Apparatus and method for stereo audio encoding of digital audio signal data

• US 5,636,324 A
• Filed: 03/31/1995
• Issued: 06/03/1997
• Est. Priority Date: 03/30/1992
• Status: Expired due to Fees

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Teh, Do-Hui, Tan, Ah-Peng
• Primary Examiner(s): MacDonald, Allen R.
• Assistant Examiner(s): Sartori, Michael A.
• Application Number:
  US08/415,159
• Time in Patent Office: 795 Days
• Field of Search: 395/2.1, 395/2.15, 395/2.2, 395/2.35-2.39, 381/1, 381/2, 381/17, 381/29, 381/30, 381/34, 381/36, 381/94
• US Class Current: 704/226
• CPC Class Codes: H04B 1/665 using psychoacoustic properties of the ear, e.g. masking effect
• CPC Class Codes: H04H 20/88 Stereophonic broadcast systems multiplex system in general H04J
• CPC Class Codes: H04S 1/007 in which the audio signals are in digital form data reduction aspects thereof based on psychoacoustics G10L19/02

Abstract

A stereo audio encoding method for encoding left and right original signals to a left and right reproduced signals for suppressing a loss of quality in the reproduced audio signal. The correlation between the right and left channel signals is determined, and the phase of each signal is compared. If the two signals have the same phase, a modified scale factor is calculated based on a power equalization method, but if the two signals are in opposite phase, another modified scale factor is calculated based on an error minimization method. The modified scale factors are used for calculating the reproduced signals.

Claims

1. A stereo audio encoding method for encoding audio signals to left and right reproduced signals, comprising the steps of:

(a) calculating a correlation between left and right original signals to detect a pair of the left and right original signals having opposite phase characteristics and a pair of the left and right original signals having same phase characteristics, where the audio signals have been transformed to the left and right original signals each defined by a train of frames where each frame contains a plurality of sampled data;

(b) processing the left and right original signals having the opposite phase characteristics according to a power equalization method to obtain said left and right reproduced signals; and

(c) processing the left and right original signals having the same phase characteristics according to an error minimization method to obtain said left and right reproduced signals.

3. A stereo encoding method as claimed in claim 1, wherein said processing step (b) comprises the steps of:

(b1) calculating an average signal which is an average between said left and right original signals;

(b2) respectively selecting a left maximum absolute sampled data and a right maximum absolute sampled data in each frame for each of said left and right original signals;

(b3) calculating (1) a left intervening signal by multiplying said average signal by said left maximum absolute sampled data and (2) a right intervening signal by multiplying said average signal by said right maximum absolute sampled data;

(b4) respectively calculating a left intervening power, a right intervening power, a left original power, and a right original power for each of said left and right intervening signals, and said left and right original signals;

(b5) calculating said left reproduced signal by multiplying said left intervening signal by (1) said left maximum absolute sampled data and (2) a square root of a ratio of the left original power to the left intervening power; and

(b6) calculating said right reproduced signal by multiplying said right intervening signal by (1) said right maximum absolute sampled data and (2) a square root of a ratio of the right original power to the right intervening power.

4. A stereo audio encoding method as claimed in claim 1, wherein said processing step (c) comprises the steps of:

adding said left and right original signals; and

dividing the added signal into left and right reproduced signals relative to a ratio between (1) a power of the left original signal to (2) a power of the right original signal.

5. A stereo audio encoding method as claimed in claim 1, wherein said processing step (c) comprises the steps of:

(c1) respectively calculating a left power and a right power for each of said left and right original signals;

(c2) calculating an interaction term by multiplying each of said plurality of sampled data in said left original signal with corresponding data of said plurality of sampled data in said right original signal;

(c3) calculating an average signal which is an average between said left and right original signals;

(c4) selecting a maximum absolute sampled data in the average signal;

(c5) calculating a first ratio of (1) a product of said maximum absolute sampled data multiplied by a sum of said left power and said interaction term to (2) a sum of said right power and said left power and said interaction term;

(c6) calculating a second ratio of (1) a product of said maximum absolute sampled data multiplied by a sum of said right power and said interaction term to (2) a sum of said right power and said left power and said interaction term;

(c7) calculating said left reproduced signal by multiplying said average signal by said first ratio; and

(c8) calculating said right reproduced signal by multiplying said average signal by said second ratio.

6. A stereo encoding method as claimed in claim 1, wherein said processing step (b) comprises the steps of:

(b1) respectively calculating a left original power, a right original power, a left intervening power, and a right intervening power for each of said left and right original signals, and each of left and right intervening signals wherein said left and right intervening signals are obtained by multiplication of average of the said left and right original signal with a maximum sampled data of said left and right original signals; and

(b2) calculating said left reproduced signal by multiplying (1) said left intervening signal by a left maximum absolute sampled data for the left original signal together with (2) a square root of a first ratio of the left original power of said left original signal to the left intervening power of said left intervening signal; and

(b3) calculating said right reproduced signal by multiplying (1) said right intervening signal by a right maximum absolute sampled data for the right original signal together with (2) a square root of a second ratio of the right original power of said right original signal to the right intervening power of said right intervening signal.

7. A stereo audio encoding method as claimed in claim 1, wherein said processing step (c) comprises the steps of:

(c1) respectively calculating a left power and a right power for each of said left and right original signals;

(c2) calculating a first ratio of

(1) a first product of;

a maximum absolute sampled data of an average of the left and right original signals; and

a first sum of the left power of said left signal and a first crossproduct of the left and right original signals to

(2) a second sum of said left power and said right power of said left and right original signals and a second crossproduct of said left and right original signals;

(c3) obtaining a second ratio of

(1) a second product of;

said maximum absolute sampled data of the average of said left and right original signals;

a third sum of said right power of said right signal and a third crossproduct of said left and right original signals to

(2) a fourth sum of said left power and said right power of said left and right original signals and a fourth crossproduct of the left and right original signals;

(c4) obtaining said left reproduced signal by multiplying the average of the left and right original signals by said first ratio; and

(c5) obtaining said right reproduced signal by multiplying the average of the left and right original signals by said second ratio.

2. A stereo audio encoding method as claimed in 1, wherein said processing step (b) comprises the steps of:

adding said left and right original signals; and

dividing the added signal into left and right reproduced signals relative to a first ratio between (1) a maximum absolute data of an average signal which is an average between said left and right original signals to (2) a maximum absolute sampled data in the left and right original signals and also to a second ratio between (1) a power of the left original signal to (2) a power of d right original signal.

8. A stereo audio encoding apparatus for encoding audio signals to left and right reproduced signals, said apparatus comprising:

calculating means for calculating a correlation between left and right original signals to detect a pair of the left and right original signals having opposite phase characteristics and a pair of the left and right original signals having same phase characteristics, where the audio signals have been transformed to the left and right original signals each defined by a train of frames where each frame contains a plurality of sampled data;

first processing means for processing said left and right original signals having the opposite phase characteristics according to a power equalization method to obtain said left and right reproduced signals; and

second processing means for processing said left and right original signals having the same phase characteristics according to an error minimization method to obtain said left and right reproduced signals.

9. A stereo audio encoding method for encoding audio signals to produce modified scale factors, comprising the steps of:

(a) calculating a correlation between left and right original signals to detect a pair of the left and right original signals having opposite phase characteristics and a pair of the left and right original signals having same phase characteristics where the audio signals have been transformed to the left and right original signals each defined by a train of frames where each frame contains a plurality of sampled data;

(b) processing said left and right original signals having the opposite phase characteristics according to a power equalization method to obtain left and right scale factor modifiers; and

(c) processing said left and right original signals having the same phase characteristics according to an error minimization method to obtain left and right scale factor modifiers.

10. A stereo audio encoding method as claimed in claim 9, wherein said processing step (b) comprises the steps of:

(b1) calculating an average signal which is an average between said left and right original signals;

(b2) calculating an average maximum absolute data of said average signal;

(b3) respectively selecting a left maximum absolute sampled data and a right maximum absolute sampled data in each frame for each of said left and right original signals;

(b4) calculating a left intervening signal by multiplying said average signal by said left maximum absolute sampled data and a right intervening signal by multiplying said average signal by said right maximum absolute sampled data;

(b5) calculating said left scale factor modifier by multiplying a first ratio of (1) said average maximum absolute data to (2) said left maximum absolute sampled data, by a first square root of a second ratio of (1) a left power of the left original signal to an average power of said average signal; and

(b6) calculating said right scale factor modifier by multiplying a third ratio of (1) said average maximum absolute data to (2) said right maximum absolute sampled data, by a second square root of a fourth ratio of (1) a right power of the right original signal to the average power of said average signal.

11. A stereo audio encoding method as claimed in claim 9, wherein said processing step (c) comprises the steps of:

(c1) respectively calculating a left power and a right power for each of said left and right original signals;

(c2) calculating an interaction term by multiplying each of said plurality of sampled data in the left original signal with corresponding data of the plurality of sampled data in the right original signal;

(c3) calculating an average signal which is an average between said left and right original signals;

(c4) calculating an average maximum absolute sampled data in the average signal;

(c5) respectively calculating a left maximum absolute sampled data and a right maximum absolute sampled data in each frame for each of said left and right original signals;

(c6) calculating said left scale factor modifier which is equal to a first ratio of (1) a first product of said average maximum absolute sampled data multiplied by a sum of said left power and of said left signal and said interaction term to (2) a second product of said left maximum absolute sampled data in said left original signal multiplied by a sum of said left power and said right power of said left and right original signals and said interaction term; and

(c7) calculating said right scale factor modifier which is equal to a second ratio of (1) a third product of said average maximum absolute sampled data multiplied by a sum of said right power of said right signal and said interaction term to (2) a fourth product of said right maximum absolute sampled data in said right original signal multiplied by a sum of said left power and said right power of said left and right original signals and said interaction term.

12. (Amended) A stereo encoding method as claimed in claim 9, wherein said processing step (b) comprises the steps of:

(b1) respectively calculating a left original power, a right original power, a left intervening power, and a right intervening power for each of said left and right original signals, and each of said left and right intervening signals wherein said left and right intervening signals are obtained by multiplication of average of the left and right original signal with a maximum sampled data of said left and right original signals; and

(b2) calculating said left reproduced signal by multiplying (1) said left intervening signal by a left maximum absolute sampled data for the left original signal together with (2) a square root of a first ratio of the left original power of said left original signal to the left intervening power of said left intervening signal; and

(b3) calculating said right reproduced signal by multiplying (1) said right intervening signal by a right maximum absolute sampled data for the right original signal together with (2) a square root of a second ratio of the right original power of said right original signal to the right intervening power of said right intervening signal.

13. A stereo audio encoding method as claimed in claim 9, wherein said processing step (c) comprises the steps of:

(c1) respectively calculating a left power and a right power for each of said left and right original signals;

(c2) calculating a first ratio of

(1) a first product of;

a maximum absolute sampled data of an average of the left and right original signals; and

a first sum of the left power of said left signal and a first crossproduct of the left and right original signals to

(2) a second sum of said left power and said right power of said left and right original signals and a second crossproduct of said left and right original signals;

(c3) obtaining a second ratio of

(1) a second product of;

said maximum absolute sampled data of the average of said left and right original signals;

a third sum of said right power of said right signal and a third crossproduct of said left and right original signals to

(2) a fourth sum of said left power and said right power of said left and right original signals and a fourth crossproduct of the left and right original signals;

(c4) obtaining said left reproduced signal by multiplying the verage of the left and right original signals by said first ratio; and

(c5) obtaining said right reproduced signal by multiplying the average of the left and right original signals by said second ratio.

14. A stereo audio encoding apparatus for encoding audio signals to produce modified scale factors, said apparatus comprising:

calculating means for calculating a correlation between left and right original signals to detect a pair of the left and right original signals having opposite phase characteristics and a pair of the left and right original signals having same phase characteristics, where the audio signals have been transformed to the left and right original signals each defined by a train of frames where each frame contains a plurality of sampled data;

first processing means for processing said left and right original signals having the opposite phase characteristics according to a power equalization method to obtain the left and right scale factor modifiers; and

second processing means for processing said left and right original signals having the same phase characteristics according to an error minimization method to obtain the left and right scale factors modifiers.