WO2007102687A1

WO2007102687A1 - Apparatus and method for optimizing 3d sound in multimedia device

Info

Publication number: WO2007102687A1
Application number: PCT/KR2007/001089
Authority: WO
Inventors: Sin-Lyul Lee
Original assignee: Slaudiolab Co., Ltd.
Priority date: 2006-03-07
Filing date: 2007-03-06
Publication date: 2007-09-13
Also published as: KR100703923B1

Abstract

An apparatus for optimizing three-dimensional sound in a multimedia device includes a three-dimensional sound processing unit, and a harmonic component level adjustment unit. The three-dimensional sound processing unit analyzes the frequency spectrum of an audio file, generates harmonic components to restore low frequency harmonic components for generating low frequency sounds and high frequency harmonic components for generating high frequency sounds, and adjusts the levels of the generated low and high frequency harmonic components according to a user's level setting values, mixes the level-adjusted low and high frequency harmonic components with middle frequency components, and outputs the frequency components obtained by the mixing. The harmonic component level adjustment unit alternately generates the middle frequency reference sound at a fixed level and the low and high frequency reference sounds at varying levels, and provides the levels of the corresponding reference sounds to the three-dimensional sound processing unit as level setting values.

Description

APPARATUS AND METHOD FOR OPTIMIZING 3D SOUND IN MULTIMEDIA DEVICE

Technical Field

The present invention relates to an apparatus and method for optimizing three-dimensional sound in a multimedia device which reproduces the three-dimensional sound by restoring low and high frequency sounds, lost at the time of sound reproduction via a small-sized sound converter, using a psychoacoustic effect

Low and high frequency reference sounds are generated using the harmonic components of respective low and high frequency reference noise signals, a user alternately listens to the generated low and high frequency reference sounds along with a middle frequency reference sound, that is, a middle frequency noise signal sound, and then subjectively adjusts the levels thereof, thereby allowing the levels of the low and high frequency sounds, which are restored at the time of sound reproduction via the multimedia device, to be adjusted according to level setting values depending on the user's subjective adjustment.

Background Art

Recently, with the development of information and communication technology and multimedia technology, various types of portable multimedia devices are being used. In the portable multimedia devices, reproduction of sound is mostly performed via small-sized portable sound converters, such as earphones, headphones, clip phones and canal phones, as shown in FIG. 1. Such small-sized sound converters have very different sound reproduction characteristics according to the type thereof, and exhibit large differences in sound characteristics between respective manufacturing companies even for the same type of product

For such multimedia devices, the objective of reproduction of three-dimensional sound is to make the characteristics of the reproduced sound, which is listened to through small-sized sound converters, similar to those intended by a sound engineer in a studio. Sounds that the sound

engineer listens to in the studio are reproduced through full-range loudspeakers that are disposed 30 degrees to the left and right and are capable of reproducing all sounds that humans can hear within an audible frequency range of 20 Hz ~ 20 kHz. Before reaching the sound engineer's ears, sounds emitted from the respective loudspeakers are reflected from the walls of the studio, to which diffusion plates are attached, and then reach the sound engineer's ears togelher with various reflected sounds. In this way, a kind of stage is formed between the two loudspeakers disposed in the front of the sound engineer, and respective musical instruments are disposed at appropriate locations between the loudspeakers, so that an effect in which sounds are realistically reproduced as if they came from a performance stage is achieved.

However, in the case of the use of small-sized sound converters, sounds are reproduced through sound converters inserted into both ears of a listener or through sound converters which press the listener's ears, so that most of the sounds that should be distributed in front of the listener are formed in a line in the listener's head between the two ears or around the two ears. This can irritate the listener. Unlike a general loudspeaker, the earphone or the headphone has a small-sized vibration plate, so that it has a limited capability to reproduce low frequency sounds. Since a digital audio compression process is performed due to the limited storage space of the multimedia device,

considerable portions of high frequency sounds above 16 kHz are lost through the compression process.

Particularly, for earphones, low frequency reproduction capability decreases rapidly in a frequency region below 100 Hz due to the compact size of a portable sound converter. The easiest method of overcoming such a limited low frequency reproduction capability is to increase the levels of the frequencies below 100 Hz using an equalizer. However, merely increasing the frequency levels greatly decreases durability because it causes the sound vibration plate of the small-sized sound converter to move excessively, increases the amount of power consumption because excessive movement of the sound vibration plate occurs, limits the dynamic expression of music because the dynamic range across the entire band is narrowed by the increase in the low frequency band, and increases the probability of the occurrence of clipping noise that is caused by excessive movement of the small-sized converter at low frequencies.

Accordingly, in order to solve such problems, various three-dimensional sound reproduction algorithms, such as a front sound image localization algorithm, a low frequency extension algorithm and a high frequency extension algorithm, which are installed in hardware or software form in multimedia devices and are constructed such that listeners can listen to sounds as though they were in an ideal sound reproduction environment, are being used.

The front sound image localization algorithm, the low frequency extension algorithm, and the high frequency extension algorithm, described above, are components required for reproduction of three-dimensional sound in multimedia devices. In the front sound image localization algorithm, which is based on reverberating sound, variation in a three-dimensional sound effect depending on the characteristics of a sound converter is not large. However, in the low frequency extension algorithm and the high frequency extension algorithm, which increase bandwidth using harmonic components, the three-dimensional sound effect is considerably affected by the level values of the respective algorithms, so that the appropriate level adjustment of the harmonic components is required.

In order to solve such a problem, a menu for adjusting the level values of the low frequency band extension algorithm and the high frequency band extension algorithm while the user listens to music may be constructed in multimedia devices. However, in this case there are problems in that it is difficult for a general person to accurately adjust the level values while listening to music, the exact determination or evaluation of whether the level values have been appropriately adjusted cannot be made using music, the manipulation thereof is inconvenient and a lot of time must.be spent

Disclosure of the Invention

Accordingly, the present invention has been made keeping in mind the above problems occurring in the prior art, and an object of the present invention is to provide an apparatus and method for optimizing three-dimensional sound in a portable multimedia device which reproduces the three- dimensional sound by restoring low and high fiequency sounds, lost at the time of sound reproduction via a small-sized sound converter, using a psychoacoustic effect

Low and high fiequency reference sounds are generated using the respective harmonic components of low and high fiequency reference noise signals, a user alternately listens to the generated low and high fiequency reference sounds along with a middle frequency reference sound, that is, a middle frequency noise signal sound, thereby allowing the user to conveniently adjust the levels of the low and high frequency sounds based on subjective criteria while comparing and listening to the low frequency reference sound and the middle frequency reference sound, or the high frequency reference sound and the middle frequency reference sound. Low and high frequency sounds, which are restored at the time of sound reproduction via the multimedia device, are reproduced according to the user's level set values, thereby enabling reproduction of optimized three- dimensional sound.

In order to accomplish the above object, the present invention provides an apparatus for optimizing three-dimensional sound in a multimedia device, the multimedia device reproducing sound through a sound converter, the apparatus including a three-dimensional sound processing unit for analyzing the frequency spectrum of an audio file to be reproduced, generating harmonic components to restore low frequency harmonic components, so as to generate low frequency sounds corresponding to frequency components below a predetermined low cut-off frequency, and high frequency harmonic components to generate high frequency sounds corresponding to frequency components above a predetermined high cut-off frequency, and adjusting the levels of the generated low and high frequency harmonic components according to a user's level setting values for low and high frequency sounds, mixing the level-adjusted low and high frequency harmonic components with middle frequency components between the low and high cut-off frequencies of the audio file, and outputting the frequency components obtained by the mixing; and a harmonic component level adjustment unit for alternately generating the middle frequency reference sound at a fixed level and the low and high frequency reference sounds at varying levels, and providing the levels of the corresponding reference sounds to the three-dimensional sound processing unit as level setting values if there is user input for setting the levels of the low and high frequency reference sounds, which are currently being reproduced, to the levels of the low and high frequency sounds.

In addition, the present invention provides a method of reproducing optimized three- dimensional sound in a multimedia apparatus through a sound converter, the method including the steps of (A) analyzing the frequency spectrum of an audio file to be reproduced, and generating low frequency harmonic components, for generating low frequency sounds corresponding to frequency components below a predetermined low cut-off frequency, and high frequency harmonic components, for generating high frequency sounds corresponding to frequency components above a predetermined high cut-off frequency; (B) adjusting the levels of the generated low and high frequency harmonic components according to the user's level setting values for the low and high frequency sounds, and outputting the levels-adjusted low and high frequency harmonic components; and (C) mixing the output low and high frequency harmonic components with the middle frequency components of the audio file and then outputting frequency components obtained through the mixing.

Furthermore, the present invention further includes, before step (A), the steps of (A'-l) alternately reproducing a middle frequency reference sound at a fixed level and a high frequency reference sound at varying levels when the user selects a menu item for setting the level value of the high frequency sound; and (A'-2) storing the level of the high frequency reference sound, which is currently being reproduced, as a high frequency level sound setting value in response to user input for storing the level value of the high frequency sound.

Brief Description of the Drawings

The above and other objects, features and other advantages of the present invention will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings, in which: FIG. 1 is a diagram showing small-sized sound converters;

FIG. 2 is a block diagram showing the construction of a multimedia device in which an apparatus for optimizing three-dimensional sound is provided, according to an embodiment of the present invention;

FIG. 3 is a block diagram showing the construction of the three-dimensional sound processing unit of the apparatus for optimizing three-dimensional sound, according to an embodiment

of the present invention;

FIG.4 is a block diagram showing the construction of the harmonic level adjustment unit of tiie apparatus for optimizing three-dimensional sound, according to an embodiment of the present invention; FIG. 5 is a diagram illustrating the adjustment of harmonic levels in the apparatus for optimizing three-dimensional sound, according to an embodiment of the present invention;

FIG. 6 is a flowchart illustrating a harmonic level adjustment process in a method of optimizing three-dimensional sound, according to an embodiment of the present invention; and FIG. 7 is a flowchart illustrating a three-dimensional sound reproducing process in a method of optimizing three-dimensional sound, according to an embodiment of the present invention.

Best Mode for Carrying Out the Invention

FIG. 2 is a block diagram showing the construction of a multimedia device, in which an apparatus for optimizing three-dimensional sound is provided, according to an embodiment of the present invention.

In a multimedia device 100, including a key input unit 110, a display unit 120, memory 130, and a sound converter 140, when audio files are reproduced through a small-sized sound converter 140 such as earphones, the three-dimensional sound optimization apparatus 200 according to the present invention, as shown in FIG.2, is installed in the multimedia device 100 in hardware or software form in order to reproduce optimized three-dimensional sound.

The three-dimensional sound optimization apparatus 200 according to the present invention includes a three-dimensional sound processing unit 210 for restoring low and high frequency sounds using harmonic components and adjusting the levels of the restored low and high frequency sounds using level setting values for low and high frequency harmonic components, which are set by a user, and then outputting the level-adjusted low and high frequency sounds, and a harmonic component level adjustment unit 220 for allowing the user to set the level setting values.

Sounds corresponding to frequency components below a specific low cut-off frequency cannot be heard when audio is reproduced through the small-sized sound converter 140, so the present invention generates harmonic components of low fiequency components, which cannot be heard through the small-sized sound converter, in the overall fiequency band above the low cut-off fiequency, and uses a psychoacoustic effect to trick 1he user into believing that he or she were

listening to sound including a low fiequency component

In the present invention, the sounds generated using the harmonic components of the low fiequency components are referred to as low fiequency sounds. That is, the term "low fiequency sounds" refers to the sum of the harmonic fiequency components below the low cut-off fiequency. When the small-sized sound converter 140 does not reproduce sounds below 100 Hz well, the low cut-off fiequency is 100 Hz.

Furthermore, an audio file reproduced by the multimedia device is typically a compressed file, and fiequency components above a specific high cut-off fiequency are lost through the

compression process. Accordingly, the present invention restores the lost fiequency components above the high cut-off fiequency through the generation of harmonic components of fiequency components within a middle fiequency band.

Ih the present invention, sounds corresponding to the fiequency components above the high cut-off fiequency, restored using the fiequency components within the middle fiequency band, are high fiequency sounds. Because fiequency components above 16 kHz are lost from the audio file through the compression process, the high cut-off fiequency is 16 kHz. Meanwhile, the term "middle fiequency sounds" refers to sounds corresponding to

fiequency components that can be reproduced through the multimedia device. In the case in which the multimedia device is limited so as not to be able to reproduce sounds corresponding to fiequency

components below 100 Hz and in which the compressed audio file is a file fiom which fiequency components above 16 kHz are lost, a fiequency band for the middle fiequency sounds, for example, may be a frequency band ranging from 100 Hz to 16 kHz.

In the present invention, the three-dimensional sound processing unit 210 functions to restore the lost low and high frequency sounds through the generation of the harmonic components.

Furthermore, when a menu for adjusting the levels of harmonic components is operated by the user, the harmonic component level adjustment unit 220 alternately reproduces a low frequency reference sound and a middle frequency reference sound to adjust the levels of tfie low frequency sounds, and alternately reproduces a high frequency reference sound and the middle frequency reference sound to adjust the levels of the high frequency sounds, thus allowing the user to compare the middle frequency reference sound with the low and high frequency reference sounds while listening to them, and conduct subjective level adjustment

In this case, the low frequency reference sound is used to adjust the levels of the low frequency sounds, so that it is preferred that the low frequency reference sound be a sound obtained through a generation process identical to that of the low frequency sounds. Accordingly, the harmonic component level adjustment unit 220 generates the low frequency reference sound by generating the harmonic components of the noise signals within an octave band which has a center frequency at a specific frequency (for example, 60 Hz) below the low cut-off frequency. The term Octave band' refers to a frequency band in which the highest frequency is of the octave band double the lowest frequency of the octave band.

In the same manner, it is preferred that the high frequency reference sound be a sound obtained through a generation process identical to that of the high frequency sounds. Accordingly, in order to create components above the high cut-off frequency through the generation of harmonics, the harmonic component level adjustment unit 220 generates the harmonic components of the noise signals within an octave band which has a center frequency at a specific frequency (for example, 8 kHz) within a middle frequency band, so that the high frequency reference sound can be created. Furthermore, the middle frequency sounds reproduced through the multimedia device are the sounds of the original audio file within a reproducible frequency band, so that a separate process of generating harmonic components is not necessary, therefore the middle frequency reference sound is the sound of the noise signal itself within an octave band which has a center frequency at a specific

frequency (for example, 1 kHz) within the middle frequency band.

Meanwhile, when the user presses a button for adjusting a level value upward or downward, the harmonic component level adjustment unit 220 can reproduce sound while adjusting the levels of the low and high frequency reference sounds alternately reproduced along with the middle frequency reference sound at a fixed level, or it can reproduce sound while changing the levels of the low and high frequency reference sounds step by step at predetermined intervals without the user pressing a button. In the former case, when low and high frequency reference sounds, the levels of which are determined to be appropriate, are reproduced during the button pressing for adjusting levels, the user may set final level values when setting buttons are pressed. In the latter case, the levels are automatically adjusted, so that level values may be easily set when selection buttons are finally pressed at specific levels.

The respective constructions of the three-dimensional sound processing unit 210 and the harmonic component level adjustment unit 220 are described in detail with reference to FIGS. 3 to 5.

FIG. 3 is a block diagram showing the construction of the three-dimensional sound processing unit of the apparatus for optimizing three-dimensional sound, according to an embodiment of the present invention.

The three-dimensional sound processing unit 210 is responsible for restoring low and high frequency sounds by analyzing the frequency spectrum of an audio file to be reproduced and

generating harmonic components for low and high frequency components, and is responsible for allowing the user to adjust the levels of the generated low and high frequency harmonic components, mixing them with the middle frequency components, and then outputting the frequency components obtained through the mixing. For this purpose, the three-dimensional sound processing unit 210 includes a decoding unit 211, a Digital-to-Analog (DA) conversion unit 212, a spectrum analysis unit 213, a low frequency extension unit 214, a high frequency extension unit215, and amixing unit 216. The decoding unit 211 decompresses the compressed audio file and inputs the decompressed audio file to the DA conversion unit 212. The DA conversion unit 212 converts an audio bitstream to be reproduced into an analog signal. The multimedia device 100 generally reproduces a compressed audio file. However, if the multimedia device 100 reproduces an uncompressed audio file, the decoding unit 211 may be omitted. Since the high frequency sound is sound lost through the compression process, the high frequency extension unit 215 for restoring high frequency sounds may also be omitted.

Furthermore, the spectrum analysis unit 213 analyzes the spectrum of the converted analog signal and outputs the frequency components of the audio bitstream. In this case, outputting the frequency components allows the low frequency extension unit 214 and the high frequency extension unit 215 to restore low and high frequency sounds through the analysis of the output frequency components. The output frequency component includes low frequency components below a cutoff frequency, which are not reproduced well by the small-sized sound converter, and harmonic components of the high frequency components lost through the digital audio compression process, so that generation of the low and high frequency sounds is possible if the output frequency components are used.

The low frequency extension unit 214 includes a low frequency component analysis unit 214-1 for extracting low frequency components, which are components below the low cut-off

frequency of the multimedia device based on the result of the spectrum analysis, a low frequency harmonic component generation unit 214-2 for generating a low frequency harmonic component, which is a harmonic component that belongs to the harmonic components of the extracted low frequency components and is included in the audible range of the multimedia device (for example, 100 Hz ~ 16 kHz), that is, within a middle frequency band, and a low frequency harmonic component level adjustment unit 214-3 for adjusting the level of the generated low frequency

harmonic component to a low frequency sound level setting value provided from the harmonic component level adjustment unit 220 and outputting the level-adjusted low frequency harmonic

component

Furthermore, the high frequency extension unit 215 includes a high frequency component analysis unit 215-1 for detecting harmonic components for high frequency components above the high cut-off frequency based on the result of the spectrum analysis, a high frequency harmonic component generation unit 215-2 for generating a high frequency harmonic component based on the detected harmonic components, and a high frequency harmonic component level adjustment unit 215-3 for adjusting the level of the generated high frequency harmonic component to a high frequency sound level setting value provided from the harmonic component level adjustment unit 220 and outputting the level-adjusted high frequency harmonic component

The low and high frequency harmonic components, which are output from the low frequency extension unit 214 and the high frequency extension unit 215 as described above, are mixed with the middle frequency component through the mixing unit 216 and are then reproduced through the sound converter 140. FIG.4 is a block diagram showing the construction of the harmonic level adjustment unit of

the apparatus for optimizing three-dimensional sound, according to an embodiment of the present invention, and FIG. 5 is a diagram showing an example of a user menu representing the adjustment of

harmonic levels in the apparatus for optimizing three-dimensional sound, according to an

embodiment of the present inventioa In order to allow the user to set the level values of low and high fiequency sounds by comparing and listening to the middle frequency reference sound and the low and high frequency reference sounds, the harmonic component level adjustment unit 220 functions to alternately reproduce the middle frequency reference sound at a fixed level and low and high frequency

reference sounds at varying levels, and provide corresponding reference sound levels to the three- dimensional sound processing unit 210 as level setting values if there is the user input for adjusting the levels of the low and high frequency reference sounds, which are currently being reproduced, to the levels of the low and high frequency sounds.

The harmonic component level adjustment unit 220, as shown in FIG. 4, may include a control unit 221, a menu display unit 222, a middle frequency reference signal output unit 223, a low frequency reference signal output unit 224, a high frequency reference signal output unit 225, a level variation unit 226, a reference sound reproduction unit 227, and a level setting value storage unit 228. The control unit 221 is not only responsible for control of variation in the low and high frequency reference sounds, which are alternately reproduced along with the middle frequency reference sound, and the alternate reproduction of the middle frequency reference sound and the low and high frequency reference sounds, but is also responsible for the overall operational control of the harmonic component level adjustment unit 220, including the display of the user menu.

The menu display unit 222, as in the embodiment of FIG. 5, functions to display a user menu, including an adjustment target display portion A for displaying whether a current level setting target is a low frequency sound or a high frequency sound, and a level display portion B for

displaying the level value and level setting range of a low or high frequency reference sound currently being output through the reference sound reproduction unit 227. The construction of the menu of

FIG. 5 is only an embodiment, and the present invention is not limited thereto.

Furthermore, as shown in FIG. 5, in the present invention, the level setting values of the low and high frequency sounds may be set to fall within a range from -5 to +5, but are not limited thereto.

Furthermore, the middle frequency reference signal output unit 223 generates the middle frequency reference signal at a fixed reference level, the low frequency reference signal output unit 224 generates the low frequency reference signal at a fixed level, and the high frequency reference signal output unit 225 generates the high frequency reference signal at a fixed level. In this case, each of the reference signal output units 223, 224 and 225 merely outputs sound signals stored in advance, and may include an oscillator for generating a reference signal.

Furthermore, the above-described middle frequency reference signal may be, for example, a noise signal in an octave band which has a center frequency of 1 kHz, that is, an intermediate frequency in the audible range of the multimedia device, the low frequency reference sound may be a signal for the harmonic components of a noise signal within an octave band which has a center frequency of 60 Hz, that is, an intermediate frequency that belongs to frequencies below a low cutoff frequency, and the high frequency reference signal may be a signal for the harmonic components of a noise signal within an octave band which has a center frequency of 8 kHz, that is, an intermediate frequency that belongs to frequencies above a high cut-off frequency. Since the respective center frequencies may be set within an appropriate range, the present invention is not limited thereto.

Meanwhile, the low and high frequency reference signals are reference signals for adjusting the levels of the low and high frequency sounds at the time of reproduction of three-dimensional

sound, so it is preferred that they be signals generated using a frequency extension algorithm applied when generating the low and high frequency sounds.

Accordingly, the low frequency reference signal output unit 224 includes a low frequency noise signal generating unit 224-1 for generating a low frequency noise signal sound to generate a low frequency reference sound, a low frequency spectrum analysis unit 224-2 for analyzing the frequency components of tiie low frequency noise signal sound, and a low frequency reference signal generation unit 224-3 for generating the harmonic component of the low frequency noise signal sound within the middle frequency band using the results of the spectrum analysis. The low frequency noise signal generating unit 224-1 generates a low frequency noise signal within an octave band which has a center frequency at a specific frequency below the cut-off frequency of the small-sized sound converters within a low frequency band, and inputs the generated low frequency noise signal to the low frequency spectrum analysis unit 224-2.

The low frequency spectrum analysis unit 224-2 analyzes the spectrum of the generated low frequency noise signal and outputs the frequency components of the low frequency noise signal to the low frequency reference signal generation unit 224-3. The low frequency reference signal generation unit 224-3 generates a harmonic component, which belongs to the harmonic components of the low frequency noise signal within the middle frequency band, using the frequency components of the low frequency noise signal, and outputs the generated harmonic component as a low frequency reference signal.

The high frequency reference signal output unit 225 includes a high frequency noise signal generating unit 225-1, a high frequency spectrum analysis unit 225-2, and a high frequency reference signal generation unit 225-3. The high frequency noise signal generating unit 225-1 generates a high frequency noise signal within an octave band which has a center frequency at a specific frequency within the middle frequency band, the high frequency spectrum analysis unit 225-2

analyzes the spectrum of the generated high frequency noise signal and outputs the frequency

components of the high frequency noise signal, and the high frequency reference signal generation

unit 225-3 generates a harmonic component, which belongs to the harmonic components of the high frequency noise signal within a lost high frequency band, using the frequency components of the high frequency noise signal. The harmonic component generated as described above is output as a high frequency reference signal.

As described above, the high frequency noise signal is a noise signal within the middle frequency band and a signal for generating the high frequency reference sound. Accordingly it is referred to as a high frequency noise signal.

Meanwhile, the low frequency spectrum analysis unit 224-2 and the low frequency reference signal generation unit 224-3 function to generate the low frequency reference signal using a method identical to that used to generate the low frequency sound. In the same manner, the high frequency spectrum analysis unit 225-1 and the high frequency reference signal generation unit 225-3 also function to generate the high frequency reference signal using a method identical to that used to generate the high frequency sound.

Accordingly, the reference signal output units 224 and 225 may incorporate the construction of the three-dimensional sound processing unit 210 without including the spectrum analysis units 224-2 and 225-2 or the reference signal generation units 224-3 and 225-3. In this case, the low frequency reference signal output unit 224 transfers the low frequency noise signal to the spectrum analysis unit 213 of the three-dimensional sound processing unit 210. The low frequency noise signal is converted into a low frequency reference signal through the low frequency component analysis unit 214-1 and the low frequency harmonic component generation unit 214-2. When the converted low frequency reference signal is fed back to the low frequency reference signal output unit 224, the low frequency reference signal output unit 224 can output the low frequency reference signal.

Meanwhile, since the low frequency noise signal includes only low frequency components, a separate analysis of the low frequency components may be omitted.

Furthermore, in the same manner as that described above, the high frequency reference signal output unit 225 transfers the high frequency noise signal to the spectrum analysis unit 213 of the three-dimensional sound processing unit 210. The high fiequency noise signal is converted into a high fiequency reference signal through the high frequency component analysis unit 215-1 and the high frequency harmonic component generation unit 215-2. When the converted high

frequency reference signal is fed back to the high frequency reference signal output unit 225, the high frequency reference signal output unit 225 can output the high frequency reference signal.

Under the control of the control unit 221, the level variation unit 226 increases or decreases the level of the low frequency reference signal generated by the low frequency reference signal output unit 224 and the level of the high frequency reference signal generated by the high frequency reference signal output unit 225, and then transfers the low and high frequency reference signals to the reference sound reproduction unit 227.

In this case, the control unit 221 may perform control so that the level variation unit 226 automatically increases or decreases the levels of the low and high frequency reference signals at regular intervals, and may transfer variation control commands to the level variation unit 226 according to the user's key presses for adjusting the levels. In this case, the control unit 221 may directly provide respective level values to the level variation unit 226 as variation control commands, and the level variation unit 226 adjusts the level values upward or downward according to the variation control commands, or uses received level values for adjustment

Under the control of the control unit 221, the reference sound reproduction unit 227 alternately reproduces the low frequency reference sound, that is, the low frequency reference signal sound at a predetermined level, or the high frequency reference sound, that is, the high frequency

reference signal sound, which is output from the level variation unit 226, along with the middle frequency reference sound, that is, the middle frequency reference signal sound at a fixed level. For example, the level of the low frequency sound may be set while a middle frequency reference sound, the length of which is 1 sec, and a low fiequency reference sound, the length of which is 1 sec, are alternately reproduced at intervals of 1 sec, and the level of the high frequency sound may be set while a middle frequency reference sound, the length of which is 1 sec, and a high frequency reference sound, the length of which is 1 sec, are also alternately reproduced at intervals of 1 sec. Li the present invention, the reproduction length or interval of each reference sound is not limited thereto.

Thereafter, if there is user input for setting the levels of the low and high frequency reference sounds, which are currently being reproduced by the reference sound reproduction unit 227, to the levels of the low and high frequency sounds, the level setting value storage unit 228 stores the levels of the respective reference sounds as level setting values, and provides the level setting values to the low and high frequency extension units 214 and 215 of the three-dimensional sound processing unit 210, respectively, under the control of the control unit 221.

That is, if a button for setting is pressed by the user when the middle frequency reference sound and the low and high frequency reference sound at specific levels are alternately reproduced, the control unit 221 stores the current level value in the level setting value storage unit 228.

Information stored in the level setting value storage unit 228 is used by the three-dimensional sound processing unit 210 to adjust the levels of the low and high frequency sounds.

For this purpose, the low and high frequency harmonic component level adjustment units 214-3 and 215-3 of the three-dimensional sound processing unit 210 may perform an operation of reading the level setting values 221 while directly operating in conjunction with the level setting value

storage unit 228 or requesting and receiving the level setting values from the control unit, or may receive the level setting values from the control unit 221 in advance and store the received level setting value in separate memory.

In this case, if there is user input for setting the level of the low or high frequency reference sound while the low or high frequency reference sound at a specific level and the middle frequency reference sound at a fixed level are alternately reproduced through the reference sound reproduction unit 227, the control unit 221 may stop the reproduction operation of the reference sound reproduction unit 227, and store a corresponding level value. . A process of optimizing three-dimensional sound using the above-described three- dimensional optimization sound device 200 according to the present invention is sequentially described with reference to FIGS. 6 and 7 below.

FIG. 6 is a flowchart illustrating a harmonic level adjustment process in a method of optimizing three-dimensional sound, and shows an example of the operation of the sound harmonic component level adjustment unit 220 of the three-dimensional optimization apparatus 200.

As shown in FIG. 6, when a menu item for adjusting the levels of harmonic components is selected by the user at step S301, the harmonic component level adjustment unit 220 alternately reproduces the middle frequency reference sound at a fixed level and the low frequency reference sound at a reference level (for example, a 0 level) at step S303. If a user setting button for performing setting to a current level is pressed at step S305, the level of the low frequency reference sound reproduced at step S303 is stored as a low frequency sound level setting value at step S309.

Thereafter, if the setting button is not pressed while reproduction is repeated the predetermined number of times, the harmonic component level adjustment unit 220 increases or decreases the level at step 307 by adjusting the level of the low frequency reference sound in steps, and then alternately reproduces again the middle frequency reference sound at a fixed level and the low frequency reference sound; the level of which is increased, at step S303.

Meanwhile, the present embodiment is an operational process in the case where

reproduction is performed while the level is automatically increased or decreased without user input. In another embodiment, after step S303 has been performed, the step S307 of increasing or decreasing the level of the low frequency reference sound according to the user input for varying the level may be performed, and automatic variation and variation depending on the user's input may both be performed.

Thereafter, when the value for setting the level of the low frequency sound is stored according to the input of the setting button at step S309 after the low frequency reference sound has been reproduced through the automatic variation or variation in response to user input for the low frequency reference sound, the harmonic component level adjustment unit 220 alternately reproduce again the middle frequency reference sound at a fixed level and the high frequency reference sound at a reference level to set the level of the high frequency sound at step S311. Thereafter, at step S315, the level of the high frequency reference sound is increased or decreased until the setting button is pressed at step S313. If the setting button is pressed at step S313, the level value of current high frequency reference sound is stored as a high frequency sound level setting value at step S317.

Meanwhile, although, in the embodiment of FIG. 6, a description of the , harmonic component level adjustment unit 220 sequentially performing the operation of setting the level of the low frequency sound and the operation of setting the level of the high frequency sound is made, a process of setting the respective levels may be performed in such a manner that the level of low frequency sound is set according to selection through the menu item for setting the level of the low frequency sound, and the level of high frequency sound is set according to selection through the menu item for setting the level of the high frequency sound level. Furthermore, although, in the above description, a corresponding level value is stored in the case where the user presses the setting button, a current level value may be automatically stored even

in the case where an 'exit' menu item is selected from the harmonic component level adjustment menu, instead requiring a separate setting button to be pressed.

Furthermore, although not described in the flowchart of FIG. 6, it is preferred that, at the time of the operation of setting the respective levels, the harmonic component level adjustment unit 220 not only indicates whether a current adjustment target is a high frequency sound or a low frequency sound, but also displays the level value of the current reproduced high and low frequency reference sounds by displaying the menu shown in FIG. 5.

FIG. 7 is a flowchart illustrating a three-dimensional sound reproducing process in a method of optimizing three-dimensional sound, according to an embodiment of the present invention, and is a flowchart illustrating a three-dimensional sound reproducing process performed after the user setting operation in the process of FIG.6 has been performed.

As shown in FIG. 7, when a compressed audio file to be reproduced is input at step S401, the three-dimensional sound processing unit 210 of the three-dimensional optimization apparatus 200 according to the present invention decodes the compressed file at step S403 and then converts a digital bitstream to be reproduced into an analog signal at step S405. In the case of an

uncompressed audio file, the decoding step S403 may be omitted.

Thereafter, the three-dimensional sound processing unit 210 generates the harmonic components of frequencies below the low cut-off frequency using the frequency components of the audio bitstream, and analyzes the frequency spectrum of the converted analog signal to restore frequencies above the high cut-off frequency, at step S407.

Thereafter, the three-dimensional sound processing unit 210 analyzes low frequency components included in the frequency components of the audio bitstream at step S409-1, generates a corresponding low frequency harmonic component within the audible range at step S409-3, and

adjusts the level of the low frequency harmonic component according to a level setting value at step S409-7 if corresponding level setting values have been set by the user at step S409-5. If there is no

user's level setting value, the low frequency harmonic component is set to a default level without a separate step of adjusting the level. Furthermore, the three-dimensional sound processing unit 210 analyzes and detects a high frequency harmonic component from the frequency components of the audio bitstream at step S411- 1, restores high frequency components, which were lost through a compression process, by generating the corresponding harmonic component of the detected high frequency harmonic components at step S411-3, and adjusts the levels of the high frequency components, which are restored by the user according to a level setting value set at step S411-5, at step S411-7.

When the adjustment of the levels of the low and high frequency components is completed as described above, the three-dimensional sound processing unit 210 mixes the original frequency components, that is, the frequency components of the original audio bitstream, with the level-adjusted low and high frequency components and outputs the mixed frequency components, thus allowing three-dimensional sound optimized for the user to be played through the small-sized sound converter 140 of the multimedia device 100 at step S413.

Industrial Applicability

Accordingly, in the apparatus and method for optimizing three-dimensional sound according to the present invention, a middle frequency reference sound and respective low and high frequency reference sounds are alternately reproduced in the multimedia device while the levels of the low frequency sound generated using harmonic components within an audible frequency band to reproduce three-dimensional sound and the restored high frequency sound, are set, so that the user can adjust the levels optimized for his or her subjective listening experience. Furthermore, the low

and high frequency reference sounds are generated using low and high frequency sound extension

algorithms actually applied at the time of reproduction of the three-dimensional sound, so that the levels can be precisely adjusted. Furthermore, the low and high frequency sounds are reproduced according to the level values set by the user, so that optimized three-dimensional sound can be reproduced.

Although the preferred embodiments of the present invention have been disclosed for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims.

Claims

1. AQ apparatus for optimizing three-dimensional sound in a multimedia device, the multimedia device reproducing sound through a sound converter, the apparatus comprising:

a three-dimensional sound processing unit for analyzing a frequency spectrum of an audio file ^"to be reproduced, generating harmonic components to restore low frequency harmonic components, so as to generate low frequency sounds corresponding to fiequency components below a predetermined low cut-off frequency, and high fiequency harmonic components to generate high fiequency sounds corresponding to fiequency components above a predetermined high cut-off fiequency, and adjusting the levels of the generated low and high fiequency harmonic components according to a user's level setting values for low and high frequency sounds, mixing the level- adjusted low and high fiequency harmonic components with middle fiequency components between

the low and high cut-off frequencies of the audio file, and outputting the fiequency components obtained by the mixing; and a harmonic component level adjustment unit for alternately generating a middle fiequency reference sound at a fixed level and low and high frequency reference sounds at varying levels, and providing the levels of the corresponding reference sounds to the three-dimensional sound processing unit as level setting values if there is user input for setting levels of low and high frequency reference

sounds, which are currently being reproduced, to levels of the low and high frequency sounds.

2. The apparatus according to claim 1, wherein the harmonic component level adjustment unit comprises:

a control unit for controlling an increase or decrease of the respective low and high frequency reference sounds alternately reproduced along with the middle frequency reference sound, and alternate reproduction of the middle frequency reference sound and the respective low and high frequency reference sounds; a middle frequency reference signal output unit for outputting the middle frequency reference signal at a fixed level; a low frequency reference signal output unit for outputting a low frequency reference signal at a fixed level; a high frequency reference signal output unit for outputting a high frequency reference signal at a fixed level; a level variation unit for increasing or decreasing the level of the low frequency reference signal, which is output from the low frequency reference signal output unit, and the level of the high frequency reference signal, which is output from the high frequency reference signal output unit, under control of the control unit; a reference sound reproduction unit for alternately reproducing the low frequency reference sound, that is, a low frequency reference signal sound at a predetermined level, or the high frequency reference sound, that is, a high frequency reference signal sound at a predetermined level, along with the middle frequency reference sound, that is, a middle frequency reference signal sound at a fixed level, levels of the low and high frequency reference signal sounds being increased or decreased by the level variation unit; and a level setting value storage unit for storing the levels of the corresponding reference sounds as the level setting values under control of the control unit, if there is user input for setting the levels of the low and high frequency reference sounds, which are currently being reproduced by the reference sound reproduction unit, to the levels of the low and high frequency sounds restored at a time of reproduction of the three-dimensional sound.

3. The apparatus according to claim 2, wherein the control unit stops a sound reproducing operation of the reference sound reproduction unit, and stores the levels of the low and high frequency reference sounds, which are currently being reproduced, in the level setting value storage unit as the level setting values of the low and high frequency sounds, if a button for setting the levels of the low and high frequency reference sounds is pressed by the user while low and high frequency reference sounds at a specific level and the middle frequency reference sound at a fixed level are alternately reproduced through the reference sound reproduction unit

4. The apparatus according to claim 2, wherein the control unit performs control so as to automatically increase or decrease the levels of the low and high frequency reference signals at regular intervals.

5. The apparatus according to claim 2, wherein the control unit transfers a level value depending on the user input to the level variation unit if a button for increasing or decreasing the level of the low or high frequency reference sounds reproduced through the reference sound reproduction unit is pressed by the user; and the level variation unit increases or decreases the levels of the low and high frequency reference signals output through the reference sound reproduction unit according to level values transferred from the control unit and provides the low and high frequency reference signals to the reference sound reproduction unit

6. The apparatus according to claim 2, wherein the harmonic component level adjustment unit further comprises a menu display unit for displaying a menu, the menu display unit comprising an adjustment target display portion for displaying whether a current level setting target is a low frequency sound or a high frequency sound, and a level display portion for displaying level values of low and high frequency reference sounds currently being output through the reference sound reproduction unit and level setting ranges.

7. The apparatus according to any one of claims 2 to 6, wherein the middle frequency reference signal output unit outputs a noise signal within an octave band which has a center frequency at a specific frequency within a middle frequency band, that is, an audible range of the multimedia device, as the middle frequency reference signal.

8. The three-dimensional sound optimizing device according to claim 7, wherein the middle frequency reference signal is a noise signal within an octave band which has a center frequency of 1 kHz.

9. The apparatus according to any one of claims 2 to 6, wherein the low frequency reference signal output unit comprises: a low frequency noise signal generation unit for generating a low frequency noise signal within an octave band which has a center frequency at a specific frequency within a low frequency band below an audible range of the multimedia device; a low frequency spectrum analysis unit for analyzing a spectrum of the generated low frequency noise signal and outputting frequency components of the low frequency noise signal; and

a low frequency reference signal generation unit for generating a harmonic component, which belongs to harmonic components of the low frequency noise signal and is included in the

middle frequency band, using the frequency components of the low frequency noise signal, and outputting the generated harmonic component as the low frequency reference signal.

10. The apparatus according to claim 9, wherein the low frequency noise signal is a noise signal within an octave band which has a center frequency of 60 Hz.

11. The apparatus according to any one of claims 2 to 6, wherein the high frequency reference signal output unit comprises: a high frequency noise signal generation unit for generating a high frequency noise signal within an octave band which has a center frequency at a specific frequency that is included in. a middle frequency band, that is, an audible range of the multimedia device; a high frequency spectrum analysis unit for analyzing a spectrum of the generated high frequency noise signal and outputting frequency components of the high frequency noise signal; and a high frequency reference signal generation unit for generating a harmonic component,

which belongs to harmonic components of the high frequency noise signal and is included in a high frequency band above the audible range, using the frequency components of the high frequency noise signal, and outputting the generated harmonic component as the high frequency reference signal.

12. The apparatus according to claim 11, wherein the high frequency noise signal is a noise signal within an octave band which has a center frequency of 8 kHz.

13. The apparatus according to any one of claims 1 to 6, wherein the three-dimensional

sound processing unit comprises: a Digital-to-Analog (DA) conversion unit for converting the audio bitstream to be reproduced into an analog signal;

a spectrum analysis unit for analyzing a spectrum of the converted analog signal and outputting the frequency components of the audio bitstreatn; a low frequency extension unit for extracting low frequency components, which belong to the frequency components output from the spectrum analysis unit and are below the low cut-off frequency, generating a harmonic component, which belongs to harmonic components of the extracted low frequency components and is included in the audible band, and adjusting a level of the generated low frequency harmonic component to a low frequency sound level setting value provided by the harmonic component level adjustment unit; a high frequency extension unit for detecting harmonic components for high frequency components, which belong to the frequency components output from the spectrum analysis unit and are below the high cut-off frequency, and generating a high frequency harmonic component based on the detected harmonic components and, thereby, restoring high frequency components lost through a compression process, adjusting a level of the generated high frequency harmonic component to a high frequency sound level setting value provided by the harmonic component level adjustment unit, and outputting the level-adjusted high frequency harmonic components; and a mixing unit for mixing the low frequency harmonic component, which is output from the low frequency extension unit, and the high frequency harmonic component, which is output from the high frequency extension unit, with an original frequency component, which is output from the DA conversion unit and outputting frequency components obtained through the mixing.

14. The apparatus according to claim 13, wherein the low frequency extension unit comprises: a low frequency component analysis unit for extracting low frequency components that belong to a frequency component output from the spectrum analysis unit and are below the low cutoff frequency; a low frequency harmonic component generation unit for generating a harmonic component that belongs to harmonic components of the extracted low frequency components and is included in the audible band; and a low frequency harmonic component level adjustment unit for adjusting a level of the generated low frequency harmonic component according to the user's level setting value, and outputting the level-adjusted low frequency harmonic components if there is a user's setting value corresponding to a level of the low frequency sound, and outputting a low frequency harmonic component of a preset reference level if there is no user's setting value.

15. The apparatus according to claim 13, wherein the high frequency extension unit comprises: a high frequency component analysis unit for detecting harmonic components for high frequency components that belong to frequency component output from the spectrum analysis unit and are above the high cut-off frequency; a high frequency harmonic component generation unit for generating a high frequency harmonic component based on the detected harmonic components and, thereby, restoring the high frequency components lost through the compression process; and a high frequency harmonic component level adjustment unit for adjusting a level of the generated high frequency harmonic component according to the user's level setting value, and outputting the level-adjusted high frequency harmonic component if there is a user's setting value corresponding to a level of the low frequency sound, and outputting a high frequency harmonic

component of a preset reference level if there is no user's setting value.

16. The apparatus according to claim 13, wherein the three-dimensional sound processing unit further comprises a decoding unit for decompressing the compressed audio file and inputting the decompressed audio file to the DA conversion unit.

17. An apparatus for optimizing three-dimensional sound in a multimedia device, the

multimedia device reproducing sound through a sound converter, the apparatus comprising: a three-dimensional sound processing unit for analyzing a frequency spectrum of an audio file to be reproduced, generating harmonic components to restore low frequency harmonic components, for generating low frequency sounds corresponding to frequency components below a predetermined low cut-off frequency, and high frequency harmonic components, for generating high frequency sounds corresponding to frequency components above a predetermined high cut-off frequency, and adjusting the levels of the generated low and high frequency harmonic components according to a user's level setting values for low and high frequency sounds, mixing the level- adjusted low and high frequency harmonic components with middle frequency components between low and high cut-off frequencies of the audio file, and outputting the frequency components obtained bythernixing.

18. The apparatus according to claim 17, wherein the three-dimensional sound processing

unit comprises: a Digital-to-Analog (DA) conversion unit for converting the audio bitstream to be reproduced into an analog signal;

a spectrum analysis unit for analyzing a spectrum of the converted analog signal and outputting the frequency components of the audio bitstream;

a low frequency extension unit for extracting low frequency components, which belong to the frequency components output from the spectrum analysis unit and are below the low cut-off frequency, generating a harmonic component, which belongs to harmonic components of the extracted low frequency components and is included in the audible band, and adjusting a level of the generated low frequency harmonic component to a low frequency sound level setting value provided by the harmonic component level adjustment unit;

a high frequency extension unit for detecting harmonic components for high frequency components, which belong to the frequency components output from the spectrum analysis unit and are below the high cut-off frequency, and generating a high frequency harmonic component based on the detected harmonic components and, thereby, restoring high frequency components lost through a compression process, adjusting a level of the generated high frequency harmonic component to a high frequency sound level setting value provided by the harmonic component level adjustment unit, and outputting the level-adjusted high frequency harmonic components; and a mixing unit for mixing the low frequency harmonic component, which is output from the low frequency extension unit, and the high frequency harmonic component, which is output from the high frequency extension unit, with an original frequency component, which is output from the DA conversion unit and outputting frequency components obtained through the mixing.

19. The apparatus according to claim 18, wherein Ihe low frequency extension unit comprises: a low frequency component analysis unit for extracting low frequency components that belong to a frequency component output from the spectrum analysis unit and are below the low cut¬

off frequency; a low frequency harmonic component generation unit for generating a harmonic

component that belongs to harmonic components of the extracted low frequency components and is included in the audible band; and a low frequency harmonic component level adjustment unit for adjusting a level of the generated low frequency harmonic component according to the user's level setting value, and outputting the level-adjusted low frequency harmonic component if there is a user's setting value corresponding to a level of the low frequency sound, and outputting a low frequency harmonic component of a preset reference level if there is no user's setting value.

20. The apparatus according to claim 18, wherein the high frequency extension unit comprises: a high frequency component analysis unit for detecting harmonic components for high frequency components that belong to frequency component output from the spectrum analysis unit and are above the high cut-off frequency; a high frequency harmonic component generation unit for generating a high frequency harmonic component based on the detected harmonic components and, thereby, restoring the high frequency components lost through the compression process; and a high frequency harmonic component level adjustment unit for adjusting a level of the generated high frequency harmonic component according to the user's level setting value, and outputting the level-adjusted high frequency harmonic component if there is a user's setting value corresponding to a level of the low frequency sound, and outputting a high frequency harmonic component of a preset reference level if there is no user's setting value.

21. The apparatus according to claim 18, wherein the three-dimensional sound processing unit ftuther comprises a decoding unit for decompressing the compressed audio file and inputting the

decompressed audio file to the DA conversion unit

22. An apparatus for optimizing three-dimensional sound in a multimedia device, the multimedia device reproducing sound through a sound converter, the apparatus comprising: a harmonic component level adjustment unit for alternately generating a middle frequency reference sound at a fixed level and low and high frequency reference sounds at varying levels, and

storing levels of the corresponding reference sounds as level setting values if there is user input for setting levels of low and high frequency reference sounds, which are currently being reproduced, to levels of the low and high frequency sounds.

23. The apparatus according to claim 22, wherein the harmonic component level

adjustment unit comprises: a control unit for controlling increase or decrease of the respective low and high frequency reference sounds alternately reproduced along with the middle frequency reference sound, and alternate reproduction of the middle frequency reference sound and the respective low and high frequency reference sounds; a middle frequency reference signal output unit for outputting the middle frequency reference signal at a fixed level; a low frequency reference signal output unit for outputting a low frequency reference signal

at a fixed level; a high frequency reference signal output unit for outputting a high frequency reference

signal at a fixed level; a level variation unit for increasing or decreasing a level of the low frequency reference signal, which is output from the low frequency reference signal output unit, and a level of the high

frequency reference signal, which is output from the high frequency reference signal output unit,

under control of the control unit; a reference sound reproduction unit for alternately reproducing the low frequency reference sound, that is, a low frequency reference signal sound at a predetermined level, or the high frequency reference sound, that is, a high frequency reference signal sound at a predetermined level, along with the middle frequency reference sound, that is, a middle frequency reference signal sound at a fixed

level, levels of the low and high frequency reference signal sounds being increased or decreased by the level variation unit; and a level setting value storage unit for storing levels of the corresponding reference sounds as the level setting values under control of the control unit, if there is user input for setting the levels of the low and high frequency reference sounds, which are currently being reproduced by the reference sound reproduction unit, to the levels of the low and high frequency sounds restored at a time of reproduction of the three-dimensional sound.

24. The apparatus according to claim 23, wherein the control unit stops a sound reproducing operation of the reference sound reproduction unit, and stores the levels of the low and high frequency reference sounds, which are currently being reproduced, in the level setting value storage unit as the level setting values of the low and high frequency sounds, if a button for setting the levels of the low and high frequency reference sounds is pressed by the user while low and high frequency reference sound at a specific level and the middle frequency reference sound at a fixed level are

alternately reproduced through the reference sound reproduction unit

25. The apparatus according to claim 23, wherein the control unit performs control so as to

automatically increase or decrease levels of the low and high frequency reference signals at regular intervals.

26. The apparatus according to claim 23, wherein the control unit transfers a level value depending on the user input to the level variation unit if a button for increasing or decreasing the level of the low or high frequency reference sounds reproduced through the reference sound reproduction unit is pressed by the user; and the level variation unit increases or decreases the levels of the low and high frequency reference signals output through the reference sound reproduction unit according to level values transferred from the control unit and providing the low and high frequency reference signals to the reference sound reproduction unit

27. The apparatus according to claim 23, wherein the harmonic component level adjustment unit further comprises a menu display unit for displaying a menu, the menu display unit comprising an adjustment target display portion for displaying whether a current level setting target is a low frequency sound or a high frequency sound, and a level display portion for displaying level values of low and high frequency reference sounds currently being output through the reference sound reproduction unit and level setting ranges.

28. The apparatus according to any one of claims 23 to 27, wherein the middle frequency reference signal output unit outputs a noise signal within an octave band which has a center frequency at a specific frequency within a middle frequency band, that is, an audible range of the multimedia device, as the middle frequency reference signal.

29. The three-dimensional sound optimizing device according to claim 28, wherein the middle frequency reference signal is a noise signal within an octave band which has a center frequency of 1 kHz.

30. The apparatus according to any one of claims 23 to 27, wherein the low frequency reference signal output unit comprises: a low frequency noise signal generation unit for generating a low frequency noise signal within an octave band which has a center frequency at a specific frequency within a low frequency band below an audible range of the multimedia device; a low frequency spectrum analysis unit for analyzing a spectrum of the generated low frequency noise signal and outputting frequency components of the low frequency noise signal; and a low frequency reference signal generation unit for generating a harmonic component, which belongs to harmonic components of the low frequency noise signal and is included in the middle frequency band, using the frequency components of the low frequency noise signal, and outputting the generated harmonic component as the low frequency reference signal.

31. The apparatus according to claim 30, wherein the low frequency noise signal is a noise signal within an octave band which has a center frequency of 60 Hz.

32. The apparatus according to any one of claims 23 to 27, wherein the high frequency reference signal output unit comprises: a high frequency noise signal generation unit for generating a high frequency noise signal within an octave band which has a center frequency at a specific frequency that is included in a

middle frequency band, that is, an audible range of the multimedia device; a high frequency spectrum analysis unit for analyzing a spectrum of the generated high

frequency noise signal and outputting frequency components of the high frequency noise signal; and a high frequency reference signal generation unit for generating a harmonic component, which belongs to harmonic components of the high frequency noise signal and is included in a high frequency band above the audible range, using the frequency components of the high frequency noise signal, and outputting the generated harmonic component as the high frequency reference signal.

33. The apparatus according to claim 32, wherein the high frequency noise signal is a noise signal within an octave band which has a center frequency of 8 kHz.

34. A method of reproducing optimized three-dimensional sound in a multimedia apparatus for reproducing sound through a sound converter, the method comprising the steps of: (A) analyzing a frequency spectrum of an audio file to be reproduced, and generating low frequency harmonic components, for generating low frequency sounds corresponding to frequency components below a predetermined low cutoff frequency, and high frequency harmonic

components, for generating high frequency sounds corresponding to frequency components above a predetermined high cut-off frequency; (B) adjusting the levels of the generated low and high frequency harmonic components according to the user's level setting values for the low and high frequency sounds, and outputting the levels-adjusted low and high frequency harmonic components; and

(C) mixing the output low and high frequency harmonic components with middle frequency components of the audio file and then outputting frequency components obtained through

the mixing.

35. The method according to claim 34, wherein the step (A) comprises the steps of:

(A-I) converting a bitstream of the audio file to be reproduced into an analog signal; (A-2) analyzing a spectrum of the converted analog signal and extracting frequency components that belong to frequency components of the audio bitstream and are below a low cut-off frequency of the sound converter provided in the multimedia device; and

(A-3) generating a low frequency harmonic component, that is, a harmonic component that belongs to harmonic components of the extracted low frequency components and is included in an

audible range.

36. The method according to claim 35, wherein the step (B) comprises the steps of:

(B-I) determining whether there is a user's level setting value corresponding to a level of the generated low frequency harmonic component; and (B-2) adjusting the levels of the generated low frequency harmonic component according to the user's level setting value, and outputting the level-adjusted low frequency harmonic component if there is a user's level setting value, and outputting a low frequency harmonic component of a preset reference level if there is no user's level setting value.

37. The method according to claim 34, wherein the step (A) comprises the steps of:

(A-I) converting a bitstream of the audio file to be reproduced into an analog signal; (A-2) analyzing a spectrum of the converted analog signal and, thereby, detecting harmonic components that belong to frequency components of the audio bitstream and correspond to high frequency components above an audible range; and (A-3) generating a high frequency harmonic component based on the detected harmonic components and, thereby; restoring high frequency components lost through a compression process.

38. The method according to claim 37, wherein the step (B) comprises the steps of:

(B-I) determining whether there is a user's level setting value corresponding to a level of the restored high frequency component; and

(B-2) adjusting a level of the generated high frequency harmonic component according to the user's level setting value, and outputting the level-adjusted high frequency harmonic component if there is a user's level setting value, and outputting a high frequency harmonic component of a preset

reference level if there is no user's level setting value.

39. The method according to claim 34, further comprising, before the step (A), the steps of: (A'-l) alternately reproducing a middle frequency reference sound at a fixed level and a low frequency reference sound at varying levels when the user selects a menu item for setting a level value of the low frequency sound; and

(A' -2) storing a level of a low frequency reference sound, which is currently being produced, as a low frequency level sound setting value in response to user input for storing the level

value of the low frequency sound.

40. The method according to claim 39, wherein the step (A'-2) is the step of stopping alternate reference sound reproduction operation and storing the level of the low frequency reference sound, which is currently being reproduced, as the low frequency sound level setting value in response to user input for storing the level value of the low frequency sound.

41. The method according to claim 39, wherein, at the step (A'-l), the level of the

reproduced low frequency reference sound increases or decreases automatically at regular intervals.

42. The method according to claim 39, wherein, at the step (A'-l), the level of the reproduced low frequency reference sound increases or decreases according to user input

43. The method according to claim 34, further comprising, before the step (A), the steps of: (A'-l) alternately reproducing a middle frequency reference sound at a fixed level and a high frequency reference sound at varying levels when the user selects a menu item for setting a level value of the low frequency sound; and

(A' -2) storing a level of a high frequency reference sound, which is currently being produced, as a high frequency level sound setting value in response to user input for storing the level value of the high frequency sound.

44. The method according to claim 43, wherein the step (A'-2) is the step of stopping the alternate reference sound reproduction operation and storing the level of the high frequency reference sound, which is currently being reproduced, as the high frequency sound level setting value in response to user input for storing the level value of the high frequency sound.

45. The method according to claim 43, wherein, at the step (A'-l), the level of the reproduced high frequency reference sound increases or decreases automatically at regular intervals.

46. The method according to claim 43, wherein, at the step (A'-l), the level of the reproduced high frequency reference sound increases or decreases according to user input.

47. The method according to claim 39, wherein the step (A'-l) comprises the steps of:

generating a low frequency noise signal within an octave band which has a center frequency at a specific frequency within a low frequency band below a cut-off frequency of the sound

converter of the multimedia device, analyzing a spectrum of the generated low frequency noise signal, and generating a low frequency harmonic component, which belongs to harmonic components of the low frequency noise signal and is included in the middle frequency band, using frequency components of the low frequency noise signal; and alternately reproducing a middle frequency reference sound at a fixed level, which is a low frequency noise signal sound within an octave band which has a center frequency at a specific frequency in the middle frequency band, that is, an audible range of the multimedia device, and the low frequency reference sound of the low frequency harmonic component at varying levels.

48. The method according to claim 47, wherein the middle frequency reference sound is a noise signal within an octave band sound which has a center frequency of 1 kHz.

49. The method according to claim 47, wherein the low frequency noise signal is a noise

signal within an octave band which has a center frequency of 60 Hz.

50. The method according to claim 43, wherein the step (A'-l) comprises the steps of: generating a high frequency noise signal within an octave band which has a center frequency at a specific frequency within a middle frequency band, that is, an audible range of the multimedia device, analyzing a spectrum thereof, and generating a high frequency harmonic component, which belongs to harmonic components of the high frequency noise signal and is included in a high frequency band above the audible range, using frequency components of the high frequency noise signal; and

alternately reproducing a middle frequency reference sound at a fixed level, which is a noise signal sound within an octave band which has a center frequency at a specific frequency within the

middle frequency band, that is, an audible range of the multimedia device, and the high frequency reference sound of the high frequency harmonic component at varying levels.

51. The method according to claim 50, wherein the middle frequency reference sound is a noise signal within an octave band sound which has a center frequency of 1 kHz.

52. The method according to claim 50, wherein the high frequency reference sound is a noise signal within an octave band sound which has a center frequency of 8 kHz.

53. A method of reproducing optimized three-dimensional sound in a multimedia apparatus for reproducing sound through a sound converter, the method comprising the steps of:

(A'-l) alternately reproducing a middle frequency reference sound at a fixed level and a low frequency reference sound at varying levels when the user selects a menu item for setting a level value of a low frequency sound; and

(A' -2) storing a level of a low frequency reference sound, which is currently being produced, as a low frequency level sound setting value in response to user input for storing the level value of the low frequency sound.

54. The method according to claim 53, wherein the step (A' -2) is the step of stopping alternate reference sound reproduction operation and storing the level of the low frequency reference sound, which is currently being reproduced, as the low frequency sound level setting value in response to user input for storing the level value of the low frequency sound.

55. The method according to claim 53, wherein, at the step (A'-l), the level of the reproduced low frequency reference sound increases or decreases automatically at regular intervals.

56. The method according to claim 53, wherein, at the step (A'-l), the level of the reproduced low frequency reference sound increases or decreases according to user input

57. The method according to claim 53, further comprising after the step (A' -2), when reproduction of a specific audio file is requested in the multimedia device, the steps of analyzing a spectrum of the audio file to be reproduced, generating the low frequency harmonic component, adjusting the level of the generated low frequency harmonic component according to the low frequency sound level setting value set by the user, and outputting the level-adjusted low frequency harmonic component

58. The method according to claim 53, wherein the step (A'-l) comprises the steps of: generating a low frequency noise signal within an octave band which has a center frequency at a specific frequency within a low frequency band below an audible range of the multimedia device, analyzing a spectrum thereof, and generating a harmonic component, which belongs to harmonic components of the low frequency noise signal and is included in the middle frequency band, using frequency components of the low frequency noise signal; and alternately reproducing a middle frequency reference sound at a fixed level, which is a noise

signal sound within an octave band which has a center frequency at a specific frequency within the middle frequency band, that is, an audible range of the multimedia device, and a low frequency

reference sound of the low frequency noise signal at varying levels.

59. The method according to claim 58, wherein the middle frequency reference sound is a

noise signal within an octave band sound which has a center frequency of 1 kHz.

60. The method according to claim 58, wherein the low frequency noise signal is a noise signal within an octave band which has a center frequency of 60 Hz.

61. A method of reproducing optimized three-dimensional sound in a multimedia apparatus for reproducing sound through a sound converter, the method comprising the steps of:

(A'-l) alternately reproducing a middle frequency reference sound at a fixed level and a high frequency reference sound at varying levels when the user selects a menu item for setting a level value of a low frequency sound; and (A' -2) storing a level of a high frequency reference sound, which is currently being produced, as a high frequency level sound setting value in response to user input for storing the level value of the high frequency sound.

62. The method according to claim 61, wherein the step (A' -2) is the step of stopping the alternate reference sound reproduction operation and storing the level of the high frequency reference sound, which is currently being reproduced, as the high frequency sound level setting value in response to user input for storing the level value of the high frequency sound.

63. The method according to claim 61, wherein, at the step (A'-l), the level of the reproduced high frequency reference sound increases or decreases automatically at regular intervals.

64. The method according to claim 61, wherein, at the step (A'-l), the level of the reproduced high frequency reference sound increases or decreases according to user input.

65. The method according to claim 61, further comprising after the step (A'-2), when reproduction of a specific audio file is requested in the multimedia device, the steps of analyzing a spectrum of the audio file to be reproduced and generating the high frequency harmonic component above a predetermined frequency, thus restoring lost high frequency components of the audio file, adjusting a level of the generated high frequency harmonic component according to a high frequency sound level setting value set by the user, and outputting the level-adjusted high frequency harmonic

component

66. The method according to claim 61, wherein the step (A'-l) comprises the steps of: generating a high frequency noise signal within an octave band which has a center frequency at a specific frequency within a middle frequency band, that is, an audible range of the multimedia device, analyzing a spectrum thereof, and generating a high frequency reference singal, which belongs to harmonic components of the high frequency noise signal and is a harmonic component within a high frequency band above the audible range, using frequency components of the high frequency noise signal; and alternately reproducing a middle frequency reference sound at a fixed level, which is a noise signal sound within an octave band which has a center frequency at a specific frequency within the middle frequency band, that is, an audible range of the multimedia device, and the high frequency reference sound of the high frequency harmonic component at varying levels.

67. The method according to claim 66, wherein the middle frequency reference sound is a noise signal within an octave band sound which has a center frequency of 1 kHz.

68. The method according to claim 66, wherein the high frequency reference sound is a noise signal within an octave band sound which has a center frequency of 8 kHz.