JPH11113097A - Audio system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To extend the range of sound image without changing an original sound quality by calculating components included in right/left channels in common and generating the waveform of a front channel and a surround channel based on the calculated common components. SOLUTION: A common component calculating part 13 calculates spectrum component included in the right/left channels in common. An output spectrum calculation part 15 separates right and left spectrum calculated by a spectrum analytic part 12 by a spectrum separation coefficient updating part 14 into four channels in total being right and left front channels and surrounding channels. Then, an output waveform calculating part 16 calculates an output waveform. Waveform D/A converted by a D/A conversion part 17 is power- amplified by an amplifier 18 and electroacoustically converted by speakers 19 to 22 within a listening room. This sound is heard in the state of a wide sound image of front 180 deg. when listening this sound at the spot of a listener 23.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention provides a two-channel audio signal that is subjected to predetermined processing and reproduced from a plurality of speakers including a surround speaker.
It is intended to obtain a good spread feeling.

[0002]

2. Description of the Related Art Generally, a sound image of a stereophonic sound reproduction system is located between speakers, and the sense of spread varies depending on the position of the speakers. For example, when the same waveform is output from the left and right speakers, the sound image is localized at the center, and it feels as if sound is coming out from the center. Conversely, if the waveforms of the left and right speakers are different, the range of the sound image expands between the two speakers. 2 in FIG.
An example of a typical reproduction system using a channel speaker will be described. However, sound image localization is limited to an angle α in front of a listener. This is because the sound image is empirically localized by comparing the phase difference and the level difference of the waveforms entering both ears from the left and right speakers with the phase difference and the level difference obtained from the actual sound source.

In order to increase the sound image localization angle, the distance between speakers may be increased, but there is a physical limit. Also, as shown in FIG. 3, when the speaker is moved to the side, the spread between the right and left sides increases, but a sense of incongruity appears such that the sound that should be localized forward is localized above the sound image. Further, if a speaker is added to the side and the same signal as that of the front speaker is also output from the side, localization of the sound image becomes difficult.

As a technique for freely controlling the sound image localization position beyond the restriction of the positional relationship of the speakers, for example, a method of canceling crosstalk between both ears is disclosed in Japanese Patent Application Laid-Open No. 8-146974. I have. In addition, a method of inputting a difference signal between left and right signals to a surround channel and amplifying a reverberation component to expand a sound field may be simple in processing, and is generally widely known.

[0005]

However, the technique of controlling the sound image localization position to localize the sound image outside the speaker has a great effect on the arrangement of the speakers and the shape of the room. There is a problem that changes.

The method using the difference signal between the left and right signals has a problem that the reverberation component is amplified so that the timbre is different from the original, and the sound image is blurred. For example, a difference signal of two-channel music in which a vocal is located in the center, a guitar is located on the left, and a piano is located on the right is a signal obtained by mixing the guitar and the piano with the inverted phase. Since this signal is output from the surround speakers, the sound of the guitar and the piano is emphasized, and the sound image of the guitar and the piano is blurred.

The present invention solves the above-mentioned problems in the prior art, and is less affected by the arrangement of speakers, the shape of the room, the listening position, and the like, and can expand the sound image without changing the original sound quality. It is intended to provide such an audio device.

[0008]

According to the first aspect of the present invention, in order to solve the above-mentioned problems, a two-channel stereo signal is input from various audio sources and subjected to predetermined processing. In an audio device for reproducing from a plurality of speakers including a surround speaker,
Input spectrum analysis means for analyzing the spectra of the left and right channel signals, common component calculation means for calculating components commonly included in the left and right channels from the spectrum analyzed by the input spectrum analysis means, and the common component calculation means Output spectrum calculating means for separating an input spectrum into a front channel and a surround channel based on the common component calculated in step (a) to generate a spectrum of the output channel; and a four-channel waveform from the output spectrum calculated by the output spectrum calculating means. And an output waveform generating means for generating

According to a second aspect of the present invention, in order to solve the above-mentioned problem, the output spectrum calculating means calculates so that the common component is allocated more to the spectrum of the front channel. And

According to a third aspect of the present invention, in order to solve the above-mentioned problems, the audio apparatus according to the third aspect of the present invention has one front speaker.
In the case of the book, an output waveform generating means for generating a monaural front signal by adding the left and right front channel signals is provided.

As described above, according to the present invention, when a two-channel stereo signal is input, the input spectrum analyzing means analyzes the spectrum of each channel signal, and the common component calculating means analyzes the common component included in both channels. Is calculated. Next, the output spectrum calculating means separates the input spectrum into a front channel and a surround channel, and calculates the spectrum of the output channel.
The output spectrum is converted into a waveform by an output waveform generating means and reproduced from a speaker.

As a result, the arrangement of the speakers, the shape of the room,
It is hardly affected by the listening position and the like, and furthermore, it is possible to widen the sound image without changing the original sound quality.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described conceptually with reference to FIG. 4 (a).
The sound image is separated into α + 2β as shown in FIG. 4 (b) by separating the common component from the front speakers and outputting the R and L components from the respective surround speakers. It is something to spread.

FIG. 1 is a block diagram showing one embodiment of the audio apparatus of the present invention. In FIG. 1, 10 is C
It is an audio source such as a D playback device, an MD playback device, a tape playback device, and a radio tuner. 11 is A / D
A conversion unit that converts an analog signal output from the audio source into a digital signal,
The spectrum analysis is performed for each input channel, and the spectrum information is sent to 13 common component calculation units and 15 output spectrum calculation units. Next, after the common component calculation unit 13 calculates the spectrum components included in the left and right channels, the spectrum separation coefficient calculation unit 14 calculates the coefficients to be separated into the front channel and the surround channel.

The 15 output spectrum calculators calculate the left and right spectra calculated by the spectrum analyzer 12 based on the spectrum separation coefficients calculated by the spectrum separation coefficient updater 14 based on the left and right front channels and surround channels. The output waveform is calculated by 16 output waveform calculation units. Next, the D / A conversion unit 17 provides D / A
The A-converted waveform is power-amplified by 18 amplifiers,
Speakers 19, 20, 2 in 24 listening rooms
Electroacoustic conversion is performed by 1, 22. Listener 2
When listening at the point 3, the sound is heard as a wide sound image 180 degrees forward. When the audio source is a digital output, the A / D converter 11 is passed through. The entire processing is controlled by the CPU 25.

Next, the flow of processing for creating a four-channel waveform from a two-channel waveform will be described with reference to the flowchart of FIG. In step S1, L points (for example, L = 512) for each channel from the audio source
Input the sampling sequence. Note that the left channel signal is xl [n], the right channel signal is xr [n], and n is a sampling number.

In step S2, the spectrum analyzer 12 performs a short-time Fourier transform on each channel signal as shown in the following equation. Note that w (n) is an arbitrary window sequence having a length L, and a square window, a triangular window, a Hanning window, a Hamming window, a Blackman window, and the like are often used, but are not particularly limited here.

[0018]

(Equation 1)

[0019]

(Equation 2)

In step S3, the common component calculator 1
At 3, the common component C (k) is calculated using the following equation. Note that MIN ｛｝ is a function that gives the minimum value.

C (k) = MIN ｛| XL (k), XR (k) |｝ (0 ≦ k <L / 2) (3) In step S4, next, the spectrum separation coefficient The updating unit 14 updates the separation coefficient k (k) by the following equation. In addition,
α is a constant (for example, 0.2 which is the best by experiment), and the initial value of k (k) is 0.5.

K (k) = c (k) / ｛| XL (k) | + | XR (k) |｝ × α + k (k) × (1−α) (0 ≦ k <L / 2) · (4) In step S5, the output spectrum calculation unit 15
Front spectra XFL (k) and XFR (k) and surround spectra XSL (k) and XSR (k)
Is calculated by the following equation.

XFL (k) = XL (k) × sin (π × k (k)) (0 ≦ k <L / 2) (5) XSL (k) = XL (k) × cos (π × k (k)) (0 ≦ k <L / 2) (6) XFR (k) = XR (k) × sin (π × k (k)) (0 ≦ k <L / 2) (7) XSR (k) = XR (k) × cos (π × k (k)) (0 ≦ k <L / 2) (8) The reason why the trigonometric function is used to separate the front spectrum and the surround spectrum is to keep the tone of the original signal unchanged by keeping the power spectrum constant. Also, as is apparent from the above equation, the separation coefficient k (k)
Has a positive value of 0.5 or less, and the larger k (k), the more common components are included, and the sound is collected in the front direction. Conversely, the sound moves in the surround direction as k (k) decreases.

In step S6, the output waveform calculator 1
At 6, the front left channel xfl (n), the front right channel xfr (n), the surround left channel xsl (n), and the surround right channel xsr (n)
Are obtained by performing an inverse Fourier transform by the following equation using the spectrum calculated in step S5.

[0025]

(Equation 3)

[0026]

(Equation 4)

[0027]

(Equation 5)

[0028]

(Equation 6)

In step S7, the waveform calculated in step S6 is sent to the D / A converter 17, and the processing in steps S1 to S6 is repeated for the next input waveform. By outputting the four-channel waveforms created by the above processing from the speaker, it is possible to reproduce music in which the sound image localization is expanded without deteriorating the original sound quality.

Next, a description will be given of a reproducing method in the case of a total of three loudspeakers, one front speaker and two left and right surround speakers. In this case, the waveforms of the left and right front channels calculated as described above are added on the waveform as shown in the following equation, and this is used as the front channel waveform.

Xf (n) = xfl (n) + xfr (n) (0 ≦ n <L) (13) By doing so, a sound field having a feeling of spreading even with a three-channel speaker can be obtained. Since reproduction is possible, for example, by adding two surround speakers to an existing monaural television, audio reproduction with a sense of reality can be realized, and there is an advantage that it can be realized at low cost.

[0032]

As described above, the audio apparatus of the present invention performs a predetermined process on a two-channel audio signal and reproduces the audio signal from a plurality of speakers including a surround speaker, thereby providing a good sense of spaciousness. You can enjoy some music.

[Brief description of the drawings]

FIG. 1 is a block diagram of an audio device according to the present invention.

FIG. 2 is a flowchart illustrating an operation related to generation of an output waveform of four channels from an input waveform of two channels.

FIG. 3A shows a general method of listening to stereo sound. (B) shows an example in which a speaker is placed on the side to expand sound image localization.

FIGS. 4A and 4B are schematic diagrams illustrating a method of separating an input spectrum into a front channel and a surround channel.

[Explanation of symbols]

 Reference Signs List 10 audio source 11 A / D conversion unit 12 spectrum analysis unit 13 common component calculation unit 14 spectrum separation coefficient calculation unit 15 output spectrum calculation unit 16 output waveform calculation unit 17 D / A conversion unit 18 amplifier 19, 20, 21, 22 speaker

Claims (3)

[Claims] 1. An audio apparatus which receives two-channel stereo signals from various audio sources, performs predetermined processing, and reproduces the signals from a plurality of speakers including a surround speaker. Analysis means, from the spectrum analyzed by the input spectrum analysis means, a common component calculation means for calculating components commonly included in the left and right channels, based on the common component calculated by the common component calculation means,
Output spectrum calculating means for separating an input spectrum into a front channel and a surround channel to generate a spectrum of an output channel; and output waveform generating means for generating a four-channel waveform from the output spectrum calculated by the output spectrum calculating means. An audio device comprising: 2. The audio apparatus according to claim 1, wherein said output spectrum calculating means calculates so as to allocate more of said common component to said spectrum of said front channel. 3. The apparatus according to claim 1, further comprising output waveform generating means for generating a monaural front signal by adding left and right front channel signals when the number of front speakers is one. Audio equipment.