JPH09224300A - Method and device for correcting sound image position - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the feeling of incompatibility or unnaturality caused by the non-coincidence of video and sound image positions by correcting the sound image position of an audio signal to be reproduced by performing the correcting processing of frequency characteristics to frequency data provided by extending processing. SOLUTION: Based on the frequency data outputted from an extender 16, frequency data extracting parts 31-35 at a correction part 17 extract frequency data SFL, SFR, SCT, SRL and SRR concerning respective front left/right, center and rear left/right channels. Correction processing parts 41-45 perform the correcting processing by adding correction data DFL, DFR, DCT, DRL and DRR read out of a memory 52 to the respective data SFL, SFR, SCT, SRL and SRR. As a result, the vertical deviation of the sound image position caused by no arrangement of the speaker at an ideal position is corrected, and unnaturality caused by the non-coincidence of video and sound image positions is reduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method in which an audio signal is converted from a time domain into a frequency domain and then compressed, and then compressed data is decompressed to obtain frequency data. The present invention relates to a method and apparatus for correcting a sound image position in an audio reproduction system configured to perform inverse conversion into audio and reproduce an audio signal.

[0002]

2. Description of the Related Art In recent years, in the audio-visual world, by devising a larger TV screen, higher definition, and adopting a surround system for audio, various efforts have been made to create a richer sense of presence and power. .
However, in a general home, the room in which the AV system is actually installed is not in an ideal state, and the placement position of the speaker is limited by windows and furniture. Therefore, A
Fill in the difference between the planned position of the sound source of the V system (sound reproduction system) and the position of the speaker that is actually placed,
In order to maximize the acoustic effect, it is necessary to correct the sound image position in the AV system.

For example, AC-3, which is a multi-channel compression / decompression method developed by Dolby, is used for movies and LDs.
(Laser disc), DVD (digital video disc), etc. surround system,
Front left, front right, center, rear left, rear right,
It is possible to reproduce the signals of a total of 6 channels of the subwoofer by the speakers arranged at the respective positions. Generally, since the line connecting the listener and the video screen is horizontal, these speakers are preferably arranged on the same horizontal plane as the listening position of the listener, that is, the position of the ear. However, depending on the arrangement of windows and furniture and the convenience of the room, the speaker may be mounted on the wall surface, that is, near the ceiling.
In particular, the center speaker that overlaps with the position of the television receiver or the display device that displays an image and the rear speaker for obtaining the surround effect often have to be installed near the ceiling. Therefore, a sense of discomfort occurs due to the mismatch between the image position and the sound image position. For example, in a video, the sound source is in the front, but in a extreme sense, the sound comes from above the head, which is uncomfortable.

Further, the audible frequency characteristic changes depending on the incident angle of the sound to the ear and is influenced by the shape and condition of the head such as auricle, face and hair. Therefore, the sound from the speaker installed on the ceiling becomes unnatural sound different from the sound originally intended by the AV system.

[0005]

As described above, in the conventional audio reproduction system, the sense of discomfort due to the inconsistency between the image position and the sound image position, or the unnatural feeling due to the change of the frequency characteristic due to the influence of the head shape. There was a problem that is caused.

In order to solve this problem, a method of correcting the frequency characteristic and the phase characteristic by using an FIR filter or the like and correcting the sound image position can be considered. However, according to this method, the processing amount of the convolution calculation for controlling the sound image position becomes enormous, so that an extremely high-performance arithmetic processing device is required to increase the processing speed. Therefore, the cost of the audio reproduction system increases significantly.

The present invention has been made in view of the above problems, and a sound image position capable of reducing a sense of discomfort and unnaturalness due to a mismatch between a video image and a sound image position without increasing the cost of the audio reproducing system. It is an object of the present invention to provide a correction method and device.

[0008]

According to a first aspect of the present invention, a method of transforming an audio signal from a time domain into a frequency domain and performing compression processing on the compressed data is expanded to obtain frequency data. A method of correcting a sound image position in an audio reproduction system configured to reproduce the audio signal by inversely converting the frequency domain into the time domain after obtaining the frequency characteristic with respect to the frequency data obtained by the expansion processing. The sound image position of the reproduced audio signal is corrected by performing the correction process of.

According to a second aspect of the present invention, the apparatus converts the audio signal from the time domain to the frequency domain and compresses the compressed data. Then, the compressed data is decompressed to obtain the frequency data. A sound image position correction device in an audio reproduction system configured to perform inverse conversion into a time domain to reproduce an audio signal, wherein the reproduced audio signal of the audio signal is reproduced with respect to the frequency data obtained by the expansion processing. It has a frequency characteristic correction means for correcting the frequency characteristic in order to correct the sound image position.

According to a third aspect of the present invention, the frequency characteristic correcting means uses a specific position on a vertical plane including the listening position on the same horizontal plane as the listening position by the listener, and A position on the same vertical plane having various elevation angles with respect to the listening position is defined as a displacement position, and a difference in the displacement position with respect to the reference position with respect to a spatial frequency characteristic from a sound source to the listener is corrected as a frequency characteristic. A memory stored as data, an input unit for inputting elevation angle information of a sound source for emitting the audio signal toward the listener, and the frequency from the memory based on the elevation angle information input by the input unit. It has a calculation means for reading the characteristic correction data and calculating the read frequency characteristic correction data and the frequency data.

In the device according to the fourth aspect of the present invention, the spatial frequency characteristic is a head-related transfer characteristic which is a combination of the auricle characteristic of the listener and the head spatial characteristic of the listener. To explain the principle of the present invention, as shown in FIG. 4, the acoustic spatial frequency characteristic (spatial transfer characteristic) from the speaker b arranged at the ideal position at the same height position as the listening position to the listening position is Y. [B] and [f], and the acoustic spatial frequency characteristic from the speaker a arranged at the actual arrangement position having the elevation angle α above the ideal position to the listening position is X [a].
[F] is assumed.

The generated sound from the speaker b at the ideal position is P
If it is [f], the arrival sound Zb to the listening position (ear)
Is Zb = P [f] + X [b] [f].

However, for the speaker a at the actual arrangement position, even if the generated sound is the same P [f], the arrival sound Za at the listening position (ear) is Za = P [f] + X [a] [ f] In order to make the reaching sound Za equal to the reaching sound Zb, the difference between them may be added to the sound generated by the speaker a. That is, difference = P [f] + X [b] [f] −P [f] + X [a] [f] = X [b] [f] −X [a] [f] This difference is the elevation angle α. This is the correction amount for the speaker a. Therefore, with respect to each speaker 21, various correction data for each elevation angle α are measured in advance by using the pseudo head or the pseudo ear. By using a pseudo-head or the like, correction data for a standard listener can be obtained. The obtained correction data is added or multiplied to the frequency data obtained by the expansion processing. Thereby, the correction process is performed.

In the present invention, the horizontal plane is not limited to a horizontal plane about the gravity field of the earth, but also a horizontal plane in the sound field of sound reproduction, such as a listening position and a video position, or a plane including left and right front speakers. It is a concept that includes. The elevation angle includes the depression angle.

[0015]

1 is a block diagram showing the configuration of the audio portion of an AV system 1 using an audio reproduction system 4 according to the present invention.

The AV system 1 includes an audio recording system 2,
The recording medium 3 and the audio reproduction system 4 are included.
The voice recording system 2 includes an amplifier 12 for amplifying a voice signal S1 from a sound source 11, an MDCT (modified discrete cosine transformer) 13 for converting a time domain voice signal S2 into frequency data S3 which is frequency domain spectrum data, and data. It is composed of a compressor 14 which performs compression processing to reduce the amount and outputs compressed data S4. Recording on the recording medium 3 is performed based on the compressed data S4.

As the recording medium 3, for example, a movie film, a laser disc, a digital video disc, a mini disc, a digital compact cassette tape or the like is used. The compressed data S4 is recorded on the recording medium 3 as a digital signal, and the same digital signal as the compressed data S4 is read out as reproduced data S5 during reproduction.

The audio reproduction system 4 outputs from the expander 16 and the expander 16 that perform expansion processing on the reproduction data S5 read from the recording medium 3 and output the same frequency data S6 as the original frequency data S3. Frequency data S6
Correction unit 17 for performing a correction process on the IMD for inversely converting the corrected frequency data S7 into time domain data S8
CT (Inverse Modified Discrete Cosine Transform) 18, post-processing unit 1 for applying various post-processing to the inversely transformed data S8
9, an amplifier 20 for amplifying the audio signal S9 output from the post-processing unit 19, and a speaker 21 for converting the audio signal output S10 output from the amplifier 20 into sound.

In the compressor 14, compression processing by various compression algorithms is performed. It is sometimes called coding. For this processing, for example, PASC and MD used in DDC (Digital Compact Cassette)
There are ATRAC used in (Mini Disc) and AC-3 used in the surround system of Dolby. Since these compression processes are performed in the frequency domain in order to improve the efficiency of compression, conversion to the frequency domain is performed before the compression process.

The decompressor 16 performs a decompression process that is the reverse of the compression performed by the compressor 14, and decompresses the reproduced data S5 (compressed data S4). Sometimes called decoding. The frequency data S6 output from the decompressor 16 is frequency domain data. The frequency data S6 is subjected to arithmetic processing by the correction unit 17 using the correction data having the frequency characteristic for correction, and the frequency characteristic is corrected. The correction of the frequency characteristic corrects the frequency characteristic of the sound emitted from the speaker 21 that changes depending on the incident angle of the sound on the ear, and the sound image position according to the actual installation position of the speaker 21.

The post-processing section 19 performs, for example, conversion processing of the number of channels and processing of adding various acoustic effects. FFT instead of MDCT and IMDCT
(Fast Fourier transform) and IFFT may be used. In the AV system 1, each unit is provided in the required number according to the number of channels.

FIG. 2 is a block diagram showing the configuration of the correction unit 17. The correction unit 17 includes frequency data extraction units 31 to 31.
5, correction processing units 41 to 45, an input unit 50, an address designation unit 51, and a correction data memory 52.

Based on the frequency data S6 output from the decompressor 16, the frequency data extraction units 31 to 35 generate frequency data SFL and SF for the front left, front right, center, rear left, and rear right channels, respectively.
Extract R, SCT, SRL, SRR. However, when the decompressor 16 outputs the frequency data of each channel in parallel, the frequency data extraction units 31 to 35 are unnecessary.

The correction processing units 41 to 45 correct the respective frequency data SFL, SFR, SCT, SRL, SRR using the correction data DFL, DFR, DCT, DRL, DRR read from the correction data memory 52. Perform processing. The contents of the correction process will be described later.

The correction data memory 52 uses a specific position on the vertical plane including the listening position on the same horizontal plane as the listening position by the listener as a reference position and has various elevation angles with respect to the listening position and is the same. This is a ROM in which a position on the vertical plane is defined as a displacement position, and the difference between the displacement position and the reference position regarding the spatial frequency characteristic from the sound source to the listener is stored as frequency characteristic correction data. In other words, in the correction data memory 52, the correction data DFL, DFR, for the various elevation angles α at which the speakers are arranged for each of the front left, front right, center, rear left, and rear right channels.
DCT, DRL, DRR are stored.

The addressing unit 51, based on the elevation angle α input from the input unit 50 corresponding to the actual arrangement position of the speaker 21, the correction data DFL, DF corresponding to the elevation angle α.
Addressing is performed so that R, DCT, DRL, and DRR are read from the correction data memory 52.

FIG. 3 is a diagram schematically showing an ideal position and an actual placement position of the speaker 21, FIG. 4 is a diagram for explaining the correction process, FIG. 5 is a diagram showing head-related transfer characteristics, and FIG. 6 is correction data. It is a figure which shows an example of DCT, DRL, and DRR.

As shown by a solid line and a chain line in FIG. 3, the speaker 21CT at the center is on the back surface of the screen SCR where the image is displayed, the front left and right speakers 21FL and 21FR are on both sides of the screen SCR, and at the rear. Ideally, the left and rear right speakers 21RL and 21RR are respectively arranged at the same height positions as the screen SCR on the left and right wall surfaces.

However, in practice, as shown by the solid line in FIG. 3, the center speaker 21CT is located above the screen SCR and the rear left and right speakers 21RL and 21R are arranged.
The Rs are often arranged at positions close to the ceilings of the left and right wall surfaces. Also in this embodiment, it is arranged as such. Therefore, it is necessary to perform correction processing in the correction unit 17 for each of the center, rear left, and rear right channels.

In FIG. 4, the acoustic spatial frequency characteristic (spatial transfer characteristic) from the speaker b arranged at the ideal position at the same height as the listening position to the listening position is Y [b].
It is [f], and it is assumed that the acoustic spatial frequency characteristic from the speaker a arranged at the actual arrangement position having the elevation angle α right above the ideal position to the listening position is X [a] [f].

The unit of the spatial frequency characteristic, the generated sound, the arrival sound, etc. is dB. That is, if Y is the amplitude ratio,
X [a] [f] = 20 · log 10 (Y [a] [f]).

The generated sound from the speaker b at the ideal position is P
If it is [f], the arrival sound Zb to the listening position (ear)
Is Zb = P [f] + X [b] [f].

However, for the speaker a at the actual arrangement position, even if the generated sound is the same P [f], the arrival sound Za at the listening position (ear) is Za = P [f] + X [a] [ f] In order to make the reaching sound Za equal to the reaching sound Zb, the difference between them may be added to the sound generated by the speaker a. That is, difference = P [f] + X [b] [f] −P [f] + X [a] [f] = X [b] [f] −X [a] [f] This difference is the elevation angle α. This is the correction amount for the speaker a. That is, correction frequency characteristic = X [ideal position direction] [frequency] (correction data) -X [actual arrangement position direction] [frequency].

Here, the actual arrangement position direction is equal to the elevation angle α which is the incident angle of sound. That is, the correction data (correction frequency characteristic) is the difference between the spatial frequency characteristics at the listening position between the speaker b arranged at the ideal position and the speaker a arranged at the actual arrangement position. Therefore, each speaker 21
With respect to, the correction data for every 10 ° in various elevation angles α, for example, in the range of −30 ° to 60 ° are measured in advance using the pseudo head or the pseudo ear.

As shown in FIG. 5, the spatial frequency characteristic is the sum of the spatial characteristic determined according to the situation in the room and the head-related transfer characteristic determined according to the shape of the head. The head-related transfer characteristic is the sum of the head space characteristic determined according to the shape of the face and the state of the hair, and the auricle characteristic determined according to the shape of the auricle.

Therefore, the correction data for a standard listener can be obtained by using the pseudo head or the like.
Regarding the distance from the listening position to the speaker, if the ideal position and the actual arrangement position have the same relative relationship, the change in the distance between the listening position and the speaker has little effect on the correction data. It suffices to measure the correction data for different distances and use it.

As shown in FIG. 6, the correction data DCT, D
RL and DRR are functions of frequency, and the correction amount increases when the frequency exceeds 1 KHz. Correction data DCT for the center channel and correction data DRL for the rear channel,
It is different from DRR.

Measured correction data DFL, DFR, DC
T, DRL, and DRR are stored in the correction data memory 52 as a data table. Elevation angle α of speaker 21
Is input by a listener from a ten-key pad or the like which is the input unit 50, or is detected and input by a sensor attached to the speaker 21. The correction data memory 52 for each channel corresponding to the input elevation angle α.
The correction data is read from. If there is no correction data that matches the input elevation angle α, the correction data of the elevation angle α that is closest to the input elevation angle α is read.

The correction data read out is the frequency data SFL, SFR, SCT, SRL,
It is added to SRR. As a result, the difference in the frequency characteristics on the auditory sense caused by the fact that the actual arrangement position of each speaker 21 is not the ideal position is corrected, and the listener's discomfort is reduced. Further, the vertical shift of the sound image position due to the speaker 21 not being arranged at the ideal position is corrected, and unnaturalness due to the mismatch between the image position and the sound image position is reduced.

Moreover, the correction process in the correction unit 17 is as follows.
Since the correction data read from the correction data memory 52 is added to the frequency data SFL, SFR, SCT, SRL, SRR, the content of the arithmetic processing for correction is simple and a high-performance arithmetic device is not used. Both are processed in a short time. In a system that compresses / decompresses audio data, processing in the frequency domain is performed in order to improve compression efficiency, and therefore frequency data is always generated in the process, so use generated frequency data as is. Therefore, it is not necessary to newly generate frequency data.

Therefore, in the sound reproduction system 4, the DSP used for the processing other than the correction processing.
The correction process can be performed by, for example, and even in that case, the correction process can be performed at high speed without affecting other processes. Therefore, it is possible to provide a low-cost audio reproduction system without increasing the cost of the audio reproduction system 4.

In the above embodiment, the unit of the correction data and the like is dB, but when the data is represented by the amplitude ratio, the correction may be made by multiplying the frequency data and the correction data. Audio playback system 4,
The configuration, processing content, processing order, and the like of the AV system 1 can be appropriately changed in accordance with the gist of the present invention.

[0043]

According to the first to fourth aspects of the present invention, it is possible to reduce the uncomfortable feeling and the unnaturalness due to the mismatch between the image and the sound image position without increasing the cost of the audio reproducing system.

For example, even when the speaker is installed on the ceiling, the sound can be heard with the same frequency characteristic as when the speaker is installed on the same horizontal plane as the listener's head. Also,
The sound image direction approaches the same horizontal plane as the head, and unnaturalness due to the disagreement between the image direction and the sound image direction is reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an audio part of an AV system using an audio reproduction system according to the present invention.

FIG. 2 is a block diagram showing a configuration of a correction unit.

FIG. 3 is a diagram schematically showing an ideal position and an actual arrangement position of speakers.

FIG. 4 is a diagram for explaining a correction process.

FIG. 5 is a diagram showing a head-related transfer characteristic.

FIG. 6 is a diagram showing an example of correction data.

[Explanation of symbols]

 4 audio reproduction system 17 correction unit (correction device, frequency characteristic correction unit) 41 to 45 correction processing unit (calculation unit) 50 input unit (input unit) 52 correction data memory (memory)

Claims (4)

[Claims] 1. An audio signal is obtained by converting an audio signal from a time domain into a frequency domain and performing compression processing, decompressing the compressed data to obtain frequency data, and then inversely converting the frequency domain into a time domain. A method of correcting a sound image position in an audio reproduction system configured to reproduce a signal, wherein the audio signal reproduced by performing frequency characteristic correction processing on the frequency data obtained by the expansion processing. A method for correcting a sound image position, characterized in that the sound image position is corrected. 2. A speech signal is obtained by converting an audio signal from a time domain to a frequency domain and performing compression processing, decompressing the compressed data to obtain frequency data, and then inversely converting the frequency domain into a time domain. A sound image position correcting device in an audio reproducing system configured to reproduce a signal, for correcting the sound image position of the reproduced audio signal with respect to the frequency data obtained by the expansion processing. A sound image position correcting device, comprising a frequency characteristic correcting means for performing frequency characteristic correction processing. 3. The frequency characteristic correcting means uses a specific position on a vertical plane including the listening position on the same horizontal plane as the listening position by the listener as a reference position, and various elevation angles with respect to the listening position. A position on the same vertical plane having a displacement position, a memory that stores a difference of the displacement position with respect to the reference position with respect to the spatial frequency characteristic from the sound source to the listener as frequency characteristic correction data, An input unit for inputting elevation angle information of a sound source for emitting a voice signal to the listener, and the frequency characteristic correction data is read out from the memory based on the elevation angle information input by the input unit and read out. The sound image position correcting device according to claim 2, further comprising a calculating unit that calculates the frequency characteristic correction data and the frequency data. 4. The sound image position correcting device according to claim 3, wherein the spatial frequency characteristic is a head-related characteristic that is a combination of the auricle characteristic of the listener and the head spatial characteristic of the listener. .