JP3540435B2 - Pseudo stereo conversion method and apparatus - Google Patents

Description

[0001]
[Industrial applications]
The present invention relates to a method and an apparatus for converting a monaural audio signal into pseudo-stereo, and is used for a television broadcast and a radio broadcast receiver, an audio device for reproducing a video tape, an audio tape, a compact disc, and the like. In this specification, an audio signal is a phrase for a video signal, and is used in a concept that includes not only so-called audio but also all sound signals or data such as music and natural sounds.
[0002]
[Prior art]
In the audio-visual world of recent years, the television screen has been enlarged, stereophonic sound has been adopted, and a surround system has been adopted to create richer realism and power. However, audio software, visual software, and various other sound sources are not always supplied in a multi-channel manner, and are often monaural sound sources at present. Therefore, in order to give a sense of presence to a monaural sound source, that is, a sound field for reproducing a monaural audio signal, attention has been paid to pseudo-stereo conversion of a monaural audio signal.
[0003]
Normally, pseudo-stereo conversion is performed by creating two audio signals having a small correlation from a monaural audio signal. Conventional methods for this purpose include a method using phase control and a method using a comb filter.
[0004]
In the former method, an audio signal whose phase is shifted from the original audio signal is created, and the two audio signals whose phases are shifted from each other are reproduced from the left and right speakers. In the latter method, a monaural audio signal is passed through two filters having different passing frequency bands, and the two audio signals separated by the frequency band are reproduced from left and right speakers.
[0005]
[Problems to be solved by the invention]
However, in the former method, the correlation between the left and right audio signals cannot be reduced so much that a sufficient spreading feeling cannot be obtained, and there is an unnaturalness such that the sound comes to the ear due to a sense of antiphase. There's a problem.
[0006]
In the latter method, the frequency characteristics change greatly with respect to the original audio signal.For example, the same musical instrument can be heard left and right separately according to the pitch, resulting in extreme movement of the sound image and natural reproduction sound. There is a problem that can not be.
[0007]
The present invention has been made in view of the above-described problems, and has as its object to provide a pseudo-stereo conversion method and apparatus capable of obtaining a sufficient spread feeling and obtaining a natural reproduced sound. I do.
[0008]
[Means for Solving the Problems]
The method according to the first aspect of the present invention is a method of converting a monaural audio signal into pseudo-stereo, wherein the frequency of the audio signal is changed by 1 / f fluctuation to change the frequency from the original audio signal. In this method, an audio signal is used as an audio signal of another channel.
[0009]
An apparatus according to claim 2, wherein the Fourier transform means for converting the monaural audio signal into a frequency domain, and a frequency shift means for shifting the frequency of the audio signal converted into the frequency domain so as to be changed by 1 / f fluctuation. Means and inverse Fourier transform means for converting the shifted audio signal into the time domain.
[0010]
According to a third aspect of the present invention, there is provided an apparatus for converting a monaural audio signal, which is an analog signal, into pseudo stereo, wherein the monaural audio signal is sampled at a predetermined sampling frequency Fs and converted into digital signal audio data. A / D conversion means for converting the audio data into an audio signal which is an analog signal at a fluctuation frequency Ff which varies with 1 / f fluctuation around the sampling frequency Fs. And a second DA converter for converting an audio signal which is an analog signal by the sampling frequency Fs.
[0011]
[Action]
The principle of the present invention will be described with reference to FIG. 1. A monaural audio signal MS is changed by a frequency shift means A so that the frequency changes by 1 / f fluctuation, and is changed to a first channel audio signal S2R. I do. The original audio signal MS is used as the audio signal S2L of the second channel. In that case, if necessary, the audio signal S2L is delayed by the delay means B by the same amount as the signal delay caused by the processing of the frequency shift means A, so that the audio signals S2R and S2L of the first and second channels are reproduced simultaneously. Synchronize with.
[0012]
By reproducing the audio signals S2R and S2L of the first channel and the second channel using, for example, left and right speakers, the left and right reproduced sounds have a small correlation with each other, so that the listener can obtain a feeling of expansion. The frequency changes due to the 1 / f fluctuation, and a natural spreading feeling is obtained.
[0013]
The change due to the 1 / f fluctuation means that when the energy such as the amplitude and frequency of the sound is large, the control parameter changes so that the frequency of the fluctuation (fluctuation) becomes small.
[0014]
【Example】
FIG. 2 is a block diagram of the pseudo-stereo device 1 according to the first embodiment of the present invention, FIG. 3 is a diagram showing a state where audio data Da is cut out for each section SC, and FIG. 4 is a diagram showing a state of shifting audio data Df. It is.
[0015]
In FIG. 2, the pseudo stereo apparatus 1 includes an audio input terminal Ta, an AD conversion unit 11, a cutout unit 12, an FFT (fast Fourier transform) unit 13, a frequency shift unit 14, an IFFT (inverse fast Fourier transform) unit 15, and a delay unit. 16, D / A converters 17R and 17L, power amplifiers 18R and 18L, and speakers SPR and SPL.
[0016]
The AD converter 11 converts the monaural audio signal S1 that is an analog signal input to the audio input terminal Ta into audio data Da that is a digital signal in accordance with the sampling clock CLK1. The sampling frequency Fs is, for example, 44.1 KHz.
[0017]
As shown in FIG. 3, the cutout unit 12 cuts out the audio data Da by a window function and generates audio data Da for each section SC having a fixed time width. As the window function, for example, a Hanning window is used. The section SC is set so that, for example, one section SC includes 500 pieces of audio data Da. In that case, the time width of one section SC is about 11.3 msec (= 500/44100).
[0018]
The FFT unit 13 performs a fast Fourier transform on the audio data Da of each section SC, and converts the audio data Da in the time domain into the audio data Df in the frequency domain.
The frequency shift unit 14 shifts the audio data Df in the frequency domain so as to change by 1 / f fluctuation. The frequency shift unit 14 includes memories 31a and 31b, a shift processing unit 32, and a 1 / f fluctuation generating unit 33. The audio data Df for one section SC output from the FFT unit 13 is stored in the memory 31a. The audio data Df stored in the memory 31a is shifted by a frequency determined according to 1 / f fluctuation by the shift processing unit 32 for each section SC, and the shifted audio data Dfs is stored in the memory 31b. The audio data Df in the memory 31a is deleted every time the shift processing is performed.
[0019]
The shift of the audio data Df in one section SC in the shift processing unit 32 is performed by considering that the frequency F of the audio data Dfs after the shift is represented on a logarithmic graph, where f is the frequency of the audio data Df before the shift. And is expressed by the following equation (1).
[0020]
log (F) = log (f × p) (1)
Here, p is a constant that determines the shift amount and the shift direction. When the constant p is larger than 1, the shift is performed to the high frequency side, and when the constant p is smaller than 1, the shift is performed to the low frequency side. The constant p is determined by the 1 / f fluctuation generator 33. However, the fluctuation generated by the 1 / f fluctuation generating section 33 may be approximate.
[0021]
As shown in FIG. 4, when the audio data Df and Dfs are represented on a logarithmic graph with the logarithmic scale (logarithmic scale) on the horizontal axis, the shifted audio data Dfs maintains the original audio data Df in its shape. In this state, it is translated in the plus or minus direction by the shift amount Qs. The shift amount Qs on the logarithmic graph is Log (p). By performing the shift on the log scale in this manner, the audio data DaR after being inversely transformed into the time domain by the IFFT unit 15 described later, only the pitch changes with respect to the original audio data Da, and the timbre and the sound pressure are changed. Does not change.
[0022]
The IFFT unit 15 performs inverse Fourier transform on the audio data Dfs read from the memory 31b, and returns the audio data Dfs to the audio data DaR in the time domain.
The delay unit 16 delays the audio data Da output from the AD conversion unit 11 by the same time as the total processing time in each unit from the extraction unit 12 to the IFFT unit 15, and synchronizes with the audio data DaR.
[0023]
The DA converters 17R and 17L perform DA conversion on the audio data DaR and DaL output at the same cycle as the sampling clock CLK1, and convert them into audio signals S2R and S2L, which are analog signals.
[0024]
The power amplifiers 18R and 18L amplify the audio signals S2R and S2L and drive the right and left speakers SPR and SPL, respectively. As a result, an audio signal S2R in which the frequency of the original audio signal S1 is changed by 1 / f fluctuation is reproduced from the right speaker SPR, and the same audio signal S2L as the original audio signal S1 is reproduced from the left speaker SPL. Is played. That is, the original audio signal S2L and the audio signal S2R whose frequency has been changed are reproduced from the two speakers SPR and SPL. Since the original audio signal S2L and the frequency-changed audio signal S2R have a small correlation, the listener can obtain a sufficiently wide stereo feeling, and the right-side audio signal S2R whose frequency has been changed has a timbre and Since the sound pressure does not change, there is no sense of incongruity in the listener, and a natural spreading feeling can be obtained. Also, there is no unnaturalness due to the sense of anti-phase as in the method using phase control, there is no change in the frequency characteristics as in the case of using a comb filter, and the sound image is well localized. Since the constant p changes due to the 1 / f fluctuation, and the frequency shift amount Qs is varied according to the 1 / f fluctuation, an extremely natural spreading feeling can be obtained.
[0025]
In the above-described embodiment, the clipping unit 12, the FFT unit 13, the shift processing unit 32, the IFFT unit 15, the delay unit 16, and the like can be realized using, for example, a DSP (Digital Signal Processor). As the memories 31a and 31b, a memory area built in the DSP may be used, or an externally connected memory may be used.
[0026]
In the above-described embodiment, the audio signal S1 which is an analog signal is input. However, when a digital signal output from a digital device such as a DAT is input, the AD converter 11 is unnecessary. The frequency shift unit 14 can have various configurations. Although the audio data Da is delayed by the delay unit 16 so that the left and right audio signals S2R and S2L are simultaneously reproduced, the audio data Da may be shifted by an appropriate time. Further, the delay unit 16 can be omitted. The time width of the section SC can be set to an appropriate value.
[0027]
Next, a second embodiment of the present invention will be described. FIG. 5 is a block diagram of a pseudo stereo apparatus 1a according to a second embodiment of the present invention.
In FIG. 5, the pseudo stereo apparatus 1a includes an AD converter 21, memories 22R and 22L, DA converters 23R and 23L, power amplifiers 24R and 24L, speakers SPR and SPL, a fluctuation clock generator 41, and 1 / f fluctuation. It comprises a generating unit 42.
[0028]
The AD converter 21 converts the monaural audio signal S1 that is an analog signal input to the audio input terminal Ta into audio data Da that is a digital signal in accordance with the sampling clock CLK1. The sampling frequency Fs is, for example, 44.1 KHz.
[0029]
The audio data Da output from the AD converter 21 is stored in the two memories 22R and 22L. From one memory 22R, audio data Da is read out by a fluctuation clock CLKf that changes by 1 / f fluctuation with the sampling frequency Fs as a center frequency, and is converted into an analog audio signal S2R by a DA converter 23R. You.
[0030]
The fluctuation clock generator 41 generates a fluctuation clock CLKf having a frequency Ff that changes with the 1 / f fluctuation around the sampling frequency Fs based on the fluctuation signal S3 from the 1 / f fluctuation generator 42. The fluctuation frequency Ff changes within a range of, for example, 43.8 to 44.4 KHz. The range of the fluctuation frequency Ff can be set by a setting unit (not shown) to various values that do not cause a sense of incongruity. Thus, the frequency of the original audio signal S1 is changed by the 1 / f fluctuation in the audio signal S2R.
[0031]
The audio data Da is read from the other memory 22L by the clock CLKs having the sampling frequency Fs, and is converted into an analog audio signal S2L by the DA converter 23L.
[0032]
The power amplifiers 24R and 24L amplify the audio signals S2R and S2L and drive the right and left speakers SPR and SPL, respectively. As a result, an audio signal S2R in which the frequency of the original audio signal S1 is changed by 1 / f fluctuation is reproduced from the right speaker SPR, and an audio having the same frequency as the original audio signal S1 is reproduced from the left speaker SPL. The signal S2L is reproduced.
[0033]
Therefore, similarly to the first embodiment, the listener can obtain a sufficiently wide stereo feeling, and the right-side audio signal S2R whose frequency has been shifted has no change in timbre and sound pressure. There is no natural feeling of expansion. Since the frequency changes due to the 1 / f fluctuation, a listener can enjoy a comfortable and extremely natural spreading feeling. Further, the pseudo stereo apparatus 1a of the second embodiment has a simple configuration and is low in cost. Since the time required for signal processing of both the left and right channels is the same, the audio signals S2R and S2L are reproduced simultaneously without providing a delay circuit.
[0034]
In the above-described embodiment, the audio signals S2R and S2L are reproduced by the left and right speakers SPR and SPL. However, the audio signals S2R and S2L may be reproduced in addition to the audio signals on the rear side of the surround system. The sampling frequency Fs, the range of the fluctuation frequency Ff, the number of bits for AD conversion, and the like can be variously changed. In addition, the configuration, processing contents, processing order, processing timing, and the like of the whole or each part of the pseudo stereo apparatuses 1 and 1a can be variously changed in accordance with the gist of the present invention.
[0035]
【The invention's effect】
According to the first to third aspects of the present invention, it is possible to provide a pseudo-stereo conversion method and apparatus capable of obtaining a sufficient spreading feeling and obtaining a natural reproduced sound.
[0036]
According to the third aspect of the invention, the configuration is simple and the cost is low.
[Brief description of the drawings]
FIG. 1 is a diagram for explaining the principle of the present invention.
FIG. 2 is a block diagram of a pseudo stereo apparatus according to a first embodiment of the present invention.
FIG. 3 is a diagram showing how audio data Da is cut out for each section.
FIG. 4 is a diagram showing a state of shifting audio data Df.
FIG. 5 is a block diagram of a pseudo stereo apparatus according to a second embodiment of the present invention.
[Explanation of symbols]
1,1a pseudo stereo device (pseudo stereo device)
12 Cutting part (cutting means)
13 FFT unit (Fourier transform means)
14. Frequency shift unit (frequency shift means)
15 IFFT unit (inverse Fourier transform means)
21 AD converter (AD converter)
23R DA converter (first DA converter)
23L DA converter (second DA converter)
A frequency shift means

Claims (3)

A method of converting a monaural audio signal into pseudo-stereo,
Changing the frequency of the audio signal by 1 / f fluctuation,
Using the original audio signal and the audio signal whose frequency is changed as audio signals of different channels from each other,
A pseudo-stereo conversion method characterized in that: A device for converting a monaural audio signal into pseudo-stereo,
Fourier transform means for transforming the monaural audio signal into a frequency domain,
Frequency shifting means for shifting the frequency of the audio signal converted into the frequency domain so as to change by 1 / f fluctuation;
An inverse Fourier transform means for transforming the shifted audio signal into a time domain;
A pseudo-stereo conversion apparatus characterized by having: A device for converting a monaural audio signal, which is an analog signal, into pseudo-stereo,
AD conversion means for sampling the monaural audio signal at a predetermined sampling frequency Fs and converting the monaural audio signal into audio data of a digital signal;
First DA conversion means for converting the audio data into an audio signal which is an analog signal at a fluctuation frequency Ff which varies with 1 / f fluctuation around the sampling frequency Fs;
A second DA converter that converts the audio data into an audio signal that is an analog signal at the sampling frequency Fs;
A pseudo-stereo conversion apparatus characterized by having:

Images