Patent
1994-11-21
1997-10-28
Safourek, Benedict V.
395 238, 395 239, G10L 900
Patent
active
056824610
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
The invention relates to a method of transmitting or storing digitalized, multi-channel audio signals. More particularly, one aspect of the invention is directed to a transmitting or storing method of the type in which, on the encoder side, (a) each audio channel is represented digitally by a number of spectral subband signals, wherein for each subband signal quantized sampling values that are discrete in time are present; (b) the quantization of the sampling values is changed (coded) in the individual subbands according to the channel-specific, intra-channel perceptual threshold of human hearing in the sense of a reduction in irrelevance; (c) for each subband signal a scale factor (SCF) is determined that classes the peak value of the subband signal within a certain time interval (block) and that is used in the normalization of the subband signal prior to quantization; and (d) for the control of the quantization of the sampling values of each subband signal, a piece of bit allocation information (BAL) is obtained from an audio signal analysis and is stored or transmitted together with the determined scale factors (SCF) and the coded subband signals, and in which, on the decoder side, (e) the coded subband signals are decoded according to the bit allocation information (BAL) and the subband signal levels are denormalized according to the scale factors (SCF); and (f) the decoded and denormalized subband signals of each audio channel are combined into one broadband, digital audio signal.
For transmitting and storing digital audio signals, it is known (DE 34 40 613 and DE 36 39 753) to perform a data reduction with the aid of a coding in subbands, and with the use of a psychoacoustic model. In these cases the digitalized, broadband audio signal is represented by a number of spectral subband signals, and the quantization of the sampling values is effected in the individual subbands according to the perceptual threshold of human hearing. The data reduction factor for very high-value audio signals that can be achieved with this "monophonic coding" has approximately the value of 6 at a 46 kHz sampling frequency and 16-bit linear quantization, which corresponds to a data rate of 2.times.128 kbit/s per stereophonic audio signal.
Also known are measures for increasing the data reduction factor during coding of dual-channel, stereophonic signals, which are based on suppressing redundant signal components in the left and right channel, as well as such components which are irrelevant with regard to the stereophonic effect. Two different embodiments of this "stereophonic coding" are described in Netherland Patent Application No. 90 00 338 and in "Proceedings of the ICASP 1990" Perceptual Transform Coding of Wideband Stereo Signals,"namely range, and block-wise reconstruction of the subband signal level in the left and right channels with the aid of scale factors which represent the maximum signal level in the subbands of the left and right audio channels, and (L) and right (R) audio signal in accordance with the matrixing M=L+R and S=L-R, where the decoding of the sum signals and differential signals is effected according to the perceptual threshold determined for the sum signal M.
If the separate coding in the subbands of the left and right audio channels is supplemented by stereophonic coding in such a way that one of these methods or a combination of the two methods is employed, the data reduction factor of the value 6 can be increased to approximately the value 8. If, for example, a bit rate of 256 kbit/s is required for the transmission of two independent monosignals of the same quality as a 16-bit linear quantization, during stereophonic coding only 192 kbit/s are required for the transmission of a dual-channel stereo signal of the same subjective quality.
SUMMARY OF THE INVENTION
In contrast, the object of the invention is to achieve an even more substantial data reduction in dual-channel or multi-channel stereophonic signals.
This object is accomplished, according to one aspect of
REFERENCES:
patent: 4790016 (1988-12-01), Mazor
patent: 4972484 (1990-11-01), Theile et al.
patent: 5230038 (1993-07-01), Fielder et al.
patent: 5235671 (1993-08-01), Mazor
patent: 5367608 (1994-11-01), Veldhuis et al.
patent: 5479562 (1995-12-01), Fielder et al.
patent: 5539829 (1996-07-01), Lokhoff et al.
ICASSP 91--IEEE International Conference on Acoustics, Speech, and Signal Processing, Bd. 5, 14 Mai 1991, Toronto, CA, Seiten 3601-3604, "Subband Coding of Sterophonic Digital Audio Signals" R.G. Van Der Wall et al.
ICASSP 92--IEEE International Conference on Acoustics, Speech, and Signal Processing, Bd. 2, 23 Marz 1992, San Francisco, US Seiten 569-572 "Sum-Difference Stereo Transform Coding" J.D. Johnston et al.
ICASSP 89--IEEE International Conference on Acoustics, Speech and Signal Processing, Bd. 3, 23. Mai 1989, Glasgow, GB, "Perceptual Transform Coding of Wideband Stereo Signals" J.D. Johnston.
Link Martin
Silzle Andreas
Stoll Gerhard
Theile Guenther
Institut Fuer Rundfunktechnik GmbH
Ngo Ricky
Safourek Benedict V.
LandOfFree
Method of transmitting or storing digitalized, multi-channel aud does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of transmitting or storing digitalized, multi-channel aud, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of transmitting or storing digitalized, multi-channel aud will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1031306