the original sound signal to the original sound signal, the above-described problem of signal reproduction in the audio device is solved, We are trying to get even more natural reproduction sound quality.
[0023]
In a preferred embodiment of the present invention, the noise component is dynamically added in conjunction with the output of the original audio signal component in a specific frequency band.
[0024]
It is considered that the spectrum of the original acoustic signal has continuity between the spectrum on the high-frequency side of the audible frequency band and the spectrum in the region beyond the high-frequency range, as shown in FIG. Therefore, if a high frequency band of the original sound signal is selected and the level of the noise component added to the original sound signal is dynamically adjusted in accordance with the fluctuation of the output level, natural reproduction can be obtained. It is based on knowledge.
[0025]
Here, the frequency band of the specific high-frequency range of the original acoustic signal component is preferably selected from the lower limit frequency of the signal to be added, that is, a frequency band near 20 kHz, even from the continuity of spectrum, and 6 kHz to 20 kHz. The degree is suitable. Of course, in an analog audio device that outputs up to a frequency band of 20 kHz or more, a noise component may be added in conjunction with an output through a high-pass filter that passes 6 kHz or more without cutting the high frequency band at 20 kHz. .
[0026]
Also, the dynamics of the noise component to be added can be selected from various methods, such as simply adding the dynamics in proportion to the output of a specific frequency band, or adding with a specific non-linearity. This may be determined as appropriate according to the reproduction characteristics of the audio equipment to be used and the trial listening.
[0027]
In a further preferred embodiment of the present invention, a predetermined time delay is caused in the output of the treble range of the original audio signal, and the output with the time delay is added to the noise component added to the original audio signal. (Or a noise component and a harmonic component) as a level control signal for controlling the output level.
[0028]
This is because, when the inventor uses a variation in the output level of a specific frequency band of the original audio signal as a level control signal of a noise component to be added to the original audio signal as a result of a number of experiments, a time delay is generated and used. As a result, good reproduction sound quality can be obtained, and this time delay has been found to have a great effect on the reproduction sound quality. More specifically, by adjusting the level of a noise component to be added based on a signal passed through a time constant circuit having a time constant in a specific range, it is possible to satisfy human hearing in sound quality. Was confirmed as a result. A preferable range of the time constant is approximately 2.2 mS, and preferably, a range of 2.2 mS ± 40%. It has been found that when the time constant is shortened beyond this range, the reproduced sound quality becomes noisy, and when the time constant is increased, the reproduced sound quality becomes sticky and unclear.
[0029]
In the present invention, it is preferable to use thermal noise as a signal source of random noise or similar noise.
[0030]
Usually, as a means for generating random noise, a zener diode is generally used because the noise level is high and the number of amplification stages is small. However, as a result of the audition test, the random noise using the thermal noise as the sound source has a good result that the graininess of the sound is fine even though it has the same spectrum as compared with the random noise using the Zener diode. was gotten.
[0031]
The reason for this is that the noise source is electrons in both the Zener diode and thermal noise, but in the Zener diode the noise is obtained by the collective avalanche phenomenon of electrons due to the Avalanche effect, so the sound graininess is rough, On the other hand, thermal noise is generated due to fluctuations in the thermal motion of electrons inside the resistor, and in principle, it is considered that noise is generated randomly and uniformly, so that the granularity of sound becomes fine. The thermal noise is not necessarily obtained from the actual resistance element, but may be obtained from the base resistance inside the transistor or the channel resistance inside the FET.
[0032]
Further, the random noise added to the original audio signal is preferably pink noise or a noise component similar to pink noise. Since pink noise does not occur naturally, high frequency components of thermal noise (white noise) used as a signal source of the noise signal are gradually reduced to generate the pink noise. A comparison of the case where white noise that merely cuts the audible frequency band was added to the original audio signal and the case where white noise was processed into pink noise and added to the original audio signal by trial listening showed that the former was In this case, the sound became thinner and sharper, and the latter, that is, the case where pink noise was added, resulted in a reproduced sound closer to a natural sound. This is because while white noise has a flat frequency characteristic, pink noise has a gentle downward-sloping frequency characteristic, which is considered to be closer to the characteristic of the original acoustic signal.
[0033]
It is preferable that the noise component added to the original acoustic signal contains as little as possible an audio frequency band, that is, a component of 20 kHz or less. Therefore, in a preferred embodiment of the present invention, a high-order filter is used as a filter that passes a signal exceeding the upper limit of the audible frequency band, so that noise components of 20 kHz or less are positively cut. As a result of the audition, it was found that the higher the order of the filter to be used, the lower the noise in the reproduced audio signal, and a clearer reproduced sound without smear was obtained.
[0034]
In the present invention, it is preferable that noise sources are provided independently for each channel of the reproduction circuit of the audio device.
[0035]
The sound reproducing device usually has a two-channel configuration of a stereo system, and further may have a multi-channel configuration of a surround system or the like. In any case, obtaining and distributing random noise to be added to a plurality of channels from a single signal source can be recommended from the viewpoint of simplifying a circuit configuration. However, when compared with a device in which a noise source was separately installed and listened, it was found that the device which was independently separated can obtain a natural acoustically widespread and realistic feeling.
[0036]
This is because, when considering natural sounds, in a frequency band exceeding 20 kHz, the phase is disturbed by acoustic reflection, and it becomes easy to lose the relevance of the phase of the left and right channels. This is because it is considered that the independent and separate installation of the noise sources can obtain a reproduced sound closer to a natural sound.
[0037]
Further, in another preferred embodiment of the present invention, a signal (random noise component, or a harmonic component of the original audio signal and a random noise signal) to be added to the original audio signal is added via a capacitor. .
[0038]
This is because adding random noise or harmonic components and random noise via a capacitor causes the additional signal to be separated from the original audio signal circuit in a DC manner, so the output of the additional signal is the buffer amplifier of the original audio signal circuit. This is because it is possible to reduce the load of, and to reduce the sense of auditory suppression in the middle range and the low range. The capacitor to be inserted is not particularly specified, but may be any as long as it has an impedance in a range that does not substantially affect the passage of at least a signal in a band exceeding 20 kHz, which is an added signal.
[0039]
The following effects are obtained by the configuration of the present invention described above.
First, in the first invention of the present application, a high-frequency range exceeding an audible frequency band, which is a region where reproduction output is extremely attenuated in an analog audio device, and a reproduction frequency range exceeding an upper limit of a specific frequency regulated by a sampling frequency in a digital audio device. The unnaturalness of the reproduced sound quality due to the lack of the frequency characteristic generated in the above is removed by adding noise components obtained from different signal sources having a spectrum similar to the acoustic signal to the original acoustic signal, and the spectrum of the reproduced signal is removed. By approximating the spectrum of the original acoustic signal, naturalness in audibility is obtained.
[0040]
FIG. 2 illustrates the principle of the present invention, and shows the spectrum of an acoustic signal reproduced by an acoustic signal reproducing method to which the present invention is applied. An acoustic signal such as an original musical tone has a frequency component covering both the regions A and B in the figure, and the shaded area in the figure fluctuates depending on the magnitude of the sound. However, in reproduction by a digital audio device such as a CD, the high-frequency side is cut off at 20 kHz, and only the region A in FIG. Also, in analog audio equipment, the reproduction output is attenuated in a region exceeding the upper limit of the audible frequency of 20 kHz.
[0041]
Therefore, in the present invention, for example, an appropriate filtering is performed on the output of a noise generator or the like, and an audio signal in a frequency band corresponding to the region B is created separately from the original audio signal. By adding the range of the shaded area of the region, the reproduction state similar to the spectrum band of the original audio signal is reproduced to improve the sound quality.
[0042]
Further, according to the second invention of the present application, the reproduction output of the analog audio equipment in a treble range exceeding the audible frequency band is remarkably attenuated, and the reproduction output of the digital audio equipment exceeds the upper limit of a specific frequency regulated by the sampling frequency. The unnaturalness of the reproduced sound quality caused by the lack of the characteristic is added to both the harmonic component included in the original reproduced sound signal and the noise component in a band exceeding the audible frequency obtained from a signal source different from the sound signal. In this way, a reproduced sound signal that is closer to the spectrum of the original sound signal is obtained. With this configuration, it is possible to further approximate the spectrum of the original sound signal and obtain a more natural reproduction sound quality than in the case where only the harmonic component is added or the case where only the noise component is added. Can be.
[0043]
Further, in the present invention, the level of the signal added to the original audio signal is changed in accordance with the output fluctuation in the range of, for example, 6 kHz to 20 kHz in the original audio signal dynamically, so that the level is closer to the original audio signal. Try to reproduce the sound.
[0044]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 3 is a block diagram showing the configuration of the first embodiment of the sound reproducing device according to the present invention.
FIG. 3 shows a case where the device of the first invention of the present application is applied to an output circuit portion of an analog sound reproduction device or an output circuit portion after D / A conversion of a digital sound reproduction device. The audio signal enters the buffer amplifier 1 from the input terminal 9, is branched, and one is directly connected to the output terminal 10, and the other is input to the high-pass or band-pass filter 2. After passing a signal in a specific band of the acoustic signal through the filter 2, the signal is amplified to an appropriate level by the amplifier 3 and detected by the detection circuit 4. This signal is input to the multiplication circuit 8 via the time constant circuit 5 and used as a level control signal for adjusting the level of a noise component added to the original audio signal.
[0045]
On the other hand, a noise component added to the original audio signal is obtained from the output of the random noise generator 6. The output of the noise generator 6 is passed through a high-pass or band-pass filter circuit 7 through a frequency band of 20 kHz or more, and is input to a multiplication circuit 8 to be multiplied by the level control signal and added to the original sound signal. To do. Therefore, the output of the random noise generator 6 is added to the original sound signal as a signal proportional to the output level of the high frequency sound of the original sound signal, and is taken out from the output terminal 10. There is no particular restriction on the high frequency side of this additional signal, but even if noise components are included in a meaninglessly high frequency band, different troubles may occur, so that the band up to approximately 100 kHz is covered. It should just be. The acoustic signal to which the noise component has been added in this manner is supplied from an output terminal 10 to an electroacoustic transducer such as a speaker or an earphone through an appropriate amplifier or the like, and the music or the like is reproduced.
[0046]
FIG. 4 shows a specific circuit example of the block diagram shown in FIG. 3. Elements that function similarly to the blocks shown in FIG. 3 are given the same reference numerals, and descriptions thereof are omitted. It shall be.
This circuit is installed after the D / A conversion circuit in digital audio equipment, and is installed, for example, before the power amplification stage of an amplifier in analog equipment.
In FIG. 4, an original sound signal is input from an input terminal 9 and is amplified through a buffer amplifier 1. Part of the output of the buffer amplifier 1 is connected to the output terminal 10 and part of the output is guided to the high-pass filter circuit 2. The cutoff frequency of the high-pass filter 2 is approximately 6 kHz.
The sound signal in the band of 6 kHz to 20 kHz selected by the filter circuit 2 is detected by the detector 4 and then sent to the multiplication circuit 8 via the time constant circuit 5 to be added to the original sound signal. Used as a noise component level control signal.
[0047]
At this time, by selecting the time constant of the time constant circuit 5 to a predetermined value for the capacitor 12 and the resistor 13 constituting the circuit, the added noise component is adapted to the audibility and the effect of improving sound quality is enhanced. be able to. In actual trial listening, it was found that in the case of C = 0.047 μF and R = 47 kΩ, excellent reproduction sound quality was obtained near a time constant of 2.2 mS. Thereafter, the preferred range was examined by listening while changing C and R, and as a result, the preferred range was 2.2 mS ± 40%. In this embodiment, a desired time constant is obtained by one CR element. However, the time constant circuit is not limited to this form. Alternatively, a configuration including more CR elements may be employed. Further, it may be constituted by a delay circuit.
[0048]
The noise generator 6 amplifies the thermal noise of the internal resistors and feedback resistors of the operational amplifiers (TL072) 6-1 and 6-2 that randomly generate noise, and uses the amplified noise as a signal source of noise added to the original acoustic signal. . This noise output was passed through the high-pass filter circuit 7 and passed only in a frequency band of 20 kHz or more. Since the thermal noise of the semiconductor tends to decrease due to the effect of the gain characteristic of the semiconductor when the frequency becomes higher, a band-pass filter for cutting the high frequency side is not used in this embodiment. As the thermal noise source, not only a resistance element but also a base resistance inside a transistor or a channel resistance inside a FET may be used.
[0049]
This noise output is input to the multiplication circuit 8 and is multiplied by the above-mentioned level control signal. For this reason, it is added to the original sound signal while changing according to the change in the level of the original sound signal.
[0050]
The switch 11 is provided in the signal mixing path so that the addition of the noise output can be arbitrarily switched.
[0051]
Using the circuit as described above, the music was actually sampled and the effect was confirmed. When the switch 11 is opened and closed to compare the sound signal when the random noise component is added and the sound signal when the random noise component is not added, and the difference in the reproduced sound quality is examined. It was confirmed by hearing that the reproduced sound quality was very natural.
[0052]
FIG. 5 is a block diagram showing the configuration of the second embodiment of the present invention. The same components as those in the first embodiment shown in FIGS. 3 and 4 are denoted by the same reference numerals, and description thereof will be omitted.
[0053]
In the second embodiment, a capacitor 14 is provided in a circuit for adding a random noise component sent from the multiplication circuit 8 to an original audio signal. By attaching the capacitor 14 in this manner, the load applied to the buffer amplifier 1 can be reduced. By listening to the results, it was possible to obtain sound quality with reduced feeling of suppression in the middle and low ranges.
[0054]
FIG. 6 is a circuit diagram showing a configuration of the third embodiment of the audio signal reproducing device of the present invention.
In the example shown in FIG. 6, the operational amplifiers constituting the random noise generating circuit are configured in three stages (6-3, 6-4, 6-5), and the CR elements 6a, 6b, 6c are added to the feedback resistors of the operational amplifiers. Are arranged in parallel, so that generated noise is processed into so-called pink noise. As described above, by adding the pink noise component to the original sound signal, it is possible to obtain a more mellow reproduced sound signal close to a natural sound.
[0055]
In the example shown in FIG. 6, the filter circuit 7 is configured to have an eighth order so as to positively cut noise components of 20 kHz or less. As described above, when the order of the filter is increased, a noise component is hardly added to the audible band, so that clearer sound quality with less blur can be obtained. Note that the order of the filter is not limited to this embodiment, and can be appropriately changed according to the type of acoustic equipment to be used. According to the trial listening, it was confirmed that the higher the order of the filter, the higher the sound quality of the reproduced sound.
[0056]
FIG. 7 is a block diagram showing the configuration of the fourth embodiment of the present invention.
FIG. 7 shows a case where the device of the second invention of the present application is used for an output circuit portion in an analog sound reproducing device and for an output circuit portion after D / A conversion in a digital sound reproducing device. The acoustic signal is branched from the input terminal 30 via the buffer amplifier 21, one of which is sent to the output terminal 31 as it is, and the other is input to the high-pass or band-pass filter 22. The filter 22 allows only a specific band of the acoustic signal to pass therethrough, and inputs this signal to the harmonic generation circuit 23 to generate a harmonic. The output of the harmonic generation circuit 23 is input to a multiplication circuit 27 through an appropriate time constant circuit 24, and is used as a level control signal for controlling the level of a noise component added to the original audio signal.
[0057]
On the other hand, the noise component added to the original audio signal is obtained from the output of the noise generator 25. The noise generated by the noise generator 25 is input to a multiplication circuit 27 via a high-pass or band-pass filter circuit 26, and the noise level is changed according to the fluctuation of the output level of the level control signal. The output of the multiplication circuit 27 and the output of the above-described harmonic generation circuit 23 are input to a mixing circuit 28, and further passed through a high-pass filter 29 to remove the influence of noise on the audible band, and An additional signal to be added to the audio signal is obtained, and the additional signal is added to the original audio signal line and taken out from the output terminal 31 via the capacitor 33.
[0058]
Here, the frequency band of the signal to be added to the original acoustic signal is not particularly limited, but it is sufficient to cover a band up to about 100 kHz as in the first embodiment.
[0059]
FIG. 8 shows a specific circuit example of the block diagram shown in FIG. 7, and the same reference numerals are given to components that function similarly to the components shown in FIG.
As shown in FIG. 8, this circuit is installed after the D / A conversion circuit in digital audio equipment, and is installed before the power amplification stage of an amplifier in analog audio equipment. Similarly to the first embodiment, a random noise signal source is provided for each of the left and right channel circuits in order to form a stereo circuit, and a random noise signal source for each channel when used in a multi-channel device. Just fine.
[0060]
Now, the original sound signal is input from the input terminal 30 and is amplified via the buffer amplifier 21. One end of the output of the buffer amplifier 21 is connected to the output terminal 30, and part of the output is guided to the high-pass filter circuit 22. The cut-off frequency of this high-pass filter is about 6 kHz.
[0061]
The acoustic signal in the band of 6 kHz to 20 kHz selected by the filter circuit 22 is input to the harmonic generation circuit 23, where a harmonic component is generated. The harmonic circuit 23 forms a non-linear circuit by double-wave rectification, and the non-linear circuit intentionally generates distortion to obtain a harmonic component.
[0062]
A part of the output of the harmonic generation circuit 23 is sent to a multiplication circuit 27 via a time constant circuit 24 having a predetermined time constant, and is used as a level control signal of a noise component added to the original acoustic signal. As described in the first embodiment, it has been found that the selection of the time constant of the time constant circuit 24 has a high correlation with the naturalness of the audibility, and the preferable time constant suitable for the audibility is 2.2 mS ±. It was confirmed by listening that the range was 40%.
[0063]
The other output of the harmonic generation circuit 23 is guided to a mixing circuit 28 described below, mixed with the output of the multiplication circuit 27, and then added to the original sound signal.
[0064]
The noise generator 25 amplifies the thermal noise generated from the resistance elements in the operational amplifiers 25-1, 25-2, 25-3, and 25-4 to obtain a noise component added to the original acoustic signal. . The capacitors 26a to 26d connecting the operational amplifiers constituting the noise generator 25 constitute a filter circuit 26. Of the noise components generated here, components of 20 kHz or less are cut to some extent. Since the thermal noise of the semiconductor tends to decrease due to the effect of the gain characteristic of the semiconductor when the frequency becomes higher, a band-pass filter for cutting the high frequency side is not used in this embodiment.
[0065]
The output of the noise generation circuit 25 is input to the multiplication circuit 27 and is multiplied by the level control signal. As a result, the output level of the noise component fluctuates according to the output level of the original audio signal. The noise component is input to the mixing circuit 28 and mixed with a harmonic component dependent on the other original sound signal generated by the harmonic circuit 23, thereby preventing interference with a signal in the audible band. Is added to the original sound signal through a high-pass filter 29 that passes a signal of 20 kHz or more.
[0066]
An 8th-order high-order filter is used for the high-pass filter circuit 29, and is configured to positively cut a component of 20 kHz or less among harmonic components and noise components added to the original acoustic signal. The noise component is also cut off by the filter circuit 26 (4th order), and the 11th order filter including the capacitor 33 is used to remove components below 20 kHz. Therefore, since a noise component hardly remains in the audible band, clear sound quality can be obtained.
[0067]
In the fourth embodiment, the changeover switch 32 is provided so that the addition amount of the harmonic component and the random noise component to the original audio signal can be switched in a plurality of stages, and the type of the original audio signal and the viewer's The amount of addition is adjusted according to preference or the like so that an arbitrary sound quality can be obtained. Note that the changeover switch 32 is not necessarily provided, and may be configured so that a predetermined fixed amount is fixedly added.
[0068]
By actually listening to music using the circuit as described above, the effect of the second invention of the present application was confirmed. Similar to the first invention, when the case where the random noise component and the harmonic component are added and the case where the random noise component and the harmonic component are not added are compared, it can be confirmed that when added, the naturalness in the high frequency range increases, and the acoustic signal is It was confirmed that the reproduction could be suitably performed.
[0069]
FIG. 9 is a circuit diagram showing a configuration of a fifth embodiment of the sound reproducing device of the present invention. In this example, a part of the output of the harmonic circuit 23 is input to a stage preceding the input stage of the noise component of the multiplying circuit 27, and the level of the harmonic component added to the original acoustic signal is also adjusted by the time constant circuit. It is configured to control using a level control signal passed through 24. With such a configuration, it is possible to obtain a sound reproduction signal that is closer to a natural sound signal.
As in the second embodiment, the additional signal is added to the original audio signal via the capacitor 33.
[0070]
FIG. 10 is a circuit diagram showing a configuration of the sixth embodiment of the sound reproducing device of the present invention. In this example, a capacitor and resistance elements 25a to 25d are provided in parallel with the feedback resistance of each operational amplifier constituting the noise generator 25, and the level of the high frequency noise component is gradually lowered to generate pink noise. It is added to the original sound signal. As described in the description of the third embodiment, a more natural reproduced sound signal can be obtained by adding pink noise.
[0071]
FIG. 11 is a block diagram showing the configuration of the seventh embodiment of the present invention. In the present embodiment, a nonlinear circuit for generating harmonics is configured in two stages. That is, after a part of the original sound signal is input to the first absolute value circuit 23 ′ through the filter circuit 22, the output is sent to the mixing circuit 41, where the output of the absolute value circuit 23 ′ is referred to as the original sound signal. Mix. The output of the mixing circuit 41 is further input to a second absolute value circuit 43 via a filter circuit 42 so as to generate harmonics of the original acoustic signal. Other configurations are the same as those of the embodiment shown in FIG. 7, and the description thereof is omitted here. In this way, it has been confirmed by listening that a more natural reproduced sound can be obtained by configuring the nonlinear circuit for generating harmonics in two stages. In this embodiment, the nonlinear circuit for generating harmonics has a two-stage configuration, but may have a configuration of three or more stages. In addition, not only a filter circuit but also a transfer device and a delay circuit may be provided between each stage of the nonlinear circuit. Further, in another combination such as mixing output signals of the nonlinear circuits of each stage. You can also. Further, in the present embodiment, the absolute value circuit is used as the nonlinear circuit for generating harmonics. However, other nonlinear circuits such as a clip circuit, a square circuit, a square root circuit, and a logarithmic circuit can be used.
[0072]
In the above-described first to seventh embodiments, only the configuration of the audio signal reproduction circuit for one channel is shown. However, generally, a stereo circuit is provided with two channels on the left and right, and furthermore, a surround system or the like. May form a multi-channel. In such a case, as described above, it is preferable to separately install noise sources by the number of channels included in the audio device to be used, from the viewpoints of audible sound spread and presence.
[0073]
In the above-described embodiment, the addition of the noise component is performed by the analog circuit. However, the present invention is not limited to these embodiments. After performing on the area, the output can be taken out by performing D / A conversion. However, in this case, it is necessary to keep in mind that a circuit for signal processing is complicated and quantization noise is conspicuous.
[0074]
Rather, according to the present invention, it can be said that the noise generated in an analog manner also acts as a dither and has the effect of making non-linearity in quantization and regular quantization noise inconspicuous.
[0075]
Further, in the above-described embodiment, a time constant circuit is used in order to cause a time delay in the level control signal of the random noise component. However, the present invention is not limited to this embodiment. Other means, such as a delay circuit, may be used as long as they cause the time delay described above.
[0076]
In addition, a signal to be added to the original sound signal is created by the method and apparatus according to the present invention, amplified by a dedicated amplifier, and then subjected to sound conversion separately from the original sound signal by using a dedicated speaker to convert the sound. You may make it synthesize | combine in space. In this case, the level of the sound signal is detected from the speaker terminal that converts the original sound signal into sound or from the sound output from the speaker, and the level of the signal to be added is adjusted according to this level. When the signals are combined, the same effect as when the signals are combined before the sound conversion can be obtained.
[0077]
Further, the apparatus of the present invention can be suitably used not only for record players and CD players, but also for reproducing sound signals in electronic musical instruments. Also, after reproducing the conventional master tape in which the treble is cut off and adding the signal produced by the method of the present invention to the reproduced signal, the reproduced signal is recorded on a better recording tape or CD. Thus, it is possible to produce music software having a new value with improved sound quality.
[0078]
【The invention's effect】
As described above, according to the present invention, in the conventional analog audio equipment, the reproduction frequency characteristic of the high frequency range exceeding the audible frequency band is reduced, and in the digital audio equipment, the frequency band that can be reproduced is determined by the sampling frequency and the like. As a result, it is possible to overcome the unnaturalness in audibility caused by the reduction or omission of a signal in a band higher than this frequency band. According to the present invention, a sudden decrease in the reproduction frequency characteristic of the high-frequency range or a missing acoustic spectrum component is compensated by adding a noise component obtained from a random noise signal source to the original acoustic signal under predetermined conditions. Thus, in the sound reproducing apparatus using the present invention, very natural reproduced sound quality can be obtained.
[0079]
In addition, since the noise component added to the original audio signal is of an audible frequency or higher, there is no need to perform complicated circuits or signal processing. In addition, a missing acoustic spectrum can be obtained, and the improvement effect can be said to be extremely large.
[0080]
Further, according to the second aspect of the present invention, both the harmonic component dependent on the original sound signal exceeding the upper limit of the high frequency sound in the audible band or the reproduction band, and the random noise component obtained from the signal source independent of the original sound signal. By taking the means of adding, it is possible to further eliminate unnaturalness in audibility and to improve reproduction sound quality.
[Brief description of the drawings]
FIG. 1 is a graph showing a spectrum of a frequency component of an audio signal. FIG. 1 (a) shows a spectrum of an original audio signal, and FIG. 1 (b) shows a spectrum of a frequency component when a digital audio device reproduces an audio signal. It was done.
FIG. 2 is a graph for explaining the principle of the audio signal reproducing method according to the present invention, and shows a state of a frequency spectrum of the audio signal.
FIG. 3 is a block diagram showing a configuration of a first embodiment of an audio signal reproducing device according to the present invention.
FIG. 4 is a circuit diagram showing a more specific configuration of the first embodiment shown in FIG. 3;
FIG. 5 is a diagram showing a configuration of a second embodiment of the audio signal reproducing device according to the present invention.
FIG. 6 is a diagram showing a circuit configuration of a third embodiment of the audio signal reproducing device according to the present invention.
FIG. 7 is a block diagram showing a configuration of a fourth embodiment of the audio signal reproducing device according to the present invention.
FIG. 8 is a circuit diagram showing a more specific configuration of the fourth embodiment shown in FIG. 7;
FIG. 9 is a diagram showing a circuit configuration of a fifth embodiment of the sound reproducing device of the present invention.
FIG. 10 is a diagram showing a circuit configuration of a sixth embodiment of the audio signal reproducing device according to the present invention.
FIG. 11 is a diagram showing a circuit configuration of a seventh embodiment of the audio signal reproducing device according to the present invention.
[Explanation of symbols]
1,21 buffer amplifier
2,22 filter circuit
3 Amplifier
4, 24 detection circuit
5 Time constant circuit
6, 25 Noise generator
7,26 Filter circuit
8,27 multiplier
9 Input terminal
10 Output terminal
11, 32 switch
14,33 Capacitor
23 Harmonic generation circuit
28 Addition circuit
29 Filter circuit
23 'first absolute value circuit
41 Mixing circuit
42 Filter Circuit
43 Second Absolute Value Circuit

Claims (14)

In the reproduction method of the acoustic signal, in which the reproduction of the acoustic signal is performed by adding an output having a spectrum in a frequency range exceeding the upper limit of the reproduction frequency in the reproduction frequency or the upper limit of the audible frequency to the original audio signal,
The signal added to the original sound signal is a component obtained by selecting a predetermined frequency band from a noise component similar to a random or random signal obtained from a signal source different from the original sound signal, and this component is referred to as an original sound signal. Extracting a level control signal from the signal, giving the level control signal a time delay of 1.32 to 3.08 mS, and controlling the level of the component in conjunction with the level control signal having a time delay; A sound signal reproducing method comprising adding the level-controlled component to an original sound signal . In the reproduction method of the acoustic signal, in which the reproduction of the acoustic signal is performed by adding an output having a spectrum in a frequency range exceeding the upper limit of the reproduction frequency in the reproduction frequency or the upper limit of the audible frequency to the original audio signal,
The signal to be added to the original sound signal is a component obtained by selecting a predetermined frequency band from a noise component similar to a random or random signal obtained from a signal source separate from the original sound signal , and a harmonic obtained from the original sound signal. And a level control signal extracted from the original sound signal, and at least the noise component, or both the noise component and the harmonic component, are linked to the level control signal to obtain a level. A sound signal reproducing method comprising controlling and adding each of the level-controlled components to the original sound signal. 3. The audio signal reproducing method according to claim 2, wherein the level-controlled noise component or the noise component and the harmonic component added to the original audio signal have a time delay of 1.32 to 3.08 mS. An audio signal reproducing method , wherein a level-controlled component having a time delay is added to the original audio signal. 4. The acoustic signal reproducing method according to claim 2, wherein the harmonic component added to the original acoustic signal is a harmonic component generated by passing the original acoustic signal through an absolute value circuit. Signal playback method. The audio signal reproducing method according to any one of claims 1 to 4,
A sound signal reproducing method, wherein the signal processing of the sound signal reproduction is performed in a digital domain. In the audio signal reproducing method according to any one of claims 1 to 5, the acoustic signal reproduction method, which comprises carrying out by the addition of signals to the original audio signal, to synthesize a sound space. 7. The sound signal reproducing method according to claim 6, wherein the level control is performed by detecting a level of the sound signal from a speaker terminal for converting an original sound signal into sound or from a sound output from the speaker. Signal playback method. In a sound signal reproduction device that adds an output having a spectrum of a frequency band exceeding a high frequency range upper limit of a reproduction frequency band or a high frequency range upper limit of an audible frequency band to an original sound signal in reproduction of an audio signal,
Random noise generating means for generating random noise to be added to the original audio signal from a signal source separate from the original audio signal,
From the random noise generated by the random noise generating means , audible frequency band, or frequency selection means for selecting random noise in a frequency band exceeding the upper limit of the high frequency range of the reproduction frequency band in the reproduction of the acoustic signal,
Level control means for extracting a level control signal from an original sound signal and controlling an output level of random noise added to the original sound signal in accordance with the level control signal ;
A time delay generating means for generating a time delay of 1.32 to 3.08 ms in the level control signal;
Signal addition for adding random noise, which is a random noise whose predetermined frequency is selected by the frequency selecting means and whose output level is controlled with the time delay by the level control means and the time delay generating means , to an original sound signal Means,
A sound signal reproducing device comprising: In a sound signal reproduction device that adds an output having a spectrum of a frequency band exceeding a high frequency range upper limit of a reproduction frequency band or a high frequency range upper limit of an audible frequency band to an original sound signal in reproduction of an audio signal,
Random noise generating means for generating random noise to be added to the original audio signal from a signal source separate from the original audio signal,
From the noise component, an audible frequency band, or a frequency selection unit that selects a frequency band that exceeds a treble upper limit of a reproduction frequency band in reproducing an audio signal,
Harmonic component generating means for generating a harmonic component obtained from the original acoustic signal,
Level control means for extracting a level control signal from an original sound signal, adding random noise to the original sound signal, or an output level of random noise and harmonic components, in accordance with the level control signal ;
A noise component whose output level is controlled by the level control means and a predetermined frequency is selected by the frequency selection means; or a harmonic component or an output level controlled by the level control means and a predetermined frequency is controlled by the frequency selection means. And a signal adding unit that adds the harmonic component whose output level is controlled by the level control unit to an original sound signal.
A sound signal reproducing device comprising: 10. The audio signal reproducing apparatus according to claim 9, further comprising a time delay generating means for generating a time delay of 1.32 to 3.08 ms in the level control signal. The acoustic signal reproducing device according to claim 9 or 10,
An acoustic signal reproducing apparatus, wherein the apparatus comprises a harmonic component generating means for generating a harmonic component of the original acoustic signal, and the harmonic component generating means comprises an absolute value circuit. The audio signal reproducing apparatus according to claim 8, wherein the signal processing of the audio signal reproduction is performed in a digital domain. In the audio signal reproducing apparatus according to any one of claims 8 to 12, wherein the additional signal to the original audio signal, sound audio signal reproducing apparatus which is characterized in that by combining in a space. 14. The audio signal reproducing apparatus according to claim 13, further comprising: means for detecting a level of the audio signal from a speaker terminal for converting the original audio signal into sound or from a sound output from the speaker, and controlling the level of the detected level. A sound signal reproducing device characterized by performing:

Images