вход по аккаунту



код для вставкиСкачать
Patent Translate
Powered by EPO and Google
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
March, 1972 Tar-day Patent Office Secretary Ito Takehisado #-Gey Oshi de 4 Shi y I 9 Dishitsu
Fist 豐 Shiki 1, multidimensional speech signal transmission method% formula% (5) (name of the
invention multidimensional Audio signal transmission method Human voice signals in a plurality
of directions corresponding to sound sources in the original sound field are converted into a
transmission signal TL and a second transmission signal TR of a brunt 1 and recorded, and also
in an appropriate band outside the frequency band of the transmission signal. 1st porting-1! In
the method of converting into cL and the carrier signal C8 of the crest 2 and recording them, the
front side input audio signal is distributed to the image transmission signals TL and T8 and the
justification transmission numbers CL and CR1 in the left and right in phase to the reference axis,
Rear side input audio signals are distributed to the image transmission signals TL and TFL and
the near-field transmission numbers CL and CR in the axis that intersects the reference axis in
right and left reverse phases, and both forward transmission signals T and T Multi-dimensional,
characterized in that the distribution ratio of the front side input signal to the rear side input
signal and the distribution ratio of the front side input signal to the rear side input signal to C R
and the rear side input signal to C L are equal. Voice signal transmission method.
DETAILED DESCRIPTION OF THE INVENTION The present invention relates to multi-dimensional
audio signal-power equations, and in particular to a novel transmission scheme compatible with
matrix transform and independent transmission schemes. Conventionally, many systems have
been proposed for multi-dimensional transmission of audio signals, but broadly classified, there
are matrix systems that transmit matrix signals of multi-dimensional signals for transmission, and
discrete systems by independent transmission. A code recording board and playback equipment
are commercially available. However, since there is no compatibility between the two methods,
the consumer needs a reproduction device suitable for each type of program source! 1−91!
It must be used. It is a factor that impedes the spread of so-called four-channel stereo. In the
novel multidimensional audio signal transmission system compatible with both the abovementioned discrete system and matrix system proposed to eliminate such an inconvenience, the i
system of the present invention is a matrix system according to the discrete system decoder.
Regardless of the method decoder, the present invention is to provide a method of obtaining the
same sound-localization from the acoustics to the EndPage: 1-on-code method. In order to help
the understanding of the present invention, the discrete method and the regular matrix method
(hereinafter abbreviated as RM method) will be briefly described. In the RM method, at the time
of encoding, the signal corresponding to the sound source on the front side in the original sound
field is distributed in the same phase to the transmission signal TL composed mainly by the left
signal, the transmission signal TL to be struck and the transmission signal mainly composed of
the right signal Also, the signal corresponding to the sound source on the rear side in the original
sound field is distributed to TL at a 90 ° advanced phase with respect to the forward signal and
to the TR with a 90 ° lagging phase. And in the reproduction sound field, the signal to be
localized to 7i, @ is more TL than TR, and the signal to be localized to the right is more
distributed to TFL than TL. At the time of decoding, the output signal supplied to the speaker
located in front of the reproduction f field is TL. The output signal supplied to the speaker which
is blended with the TR and the same phase and located on the rear side of the reproduction
sound field is TL. The basic configuration is that the phase of TR is blended with a phase delay of
9060 and a phase advance relative to the blending of the forward output signal. A configuration
in which the phase shift is not 90 ° is also included in the RM method, but in this case, TL and
TR are in opposite phase. In the case of the KM encoder, one example of the configuration of the
RM decoder is as shown by equations 111 and 12). Here, m: distribution ratio (0 (m (1) LB: a
signal LF corresponding to a sound source located at the left rear in the original sound field: a
signal corresponding to a sound source at the left front in the original sound field RIF = right
front in the original sound field Signal RB corresponding to a certain sound source = signal LB ′
corresponding to a sound source located on the right rear in the original sound field: output
signal LF of the decoder applied to a speaker placed on the left rear in the reproduced sound
field: left front in the reproduced fm The output signal RF of the decoder applied to the
loudspeaker located at
= Decoder output signal applied to the speaker located on the right front in the reproduction f
field RB ': output 91 of the decoder applied to the speaker located on the right rear in the
playback sound field. Also, + j, -j show that the phase is +1 for leading and lagging for 90 '. Since
constructing an actual circuit from the matrix display of the encoder / decoder is easy for a
carrier, it is omitted to illustrate the construction of uius. The regular matrix type 1 曾 matrix can
be obtained by substituting the equation 111 into the equation (2) and using m = f “−1 −13”
as the normally used value and arranging as follows. The +1) equation is usually stored as a
vector as shown in FIG. In the CD-4 system, which is a discrete system, as shown in FIG. 2, a 4-
channel signal as shown in FIG. 2 is a left track, a low band (30-15000 Hz) for the TL, and a CL =
l for the high band (20 to 45 KHz). , F-LB is TR = FLF in the low band, + RB is in the low band, and
CR = CR is the carrier f in the high band. Since the signal is modulated, it must be demodulated
prior to decoding. In the description of the present invention, since the modulation /
demodulation of the carrier EndPage: 2 has no essential importance in the description of the
present invention, the description 1 @ of its procedure 1 @ is omitted hereinafter. Encoding and
decoding of the CD-47j equation perform the following matrix operation. It becomes a diagonal
matrix like substituting the equation (5) into the equation (6) and determining the bonding
matrix, and it is apparent that the CD-40,000 equation is discrete. Comparing the above KM
method and CD-4 method encoding / decoding operations 11) to (7), the KM method includes
one term and performs 90 ° phase shift, but! ! It can be seen that the addition and subtraction
of the & signal component are the only ones. The applicant paid attention to this point and
proposed a new multidimensional voice transmission scheme adapted to both the KM scheme
and the CD-4 scheme. This system is abbreviated as RMC system and its structure is shown in
FIG. 3. The two transmission signals TI which are identical to the conventional aM system and the
TRt voice frequency band are combined and the other two carrier signals CLsP are arranged. And
CR outside the voice frequency band, and CL and CB are different from TL and Ta in the
distribution configuration of the input voice signal. The description of the carrier conversion and
recovery procedures will be omitted because they are not essentially electric in the present
invention. Matrix display of an example of the RMC encoding configuration is as follows.
The RMC type TI and T8 are identical to the RM type-and TR as is apparent from the comparison
of the 11) and t81 types and the 1st and 3rd charts, and the RMs that were encoded in the KMC
mode are RM When decoding by the system, it is possible to reproduce by the same machine as
the one that is dangerous by encoding by the KM system. A transmission signal lCD-4 decoder
encoded by the KMC method is added as (61 弐 K (substituting the equation 81) as shown in FIG.
As can be seen from this, the output signal has crosstalk between the front side signal and the
rear side signal, and the t side rear signal contains j and is in reverse phase, and the rear sound is
not localized in the reproduction sound field. It is necessary to correct this. . If the correction
matrix is displayed in matrix, the RMC signal is added to the CD-4 decoder and the superior
signal of equation (7) is applied to the matrix shown by equation 181, the output signal of the
correction decoder becomes EndPage: 3 as in □ As a diagonal matrix, a complete discrete
reproduction signal is obtained. The correction matrix can use a part of the R and M decoders
incorporated in the RM type and CD-4 type shared regenerator. A CD-4 correction decoder which
combines a CD-4 decoder and a correction matrix can also be easily derived from the above
equation as follows! It is displayed tricks. It is easier for the contractor to actually construct the
circuit than this matrix. Although the theory dealt with mathematically applies only to the
conversion of the signal at the time of encoding / decoding in the ELM method, mathematical
theory is well applied, but when forming the sound field by the reproduction signal, the left and
right before for the directional characteristics of the hearing It is known that f 儂 should be
positioned inward 25; The compensation of the deviation is performed at the time of recording,
and one straight line of the distribution ratio m · is often selected to be, for example, 0.3 @
degrees for the forward signal and about 0.5 for the backward signal. Since the RMC signal is
adapted to the RM decoder, the two transmission signals TL and TR perform the abovementioned compensation, and the RM decoder obtains a signal for correct sound image
localization. As shown in the above-mentioned C121, if the signal obtained by performing sound
image localization compensation on the two transmission signals TL and TR is reproduced by the
CD-4 correction decoder shown in the above-mentioned C121, the forward signal is displaced
outward and the reproduced sound field is correct. Is not formed. The present invention
eliminates such a drawback, and in order to obtain a psychoacoustically correct reproduction
tone −t in both RM regeneration and discrete regeneration, it is carried out with respect to two
transmission signals TI and SI at the time of encoding. This compensation and the reverse
compensation may be applied to each component of the two carrier signals CL and cH.
Select a distribution ratio m for the forward signal, for example 0.5, to that of the st & square
signal, for example 0.3, If this is shown by a matrix, it becomes as follows. The total matrix in
combination with the CD-4 RMC correction decoder described above is obtained, and the
diagonal crosstalk is 18.5 dB, which does not cause a real black ground problem. Thus, the
forward signal at the time of encoding of the present invention. By overcompensating the
allocation ratio for the backward signal too much, the same psychoacoustically correct sound
image localization can be obtained for both the embedded system and the RM re-reproduction
discrete reproduction, and no consideration is given to the reproduction system. In fact, it has the
advantage of fire.
BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a conventional 1M encoder
for explaining the present invention, FIG. 2 is a diagram showing a CD-4 engineering code
system, and FIG. 3 is a diagram according to the present invention. The diagram showing the
encoding method, FIG. 4 is a factor showing the output signal when added to the encode signal
tcD-4 decoder according to the present invention as a vector. In the figure, LB: left rear input
signal, LF: left front input signal, BIFI: right front input signal-RB: right rear input signal LINB: left
rear output signal, 'LF': left front output signal,顯 I: Right front output signal it, / / @ Right rear
output signal TL: Left @ 蓬 signal, TB ,: Right transmission signal, CL: Left End Page: 4 carrier
signal, CB: Right carrier signal. Patent Applicant Nippon Columbia Co., Ltd. Agent 9 Fujitsu
EndPage: 5
Без категории
Размер файла
14 Кб
Пожаловаться на содержимое документа